Compared to democratic leaders, autocratsFootnote 1 typically face fewer institutional and popular constraints, granting them greater latitude in decision making. Understanding their goals and preferences therefore is essential—not only for academic analysis but also for diplomatic engagement and strategic responses by democratic leaders. Thus, dealing with Vladimir Putin, Xi Jinping, and Middle Eastern kings is much easier if we have some idea about their respective interests and goals. That is, to what extent are they office-seekers—seeking to maximize power and wealth for their own personal gain—or policy-seekers, driven by a commitment to certain policy goals.
Exemplifying portrayals of dictators often focus on figures such as Mobutu Sese Seko, Ferdinand Marcos, and Kim Jong Il, who personify the traits of greed, egocentrism, and brutality. However, others—including Julius Nyerere, Lee Kuan Yew, Park Chung-Hee, Hugo Chávez, and Pol Pot—were driven strongly by ideological commitments, albeit not necessarily noble ones.
However, a review of the flourishing comparative politics literature on authoritarian regimes, which are defined as countries with absence of meaningful electoral contestation for executive power (Schumpeter Reference Schumpeter1974), reveals that the former set of dictators are viewed as the stereotype. Contemporary global research on autocracies focuses predominantly on political survival. The foundational Downesian assumption is that dictators aim to hold on to power as long as possible and want to maximize their personal wealth; all dictators are first and foremost office-seeking. The theoretical and empirical focus has been to investigate how different sets of institutions and constraints lead to certain outcomes by forcing dictators to behave in certain ways to fulfill their simple and narrowly selfish motives (Albertus and Menaldo Reference Albertus and Menaldo2012; Bueno de Mesquita et al. Reference de Mesquita, Bruce, Siverson and Morrow2003; Gandhi Reference Gandhi2008; Gerschewski Reference Gerschewski2013; Olson Reference Olson1993; Svolik Reference Svolik2012). However, neglecting the importance of other motivations such as genuine ideological motivations or beliefs makes these parsimonious theories insufficient for explaining more specific policies and outcomes.
There are strong reasons to expect variation in dictators’ motivations. Psychological research demonstrates that human beings differ significantly in their core motivations and preferences, and extreme egocentrism—such as is normally assumed for autocrats—is rare (Andreoni and Miller Reference Andreoni and Miller2002; Baumard, André, and Sperber Reference Baumard, André and Sperber2013). The path to autocratic rule is strewn with obstacles and may filter out individuals who do not have a high tolerance for risk and an intense drive for power, potentially introducing a selection bias that favors office-seekers. However, those with strong ideological convictions may be at least as likely to endure this struggle, particularly in high-risk routes to power such as armed rebellions and insurgencies (DeNardo Reference DeNardo1985; Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018). What makes dictators tick remains an open question.
The prevalence of policy-seeking motivation among authoritarian leaders—that is, dictators’ motivation to implement sincere ideological aims and broader beliefs—has not been investigated on a broad scale. Existing data mainly focus on institutions (Cheibub, Gandhi, and Vreeland Reference Cheibub, Gandhi and Vreeland2010; Coppedge et al. Reference Coppedge, Gerring and Knutsen2020; Geddes, Wright, and Frantz 2014, 2017; Goemans, Gleditsch, and Chiozza Reference Goemans, Gleditsch and Chiozza2009; Hadenius and Teorell Reference Hadenius and Teorell2007; Svolik Reference Svolik2012) and the leaders’ personal background, such as education, socioeconomic factors, and previous occupation (Baturo Reference Baturo2016; Ellis, Horowitz, and Stam Reference Ellis, Horowitz and Stam2015). This study expands this literature by investigating the diversity of dictators’ motives in a global perspective.
The article enlists newly collected original data on 297 dictators who held power from 1945 to 2008. It presents observable evidence linked to motivation, such as lifestyle, involvement in self-enrichment, and the extent to which dictators stepped down voluntarily, as well as more judgmental evidence based on country experts’ evaluations of the dictators’ motives. The main source for the study was obituaries, a relatively standardized text form from which comparable information can be drawn. The findings are validated in several ways, including the use of in-depth case studies (Thorsen Reference Thorsen2025).
The empirical results reveal large variation in motivation. Even when motivation is evaluated conservatively with respect to policy-seeking motivation, 25% of the dictators in the sample were driven strongly by ideological concerns. Direct statements in the obituaries suggest that 43% of the dictators were motivated by policy-seeking concerns. I also found significant variation across dictators with regard to three primary measures of more observational character. Several dictators lived ascetic lives and did not steal from the state, and many stepped down voluntarily. These results question the stereotype of selfish dictators that is the premise for much of the literature on authoritarian politics, and they increase our understanding of dictators as well as our ability to explain the political dynamics in autocracies.
THEORETICAL FOUNDATIONS OF THE POLICY-SEEKING AUTOCRAT
Why should we expect dictators’ motivations to vary? First, qualitative studies of individual autocrats reveal far more nuanced motivations than large-N comparative research suggests (Bjerk Reference Bjerk2017, 14; Jones Reference Jones2015; Mauzy and Milne Reference Mauzy and Milne2002, 5). Accounts of Julius Nyerere, Mahathir Mohamed, and Lee Kuan Yew, for instance, demonstrate that ideological views can matter deeply for policy making (Bjerk Reference Bjerk2017, 15; Khoo Reference Khoo2003, 167–69; Mauzy and Milne Reference Mauzy and Milne2002, 5, 128). Second, psychological research shows that motivations differ significantly across individuals and that extreme self-interest is rare (Andreoni and Miller Reference Andreoni and Miller2002; Baumard, André, and Sperber Reference Baumard, André and Sperber2013). Expanding on this, I argue that the way dictators attain power explains why some are more policy-seeking and others are more office-seeking.
Road to Power: Self-Selection and Socialization
Individuals’ motivations are shaped by fairness concerns, yet conceptions of fairness vary substantially—from more egocentric (i.e., everyone should serve themselves) to more prosocial orientations (Andreoni and Miller Reference Andreoni and Miller2002; Baumard et al. Reference Baumard, André and Sperber2013). In politics, office-seeking aligns with the more egocentric side, whereas policy-seeking reflects the more prosocial. Motivations stem partly from genetic dispositions but also are shaped through socialization (Alford, Funk, and Hibbing Reference Alford, Funk and Hibbing2005), which may steer a future autocrat toward self-serving or societal goals.
Although family background and education may provide formative influences (Alford, Funk, and Hibbing Reference Alford, Funk and Hibbing2005; Nie, Junn, and Stehlik-Barry Reference Nie, Junn and Stehlik-Barry1996), the road to power appears to be the more immediate and politically consequential factor in explaining the variation in dictator motivations (DeNardo Reference DeNardo1985; Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018)—a factor that also embeds formative experiences through a selection effect. The road to power may explain diverging motivations among dictators through two mechanisms: self-selection and socialization. The self-selection effect implies that only certain types of people pursue certain paths: high-risk paths attract those with strong ideological conviction and low-risk paths attract a broader mix, including opportunists. The socialization effect refers to the way that experiences on the road to power reshape identities and reinforce motivations (Berger and Luckmann Reference Berger and Luckmann1966). Together, these mechanisms likely account for systematic differences in authoritarian leaders’ motivations.
At one end of the spectrum are rulers who gain office through armed struggle or insurgency. They are the most likely to have strong ideological commitments (DeNardo Reference DeNardo1985; Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018). First, the hardship, danger, and uncertainty of such a struggle create a strong self-selection effect: only those with deep ideological commitment are willing to endure them. Second, the socialization of constant confrontation with perceived injustice further reinforces grievance and dedication. These autocrats, therefore, are likely to be among the most policy-driven once they are in office.
Military coups, by contrast, do not necessarily favor policy-seekers. Although armies may socialize officers into ideas of national service, many soldiers are simply career professionals. Coups are risky but, for military officers, they are relatively feasible because of their access to arms and organization (Powell Reference Powell2012). Because coups are short-term undertakings with the prospect of immediate gains, they attract both policy-minded guardians (Nordlinger Reference Nordlinger1977) and opportunistic power-seekers (Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018). Coups often involve multiple competing factions, especially when a government is weak and unpopular, which adds to the heterogeneity of outcomes (Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018).
Another typical road to power is advancement from within the political system, through political parties, via either incumbent parties or opposition movements. Opposition activity, espeically under a closed autocracy, can be highly risky—akin to insurgency—and thus more often self-selects for policy-seekers. In contrast, long careers within ruling parties expose elites to patronage and insulation from grievances, which likely fosters more office-seeking behavior. Advancement within the ruling party typically is a low-risk, high-reward path, where even ordinary membership can yield material benefits (Gandhi Reference Gandhi2008; Svolik Reference Svolik2012), and often fosters opportunistic careerists.
Hereditary succession is the lowest-cost route. These rulers are socialized within privilege and often lack exposure to popular grievances (Anderson Reference Anderson1991), making them less likely to be policy-seeking. Some may inherit a paternalistic sense of duty but, in general, their motivation likely skews toward office rather than policy.
Taken together, this framework suggests that insurgent and opposition leaders are the most likely to be policy-seeking; military rulers occupy an intermediate position; and elites from incumbent parties or hereditary families are the least ideologically motivated. Early predispositions and socialization shape which path, if any, is pursued, whereas the experience of attaining power reinforces or redirects these motivations. Regime type adds another layer: single-party regimes, often rooted in ideological movements, are more likely to produce policy-driven rulers (Geddes, Wright, and Frantz Reference Geddes, Wright and Frantz2018); personalist regimes tend to favor office-seeking dictators because personalist dictators, by definition, maximize their personal power; and military regimes fall in between.Footnote 3
In summary, dictators’ motivations reflect both personal traits and political trajectories, which are shaped by the institutional context in which they rise to power. This framework challenges the prevailing assumption of uniform office-seeking autocrats and instead provides theoretical grounds for expecting meaningful variation in dictators’ motivations.
…dictators’ motivations reflect both personal traits and political trajectories, which are shaped by the institutional context in which they rise to power.
MEASURING DICTATOR MOTIVATION: METHODOLOGY
To investigate dictators’ motivation on a broad scale, I collected data on 297 deceased dictators who held power at some point during the period 1945–2008. The data are based on 695 obituaries from Western newspapers and contain a variety of leader-specific characteristics. The focus is on motivation, and there are variables that measure the extent to which a dictator seems mainly preoccupied with office-seeking goals such as power and wealth maximization, in contrast to more policy-seeking motives such as sincerely held ideology or other beliefs. (See the codebook in the online appendix for more information about the sample and the variables in the dataset.Footnote 4)
Pros and Cons of Using Obituaries as Source
Motivation is difficult to assess and impossible to observe directly; coding it therefore is challenging. The closest understanding of a dictator’s motivation is from “expert” assessments—that is, assessments from those with in-depth knowledge about the dictator. However, choosing the appropriate experts whose assessments are relevant, precise, and comparable across countries and contexts is difficult and resource demanding (Skaaning Reference Skaaning2018, 110–11). The coding for this study is based on a particular type of relatively comparable expert assessment: obituaries. They are relatively brief texts (i.e., 500–3,000 words) about dictators, typically written by foreign correspondents or scholars with a thorough knowledge of politics and society in the dictator’s country or region. For example, Jeffrey Gettleman, who wrote Ethiopian dictator Meles Zenawi’s obituary, was the New York Times’ East Africa bureau chief in Kenya for more than 10 years. David Binder, who wrote Bulgarian Todor Zhivkov’s and Yugoslav Josip Tito’s obituaries, served as a correspondent in Europe and lectured on and wrote several academic articles about the Balkans (McFadden Reference McFadden2019; New York Times Reference Times2019). Obituaries typically contain a brief summary of a dictator’s life, incumbency, motives, and achievements.Footnote 5 I relied primarily on obituaries from the New York Times, Washington Post, The Independent, and The Guardian because they are major newspapers with broad spatial and temporal coverage, and they are widely recognized for independent, balanced, and high-quality reporting.Footnote 6 This makes the obituaries an easily accessible source of relatively condensed in-depth and relevant information about dictators.
My coding was based on a particular type of relatively comparable “expert assessment”: obituaries.
Source Validation
There are potential disadvantages to using obituaries. An obituary may be biased toward judging dictators as more ideologically motivated than they were. We tend not to speak critically about the deceased; therefore, obituary writers may be inclined to downplay self-interest. However, ideology is not necessarily beneficial and (especially in extreme cases) may be viewed as a foul motive. Relatedly, whereas obituaries often include information about policy, my measurement approach attempted to distinguish between more general implementation of policy and policy-seeking behavior—specifically, a committed pursuit of particular ideological policy goals.
Another potential problem is that new information about a dictator may be revealed months or years after his obituary is published. This concern is relevant mainly for dictators who die in office (30% of the sample) because information about a specific dictator often starts to surface when he loses power. Comparison of the dictators who died in power to the remainder of the sample indicates that this bias is minor because the two groups are almost identical regarding motivation (see table S1 in the online appendix). On this basis, the timing of obituaries should not be expected to create any strong bias.
To further alleviate problems regarding potential bias in the obituaries, I crosschecked 15 randomly selected cases with in-depth case studies based on biographies and historical writings about specific countries (see table S2 in the online appendix). The crosscheck revealed a significant overlap in assessments and no systematic tendency to evaluate dictators as more ideologically motivated than in in-depth case studies (see table S3 in the online appendix).
I relied on as many as four obituaries per dictator when they were available. Generally, the New York Times had the broadest coverage and, in approximately 90% of the cases, my coding included obituaries from this newspaper. I included variables that indicate substantial disagreement among obituaries in the assessment of core variables. The sources disagreed in only two of the almost 300 cases and only with respect to the lifestyle variable. Thus, the main problem with the varying number of obituaries is that the coding was more certain for some dictators than for others. To alleviate this problem, I included a binary variable indicating whether the coding of a specific dictator was particularly uncertain.
I coded the entire dataset myself to ensure consistency and internal validity, but a risk of bias may still remain (Skaaning Reference Skaaning2018). An intercoder reliability test on a random sample of 50 dictators (see table S6 in the online appendix) revealed a substantial overlap in the coding. I reassessed the divergent cases and recoded some of them.
EMPIRICAL TRENDS IN DICTATOR MOTIVATION
To investigate the relative prevalence of policy-seeking versus office-seeking motivation among dictators, I first measured direct statements in the obituaries. Policy-seeking motivation was coded as present if at least one of the following statements was directly stated in the obituary:
-
(1) The dictator was ideologically motivated, visionary, or idealistic.
-
(2) The dictator cared about the people (or a significant subset of the people).
-
(3) The dictator had ideological or visionary aims and goals.
-
(4) The dictator was a Marxist, communist, liberal, etc.7
-
(5) The dictator was trying to implement a specific ideology.
For example: “He [Gomulka, Poland] was, of course, a Communist, but in the public’s view—and the public was largely Roman Catholic and conservative—he was also a nationalist and a patriot.” One of the most elaborate examples is: “Marshall Tito [Yugoslavia] appears to have developed a vision….It was a vision of a socialist Yugoslavia, and in the next five decades he never lost sight of it.” Direct statements of office-seeking motivation portray a dictator as being driven by power, wealth, or opportunism—for instance: “Siad Barre’s [Somalia] overwhelming desire was to have, and to hold on to, power at all costs.”
The obituaries contained at least one statement about policy-seeking motivation for 43% of the dictators and about office-seeking motivation for 16%. Although the measure was crude, it strongly indicates that policy-seeking motivation may be more prevalent than existing research assumes.
In addition to the quote-based measure of motivation, I measured motivation as an overall judgment of a dictator’s motives based on the obituaries; thus, direct statements were also a part of the overall judgment. The degree of policy-seeking relative to office-seeking motivation was coded on a four-point scale. The two polar categories were cases in which the dictator was clearly driven by either office-seeking or policy-seeking motivation. The two middle categories contain cases with indications of both (i.e., mixed motives) or uncertainty in the assessment, typically caused by a significant overlap between policy- and office-seeking behavior. These dictators’ behavior was often highly consistent with both types of motives, such as when the dictator was highly constrained (see the codebook in the online appendix).Footnote 8 Generally, the coding was restrictive regarding policy-seeking motivation, which implies that very convincing evidence of prioritizing policy over power and wealth was needed to place a dictator in the highest category. Therefore, despite finding direct statements about ideological motivation, dictators frequently achieved only the second-highest score, meaning that they were assessed as relatively policy-seeking. In cases in which a dictator’s motives seemed to change over time (5% of the sample), the coding was based on the predominant motivation during the incumbency. Table 1 lists examples of well-known dictators in each category.
Table 1 Examples of Dictators on the Motivation Spectrum
Note: See the online appendix for a full list of the dictators in the dataset.
Of the dictators in the sample, 5% appeared to not want to be in power but were installed by others or as part of a compromise; therefore, they were placed in a residual category. Examples include military dictators who were installed to end a military dictatorship and hand over power to civil democratic rule.
Figure 1 illustrates the distribution of the dictators’ motivation on the judgment-based motivation measure (without residual categories): 44% (the largest category) scored as relatively policy-seeking and 25% (the second-largest category) as very policy-seeking. This indicates that more than two thirds of the dictators were substantially motivated by implementing policy according to their ideology or broader beliefs. The two middle categories on the measure were somewhat uncertain and difficult to distinguish; however, even a very conservative assessment of ideological motivation implies that 25% of the dictators in the sample were clearly and substantially motivated by policy and not predominantly by office.
Figure 1 Dictators’ Motivation: Judgment-Based Motivation
Note: N=281; residual categories are excluded.
Therefore, even with a conservative interpretation, these measures indicate that dictators have diverse motives and that many are substantially motivated by ideological concerns rather than their own power and wealth. It is noteworthy that the policy-seeking dictators represent many different ideologies.Footnote 9
…even with a conservative interpretation, these measures indicate that dictators have diverse motives and that many are substantially motivated by ideological concerns rather than their own power and wealth.
QUALIFYING THE FINDINGS
To supplement these findings, I investigated other motivation-related observable traits of dictators—namely, involvement in embezzlement, extravagant lifestyle, and voluntary stepdown (for details, see the codebook in the online appendix). Although these three measures do not directly correspond to the distinction between policy-seeking and office-seeking motivations, they measure motivational aspects. For instance, self-enrichment and cultivating a lavish lifestyle (e.g., building large palaces) correspond well with classic office-seeking motivation.
Empirically, there was considerable variation across the measures. On embezzlement (i.e., four-point scale), 13% of the dictators clearly were engaging in self-enrichment, whereas 7% of the dictators were not corrupt at all. The latter group includes several unconstrained dictators, such as Ahidjo (Cameroon), Lee (Singapore), and Chiang Kai-shek (Taiwan). Dictators were fairly evenly distributed across the three lifestyle categories—lavish, moderate, and modest. More than 29% had a modest lifestyle, which is a significant percentage considering how the conventional literature perceives dictators. An example of a reported modest lifestyle: “As chief of state, he [Ibrahim Abboud, Sudan] lived in a modest one-story brick house in Khartoum with his wife and five children.” Of course, a modest public image does not preclude personal wealth or corruption behind the scenes, as in the cases of Castro (Cuba) and Gaddafi (Libya). However, the fact that even powerful autocrats adopt a modest lifestyle in public suggests that lifestyle choices may reflect differing motivations among authoritarian leaders.
Voluntary stepdown is not expected among either policy- or office-seeking dictators—but it would be most likely among the former if they believed it to be best for their country. Somewhat surprisingly, however, 7% step down completely voluntarily and 22% partially voluntarily (e.g., after accepting a lost election in which they most likely could have stayed using force). It appears that dictators who step down voluntarily are mainly strongly ideologically motivated dictators, such as Nyerere (Tanzania) and Lee (Singapore), or former military dictators who deliberately chose to install elections and not run, such as Aramburu (Argentina) and Gizikis (Greece).
To qualify the findings further, I investigated the temporal trends in dictator motivation. I found that policy-seeking motivation was prevalent throughout the period of investigation but peaked in the decolonization period in the 1960s (see figure S1 in the online appendix).
Given the dominant focus on institutions in existing research on autocracies, I also investigated the relationship between motivation and institutions in ordinary least squares regression models with various specifications. I found that dictator motivation (i.e., relative level of policy-seeking motivation) was unrelated to degree of authoritarianism (i.e., lack of democracy) (see table S7 in the online appendix). However, consistent with Geddes, Wright, and Frantz’s (Reference Geddes, Wright and Frantz2014) framework, there is evidence that dictators in one-party regimes tend to be the most policy-seeking, which reflects the ideological or revolutionary origins of many such regimes. Over successive generations, some party regimes evolve into personalist dictatorships, possibly as initial grievances and policy-driven objectives fade over time. As expected, personalist dictators tend to be the most office-seeking. Military regimes occupy an intermediate position, presumably reflecting divided motivations (i.e., military leaders with national policy concerns versus opportunistic coup plotters). Despite these tendencies, the relationship between authoritarian regime type and the ruling autocrat’s motivation remains limited (see table S7 and figure S2 in the online appendix). Motivation thus appears to be a relatively independent explanatory factor, providing substantial nuance to the conventional view of dictators as primarily office-seekers.
IMPLICATIONS
This study argues and demonstrates that dictators’ motivations vary greatly, and that a majority of documented dictators were not motivated primarily by office (i.e., wealth and power). To that end, this article provides new and systematic empirical evidence of unprecedented scope based on obituaries—a relatively brief source that contains in-depth expert information needed to make a reasonable assessment of a dictator’s motivation. Descriptive analysis reveals that dictators have diverse motivations, both more broadly and in terms of ideological content. Even when motivation is evaluated conservatively with respect to policy-seeking motivation, 25% of the sample was strongly driven by policy—that is, ideological concerns and broader beliefs. In addition to the judgment-based measures of motivation, the analysis used motivation measures of a more observational character, which also revealed huge variation. Several dictators lived ascetic lives and did not embezzle the state coffers, and several stepped down voluntarily. These results are surprising in light of the selfish motives that typically are attributed to dictators.
In addition to these descriptive results, the study provides data that enable us to investigate the emergence of dictators with different motives and their consequences for outcomes such as development, repression, and conflict. Moreover, we may want to construct conditional arguments in which dictators’ motives appear as scope conditions such that a theoretical expectation is expected only if the dictators hold certain motives. Just as many theoretical and empirical models depend on a particular type of autocratic setting, they also may depend on certain types of motives of an autocrat. For instance, when we scrutinize arguments for why democracies or autocracies are expected to provide more economic development, they implicitly hinge on leaders’ motivations. Specifically, they depend on whether the leaders are selfish and their motives translate into predatory behavior if not constrained (e.g., by democratic institutions) or whether they actually care about policy goals such that development is better provided by a leader with autonomy and shielded from public pressure (as in dictatorships) (Knutsen Reference Knutsen2012).
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://doi.org/10.1017/S1049096525101509.
DATA AVAILABILITY STATEMENT
Research documentation and data that support the findings of this study are openly available at the PS: Political Science & Politics Harvard Dataverse at https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi%3A10.7910%2FDVN%2FIMI31S&version=DRAFT.
CONFLICTS OF INTEREST
The author declares that there are no ethical issues or conflicts of interest in this research.
Open access
