Crowdsourcing a Democracy Index – 2012 edition

It’s that time of the year again – time to crowdsource a democracy index.

First, a bit of context. Last year, I had the idea of using the Allourideas pairwise comparison software to crowdsource a ranking of countries by their degree of “democracy” in 2010. I asked students in my Dictatorships and Revolutions class to set the ball rolling, and then posted the link to the widget here, allowing anybody to vote. Surprisingly, in just a couple of months of voting the results were interestingly close to standard indexes of freedom or democracy: the crowdsourced ranking had a correlation of 0.84 with Freedom House’s widely-used ratings of political and civil liberties for 2010, and the basic crowdsourced ranking was generally plausible (see this post for further analysis of these results) . In fact, by now the correlation has increased to 0.86 (8556 votes total), which is about as good the correlation between Freedom House and Polity IV (0.87).

I am interested in seeing if this kind of crowdsourced measure can be used as a sort of quick and dirty index of democracy. To be sure, crowdsourcing the construction of an index of democracy in this way is usually not a good way of generating reliable social science data. For one thing, the exercise does not impose any restrictions on how the concept of “democracy” should be understood, which means that it implicitly aggregates all kinds of different ideas about democracy, weighting them by the degree to which larger numbers of people consider them important. (It takes a “democratic” approach to concept formation, you might say). But it does have the virtue of being cheap (total cost: about $0, compared to over US$500,000 annually for the Freedom House “Freedom in the World” report, and $120,000 annually for the Polity IV project), aggregating the dispersed information of large numbers of people from all over the world, and making it possible to generate various measures of “uncertainty” around the crowdsourced estimates. So I want to repeat the exercise, and produce a democracy ranking for 2011:

(Click here if you can’t see the widget above. Vote as many times as you'd like, and don't worry if you have to use the "I can't decide" button).

In theory, the more votes, and the more diverse the voting population – the more people from Africa, Latin America, Asia, and the rest of the world – the more informative the results should be. So please vote early and often, and share!

I am also interested in which dimensions of the complex concept of democracy people tend to weigh more when making these sorts of comparisons. Do people put more weight on the presence or absence of elections, for example, than on economic equality? You can help me to figure this out by using the widget below to rank various dimensions or components of democracy in terms of their importance to you (or adding your own):

(Click here if you can’t see the widget above. Vote as many times as you'd like, and don't worry if you have to use the "I can't decide" button).

The “seed” dimensions of democracy for these comparisons are taken from a recent piece by Michael Coppedge, John Gerring, et al. (“Conceptualizing and Measuring Democracy”, Perspectives on Politics 9(2): pp. 42-62; ungated here) that is very much worth reading if you are interested in the issue of how to measure democracy. But I make no claim that these are the only dimensions of democracy that matter; if people have other ideas, you can add them in (I will need to approve any suggestions, though).  You are also welcome to discuss in comments the kinds of considerations that you used to make distinctions between countries, or any other considerations that might improve the usefulness of this sort of exercise.

Enjoy, and please share!

The Irrelevance of Legitimacy - now as a working paper!

(This paper has now been published. The official version is available here, and an ungated version can be found here).
 
Sorry for the recent silence. I've been busy with administrative tasks, the beginning of the term here in the Southern hemisphere, and indexing a book. I've also been working on a paper: "The Irrelevance of Legitimacy":

The concept of legitimacy plays an important explanatory and normative role in political theory and political discourse. The idea is typically used both to explain the stability of a political order by pointing to acceptance of discursive justifications for that order, and to evaluate its normative appropriateness by comparing the conditions of the actual acceptance of discourses of justification to the conditions of their rational acceptability. The normative and explanatory roles of the concept of legitimacy are linked insofar as actual acceptance of justificatory discourses is usually taken to be (defeasible) evidence for their rational acceptability. I argue here that legitimacy (in the sense of acceptance of discursive justifications for political order) is generally irrelevant for the explanation of political stability: if anything, stability explains legitimacy rather than the other way around. Stability is in turn best explained by the way in which signals of commitment are generated through collective action, not by pointing to the individual acceptance of discursive justifications for political order. I illustrate the inadequacy of explanations of political order in terms of legitimacy by examining the phenomenon of cults of personality in totalitarian regimes, and raise some questions about the normative utility of the concept given its explanatory irrelevance. (Link to download).

I'll be presenting a shortened version of this paper at the Midwest Political Science Association meeting in Chicago on April 13.

Most of the arguments in the paper will be familiar to readers of this blog; in fact, many began life as blog posts (see, e.g., here, here, here, and here), though the paper ties them together and explicates them more carefully. Responses to these posts convinced me that there was something there worth researching more fully and discussing in more detail, and I want to thank readers and commenters for feedback and encouragement. Comments on the paper are also welcome; it is still work in progress (the final section, in particular, still needs a great deal of work, but every part of the argument could be tightened and subject to careful scrutiny, and will likely change a great deal before the paper gets submitted to actual peer review).

The main ideas of the paper were motivated by my dissatisfaction with the Weberian dictum (almost a cliche at this point) that power needs to be legitimated in order to endure. Though relationships of domination are often embedded within justificatory discourses, my view is that we cannot in general explain the stability of such relationships by pointing to the genuine acceptance of such justifications by the subordinate. (As I note in the paper, Weber himself seemed to be aware of this point, if inconsistently; he notes that all that matters for the stability of relationships of domination is that the claim to legitimacy be taken as valid, not that it be believed, and these are two very different things). To say that power needs legitimacy in order to endure is at best to say that power needs credible commitments in order to endure, and discourses of justification typically provide the language in which such credible commitments are expressed and measured; they are the form, not the cause of the stability of power. The title is nevertheless a bit of an exaggeration; a more appropriate title might have been "the limited relevance of legitimacy," since I admit that there are some conditions (primarily cases where exit or voice constraints on a relationship are minimal) where appeals to legitimacy have some explanatory and normative force, but I decided to go for broke. Anyway, I would be grateful for any feedback. Enjoy!

Your irregularly scheduled blogging will resume shortly.

On stability and legitimacy

Consider the following passages from David Beetham’s The Legitimation of Power (1991):

…it is a notable feature of power relations that they are themselves capable of generating the evidence needed for their own legitimation. Thus the evidence of superiority and inferiority which justifies the inequality of condition between dominant and subordinate is itself largely the product of that condition. Those who are excluded from key positions, activities or resources are thereby denied the opportunity to acquire or demonstrate the capacities and characteristics appropriate to their occupation or exercise, so justifying their subordinate position. This is true even where relatively open processes of selection are at work, once the selection is performed by an educational system which is given the task of preparing children differentially for their respective future roles. Evidence about the fitness or appropriateness of people to exercise power thus tends to be structured by the relations of power themselves, and therefore to have a self-fulfilling quality about it.

The same holds true for demonstrations of the general interest. Once some necessary social resource or activity comes to be controlled by a particular group, it follows that the interests of society at large can only be met through satisfying the interests of that group, and on terms acceptable to them. (pp. 60-61)

[…]

The capacity of power structures to generate the evidence necessary to their own justification, and to reproduce the conditions of dependency from which consent to subordination is freely given, helps to explain how it is that their legitimacy can come to be widely acknowledged by those involved in them, the subordinate included. ‘Dominant ideology’ theories tend to put far too much emphasis upon the determining influence exercised by the powerful over the ideas of the subordinate, through their preferential access to the means of ideological construction and dissemination … The account offered here suggests a different kind of explanation: that both the evidence and the interests of the subordinate that the justifications advanced for the rules of power prove plausible to them within the given social context. Their plausibility can only be challenged from a position or standpoint outside that context, e.g. by comparison with alternative rules of power, or when social changes have come to undermine from within the evidence on which they are based. (p. 62)

I think this is pretty insightful, though I still want to take issue with it. Basically, what Beetham is arguing is that to the extent that subordinate groups willingly accept their position within large-scale systems of domination (e.g., to the extent that women accept a subordinate position to men in patriarchal societies, peasants to landlords in an agrarian societies, low-caste groups in caste societies, or for that matter disenfranchised people in a dictatorships or workers under capitalism) this is not primarily because they are duped by the propaganda of the dominant (the classic “false consciousness” explanation), but rather because the operation of the system makes the claims of the dominant – their claims to greater skill, intelligence, effort, care for the common good, etc. – generally plausible.  People will accept the propaganda of the powerful only if (and so long as) it is not obviously in conflict with their everyday experience; and a system will remain legitimate only so long as it works to reduce the gap between the lived experience of people in subordinate positions and the justificatory rhetoric of the powerful by producing systematic evidence that the claims of the powerful are plausible, even true, at least within the context of shared categories of interpretation (which may themselves be structurally biased to favour the views of the powerful). Moreover, given the plausibility of these claims, it will make sense for the subordinate to pursue their interests within the terms set by the system.

It is worth stressing that Beetham does not argue that this is always the case in every social system. Not every social system is experienced as legitimate; and those which are not experienced as legitimate are precisely those where the subordinate can perceive an obvious gap between the qualities or actions the powerful say justify their position and the qualities they actually have, or between the possibilities for pursuing their interests given within the system and possibilities obviously available elsewhere. Moreover, his argument only applies to large-scale social systems. Small-scale relations of domination could not systematically generate the evidence necessary for their own justification; they must “borrow” it from the larger scale system within which they are embedded. For example, in a highly patriarchal society the claims of men for their own position – claims based on education, experience, etc. that are themselves differentially allocated - will only be generally plausible to women; exceptions will abound, and within particular relations it will not always make sense for women to “accept” subordinate positions. (There are a lot of nuances and complications packed into the term “accept,” but let that pass for the moment). If women experience this system as legitimate, it will not necessarily be because of their individual “micro-experience,” so to speak, but because their understanding of “normative” facts (that is, their understanding of what is “normal”) tells them that in general women are less educated, have fewer of the relevant qualities for rule, etc., and that given those facts it will be easier to pursue their interests within the terms set by men.

Yet I think Beetham still puts too much stress on belief as a way to explain stability and pays too little attention to the constraints that opportunities for exit and collective action impose on these same beliefs. If anything, stability explains belief, rather than the other way around. To say that a legitimate system is stable, I want to suggest, is merely to express a tautology: systems of domination that are believed to be stable (even if this belief exists only because, for example, no one can coordinate large numbers on specific alternatives) will (most of the time) produce beliefs rationalizing the legitimacy of the system.

Consider the question of why repressive dictatorships invest so much effort in monopolizing the public sphere and policing the “attention economy.” The problem such regimes face is not that the majority of the people fail to believe the justificatory rhetoric of the rulers now (though they may or may not!), but that whatever they believe now cannot stabilize the regime in the absence of the regime’s efforts to monopolize the public sphere; if their justificatory claims were self-evidently true, they would not require policing the public sphere in the same way. These regimes can close the gap between their justificatory claims and the experience of the people subject to them only by credibly threatening to punish those who might want to break their monopoly over the discourse of justification. It is only when the system is believed to be stable – that is, when the threats to those who would like to break the state’s monopoly over the public sphere are deemed to be most credible – that people will be most likely to adaptively accept the claims of the powerful. 

One might also point to a number of experiments (ungated) that suggest that it is precisely when a constraint is experienced as inevitable or absolute that people are most likely to accept it, whereas when they experience the constraint as somehow not quite as absolute – perhaps because they see a way of changing it or resisting it that they are most likely to reject it.What matters causally, in other words, is the belief in stability rather than the belief in the justifications for the relation of domination; absent the belief in stability, the belief in the justifications also goes.  

Or consider the case of the dalits in India, as described in a short piece I came across recently by Shikha Dalmia. The caste system endures even where formal institutions do not enforce it; it is an informal equilibrium which the dalits themselves help to perpetuate. But why? Dalmia notes that an individual dalit will often do best by abiding by the rules of the caste system:

How? Consider Maya’s story.

Maya assigned herself to our house in 1977. We had no choice. If we wanted our trash picked up, bathrooms scrubbed and yards cleaned, Maya was it. Indians find dealing with other people’s refuse not just unpleasant, but polluting. Hence only dalits are willing to do this work, something that both stigmatizes them and gives them a stranglehold on the market. And they have transformed this stranglehold into an ironclad cartel that closes all other options for their customers.

When Maya got married at 16, her father-in-law paid another dalit $20 for her wedding gift: the “rights” to service 10 houses in our neighborhood, including ours. Maya has no formal deed to these “rights,” yet they are more inviolable than holy writ. Maya’s fellow dalits, who own the “rights” to other houses, can’t work in hers, just as she can’t work in theirs.

Doing so, Maya insists, would be tantamount to theft that would invite a well-deserved beating and ostracism by the dalit community. No one would help a “poacher” or attend her family functions like births, weddings or funerals.

This arrangement has guaranteed Maya a monthly income of $100 that, along with her husband’s job as a “gofer” at a government lab, has helped her raise three children and build a modest house with a bathroom, a prized feature among India’s poor. But Maya’s monopoly doesn’t give her just money. It also hands her clout to resist the upper-caste power structure, not always for noble reasons.

None of Maya’s employers dares challenge her work. Maya takes more days off for funerals every year than there are members in her extended family. Complaining, however, is not only pointless but perilous. It would result in stinking piles of garbage outside the complainer’s home for days. Every time my mother gets into spats with Maya over her sketchy scrubbing, my mother loses. One harsh word, and Maya boycotts our house until my mother cajoles her back. Nor is Maya the only sweeper, or jamadarni, with an attitude. All of New Delhi is carved up among Maya-style sweeper cartels and it is a rare house whose jamadarni is not a “big problem.”

This is consistent with Beetham’s story: Maya accepts her position within the caste system (to the extent that “accept” is the right word for what is going on, but let’s pass that over) because she finds that it is plausibly in her individual  interest to do so, though collectively this results in a bad outcome for all dalits, including social segregation, lack of mobility, etc. (Prisoner’s dilemmas everywhere!). But it is plausibly in her interest to accept her position because opportunities both for collective action to change the system and for exiting the relationship are thought to be unlikely; the collective action or exit constraint is prior to Maya’s first-order beliefs about her caste position, which would easily change if these collective constraints and exit opportunities changed, as we can glimpse near the end of the story:

Maya is resigned to such discrimination, but not her oldest son, 36. He holds a government job, works as a sales representative for an Amway-style company and dreams big. He is embarrassed by his mother and lies to his customers about her work. He makes enough money to support Maya and wants her to quit, but she will have none of it. She fears destitution and poverty more, she says, than she craves social respectability.

But the choice may not be hers much longer.

Upon retirement, she had planned to either pass her “business” to her children or sell it to another dalit for about $1,000. But about six months ago, municipal authorities started dispatching vans, Western-style, to collect trash from neighborhoods, the one service that protected Maya from obsolescence in an age of sophisticated home-cleaning gadgetry.

Maya and her fellow dalits held demonstrations outside the municipal commissioner’s office to stop the vans. They finally arrived at a compromise that lets Maya and her pals collect trash from individual homes and hand it to the vans for disposal. But Maya realizes that this arrangement won’t last. “I got branded as polluted and became unfit for other jobs, for what?” she wept. “To build a business that has now turned to dust?”

Her son, however, is pleased. He believes that this will finally force his siblings to develop skills for more respectable work instead of joining their mother. But Maya shakes her head.

And she might be right. Post-liberalization, the most dogged and determined dalits are able to escape their caste-assigned destiny and get rich. But for the vast majority, as Maya says, opportunities are better within the caste system than outside it.

Where does that leave us? If I had to make a grand claim – which I probably shouldn’t – I would suggest that relationships of domination are disrupted not so much because people come to reject the claims of the powerful, but because opportunities for exit or collective action become concretely available that make these beliefs dispensable. (Another example: the end of footbinding and infibulation. Link is to a superb paper by Gerry Mackie - ungated here).

[update 3/3/2012: fixed some really unclear sentences]

Varieties of Political Competition

(Part 1.5 of a series on the history ofpolitical regimes. This gets a bit technical in the second half, which contains a sort of blurry sketch of a theory of political regimes)

Our political vocabulary has a very long history. Terms like democracy, dictatorship, autocracy, tyranny, and even words of more recent vintage like authoritarianism or totalitarianism carry a great deal of descriptive and evaluative baggage, acquired in the course of political debate over a long time. They are rich, many-layered concepts, useful for simultaneously referring to and evaluating complex and vaguely-defined combinations of institutional, cultural, and other aspects of social systems. But by the same token, all that baggage makes it very difficult to use these terms to track the history of political institutions in anything but the roughest fashion. Or at least the history of the forms of political competition for power in states.

Consider an apparently simple question like how long the USA has been democratic. Since independence in the late 18^th century? Since the 1820s? Since reconstruction? Since the introduction of women’s suffrage? Since the 1960s? Never? Since the 1960s but only until recently? Until the 1930s? Historically, the idea of democracy has provided support for all of these answers, and all of them have had advocates. And though the current usage of the word "democracy" makes some answers more plausible than others, the controversy is pretty much ineradicable. We may at best be able to agree that the USA today is likely to be more democratic in some ways than in the late 19^th century (a more inclusive suffrage, more tolerance of certain forms of dissent), but perhaps less democratic in other ways than in the 1930s (a more entrenched and pervasive national security bureaucracy that is impervious to political control, greater structural barriers to entry into political competition, etc.). Moreover, even if we agreed that the USA has been democratic for some particular period, this would not necessarily tell us much about how competition for control of the state changed during that time: how norms of leadership selection evolved, how barriers to entry into political competition changed, how the ability to constrain the actions of the winners of the competition for power waxed or waned, etc.

Now consider instead the question of how long the USA has selected its top political leadership via competitive elections where candidates must enlist the support of a substantial fraction of a relatively large group (much greater than Dunbar’s number) to gain power, and where entry into political competition is normatively regulated (there are relatively well-enforced rules about who can be a candidate for power) but where such regulation does not impose large formal barriers to entry (the rules imply the existence of large pool of potential candidates representing a wide variety of interests and with a wide variety of life experiences and skills, even if structural barriers like access to money or racism considerably reduce this pool in fact). This question is in principle answerable (a: since the early 20^th century, at least), but it is not identical to the question of whether the USA is or ever was a democracy, whether in the USA “the people” or “the rich” or “the well connected” rule or have ever ruled, or indeed the question of whether its political system is any good at all.

It seems plausible to say that this method of leadership selection (competitive election) must be a component of democracy in large states (i.e., that no state without some such method of leadership selection can be called a democracy), yet this implies a whole theory of democracy relating competitive elections to ideals of inclusion and autonomy that is itself contestable. Ancient Greeks, for example, thought elections were characteristic of “oligarchic” regimes (rule by the rich), for understandable reasons having to do with the typical barriers to entry into political competition they presuppose (when offices are open to election, only the rich are able to compete for them effectively, as the American primary season certainly suggests); in their view, sortition [selection by lottery], not election, was more appropriate to the ideals of citizen equality implicit in the notion of demokratia (the power of the demos) and isonomia (citizen equality before the law) as they understood it. (They were certainly right for their political context; and perhaps their ideas are worth taking seriously today as well). Even the idea that elections are an institution of accountability (and hence something that helps “the people” to rule) rather than a mechanism of selection (useful only as an efficient way to discover the most talented rulers for an already established hierarchy) is a relatively recent development. At any rate, even if we agreed that competitive leadership selection through vote-gathering in large electorates is part of any definition of democracy in modern states, this does not imply that we would agree on the relative importance for democracy of such elections relative to, say, the inclusiveness of the electorate, the existence of a culture of tolerance, the responsiveness of elected officials to public opinion, or the protection of various rights.

From the point of view of trying to write a history of political regimes, it may thus be best to proceed in a disaggregated way: to speak not of democracy but of regimes where leadership selection is conducted on the basis of competitive elections with large electorates and few formal barriers to entry, for example, as I tried to do in my previous post. Instead of saying “democracy” we say CE/LG/LBE regimes (competitive elections/large group support/low barriers to entry). We thus substitute collections of small and relatively unimportant ideas like political competition through elections for large but important ideas like democracy, and then check whether these individually small and unimportant institutions come together in particular ensembles that make a difference to the things we care about. (E.g., like whether some particular combination of large-group elections, low barriers to entry into political competition, etc. actually tends to evolve over time, and whether these combinations tend to make a difference to the realization of particular ideals of autonomy or freedom). This avoids the historical tangledness of existing regime concepts, though at the cost of bracketing, at least for a time, evaluations of actual institutions. But at least then the important questions become either empirical (which institutions lead to the realization of particular ideals?) or purely philosophical (how should we weigh the relative importance of political values like equality, participation, etc.?).

I am primarily interested here in the institutions that shape political competition for the control of states (so non-state spaces are out of consideration for the moment), as well as the institutions that constrain the winners of such competition: the “varieties of political competition” for short. What we want to know are the “parameters” that describe such competition. What distinctions are useful for thinking about these varieties? A sketch of a theory of political competition might look at the following questions, only some of which overlap with the traditional concerns of democratic theory:

1.  As in the previous post, we might want to know whether the selection of those who control the state is regulated by more or less “self-enforcing” norms or not; periods where people attain power primarily by force (coups, revolutions, etc) seem to be substantially different than periods where people attain power in a normatively regulated way. Whether norms are self-enforcing depends on the commitments and resources of the actors subject to them, not on the norms themselves, since no third parties outside the regime exist to enforce them by definition (in other words, constitutions are mere pieces of parchment unless everyone can credibly commit not to defect). It would then be useful to know whether we can say anything general about the conditions that make some selection norms (hereditary, competitive, etc.) self-enforcing, not just competitive electoral selection norms.
2. We might also want to know which form the norm of leadership selection takes: whether the norm is to select people who fit a certain definite description (as in monarchies, where the next ruler is the heir of the current ruler), or to select people whose names are produced by some random process (lottery), or who meet a certain explicit standard (“meritocracy”), or, more commonly, who can show sufficient evidence of support within a group of “selectors” (election, acclamation, negotiation within the Politburo, etc.). (Mixtures of all of these selection norms are possible, especially when the state contains more than one center of power; in which case we might want to know something about the relative “powers” of these centers of power). Which selection norm is in force seems to have some effect on how long a leader is in power; selection norms that require a leader to show “evidence of support” appear to make it harder for leaders to keep control of states than selection norms that do not. It might also have an impact on the quality of political leaders (ungated), or on their social background. 
3. We might wish to know the size of the group whose support must be mustered when selection is regulated by “evidence of support” or "random" norms, as well as the kinds of barriers to entry into this group. This is in part because when contenders for power must show support from large groups (say, 4-5 times Dunbar’s number, or around 600 people, the size of the typical Central Committee of a communist party), elective institutions almost invariably develop; usually the only way to show that you have the support of 100,000 people is to count them. Moreover, small selector groups almost always indicate highly hierarchical social structures, require different sorts of “payment” by contenders hoping to secure support, and are easier to “cartelize” (preventing contenders from expanding the boundaries of the group by escalating political competition outwards).
4. We might wish to know the sort of “industrial organization” prevalent in the political arena (by analogy with this). For example, in many countries where control of the state is allocated to those who can show support from large numbers of adult citizens via elections, the political arena is typically organized around a few large multipurpose “firms” (parties) who are the main competitors for control, along with a large number of small special purpose groups (think-tanks, civil society groups, PACs, etc). Though the formal barriers to entry into this arena are low, the structural barriers are large (creating a party capable of appealing to large numbers of people takes real economic and symbolic resources), so that political competition operates in practice in an “oligopolic” manner. In other countries, political competition is fully monopolized by a single multipurpose “firm;” in yet others there is a dominant political firm but many special-purpose groups that cannot directly compete for power. Moreover, competition between the main “firms” in the political arena can be more or less regulated by norms that limit the permissible methods of competition. I know there is a well-developed theory of party systems for democracies, but it is not integrated with a theory of politico-industrial organization in other political systems; and we might want to put all of this in a single framework if we are interested in the history of the varieties of political competition.

Simplifying a bit, you might end up with a typology of varieties of political competition like this:

The table could be extended further; but it might get a bit too technical, and I’m not sure how illuminating it would be (more in part 2 with actual graphs and less theory, assuming part 2 ever arrives). The interesting questions would then be about which forms of politico-industrial organization can stably coexist with particular selection norms, and which of them actually produce good consequences, if any; but it might take me a while to get there, if ever.

A Very Short Quantitative History of Democracy, Dictatorship, and Other Political Regimes, Part I

(Part I of probably two).

Readers will have to forgive me, but I find dataset blogging addictive. One can use historical datasets to tell stories, not just to test models, yet outside economic history one hardly finds much quantitative history, much less quantitative political history, out there. Nevertheless, the Polity IV dataset I described in the previous post, with its long-run coverage and wealth of information about patterns of political authority at a global level, lends itself to the sort of quantitative history of political regimes I have in mind.  Though this sort of history is not always advisable, it provides a powerful antidote to the most common failings of what passes now for the history of democracy and other political regimes: excessive Eurocentrism, and annoying tendencies towards either Whig triumphalism or Hegelian determinism. (An exception here is Adam Przeworski’s excellent book Democracy and the Limits of Self-Government, whose self-consciously global perspective and use of long-run quantitative data makes it one of the most eye-opening books I’ve read on democracy and its history). My hope is that a little bit of quantitative history can mitigate some of the nonsense people seem to believe about democracy; and since I’ll soon start teaching my “Dictatorships and Revolutions” course again (more on this in a different post), I thought I’d try to use the Polity data to graphically chart the evolution of different patterns of political authority over the last couple of centuries for the benefit of students and perhaps others.

A couple of methodological points before starting. First, though the Polity IV data is pretty substantial, going back to 1800 in some cases, it does not track every single polity within that period. The focus is on nation-states, and indeed nation-states that have survived up to the present; lots of states that did not survive to the present time (because they were annexed by other states or disappeared through other processes) are not included, though some historical polities are (Prussia, Bavaria, Wuerttemburg, the Kingdom of the Two Sicilies, Sardinia, the Papal States, Modena, Parma, Tuscany, a few others). Colonial dependencies of these independent states are not coded; the dataset codes only the regime at the imperial center (though we can correct for this bias to some extent, as we shall see). Moreover, some of the polities that are included in the dataset from 1800 onwards (Austria and Turkey, for example) have experienced so much change (from Austro-Hungarian empire to Austria, and from Ottoman empire to modern Turkey) that one doubts the wisdom of having a single time series for them (rather than, for example, a time-series for the Austro-Hungarian empire and another for Austria). And Polity does not collect information about “micronations” (less than 500,000 inhabitants), which means that its coverage of Oceania (Polynesia, Micronesia, and Melanesia) is spotty at best. Finally, it is also worth noting that in the long span of time covered by the dataset many areas of the world, some of them incorporating substantial populations, were effectively stateless: James C. Scott’s “Zomia” in Southeast Asia is one of these regions, but every continent has had (and sometimes continues to have) large non-state spaces. Statelessness does not mean that people live without political authority, but authority is far more fluid (and often has different implications) when exit is relatively easy, as it has historically been in stateless areas, than when exit is more difficult, as it has often been in state zones.

Second, as I was saying in the previous post, the dataset does not track every feature of political regimes that might be of interest. It purports to measure only three general concepts: the mechanisms of executive recruitment in states (how leaders come to hold power over a central state apparatus), the forms of political competition (how groups contend for control over a central state apparatus), and the degree of executive restraint (how much the power of the political leaders at the centre is limited: political discipline). Though all three measures are very highly correlated (above .99), I strongly suspect that while executive recruitment and political competition do measure fundamental if related aspects of the political regime, executive restraint is best understood as a function of the other two, plus temporary changes in the configuration of political forces that are not part of the political regime properly speaking. In other words, to the extent that “executive restraint” is not simply picking up paper constraints (whether the constitution says this or that about a political leader), it must be picking up the ways in which the groups that play a role in political competition and selection are able to sanction the main government leaders in the state. So in what follows I will mostly ignore the measure of executive restraint and focus on executive recruitment (this post) and political competition (in part II).

Let’s start with the mechanisms of executive recruitment. In the Polity classificatory scheme, leaders can come to power in two basic ways: through unregulated political activity (your classic coup d’etat, for example) or through some norm-regulated process (like hereditary succession, competitive election, etc.). Norm-regulated processes can in turn be divided into those where political leadership is allocated, at least in principle, on the basis of “ascriptive” characteristics (like what family you were born into), and those where political leadership is allocated by explicit selection within a particular group. (No modern polity allocates political leadership by lottery, unlike many ancient ones; this may be a modern mistake). Finally, selection can occur through relatively open competition within a relatively large group (election, though not necessarily universal suffrage election), or through informal processes within small groups ("designation;" e.g., like the process of selecting the First Secretary of the Communist Party of the Soviet Union). These three dimensions of political competition lead to a seven-fold set of categories: pure ascriptive (absolute hereditary monarchies without powerful prime ministers); ascriptive plus small-elite selection (absolute monarchies with powerful prime ministers, for example); ascriptive plus large-elite selection (powerful monarchs confronting powerful elected prime ministers, for example); pure small-elite selection (single party regimes, for example); pure large-group selection (competitive electoral regimes); small group plus large-group selection (what polity calls “transitional” or “restricted” election regimes, though they are often not very transitional at all but merely regimes where small elites cannot select the leadership without some form of large-scale electoral competition, even if unfair); and self-selection regimes (unregulated selection). Polity also has a confusing “executive-guided transition” category that I don’t much like, but basically indicates a period of transition from a self-selection regime to a norm-regulated selection regime. (In fact, for most purposes it can be replaced by the self-selection category, since it does not actually indicate a change in executive selection mechanisms and relies on uncertain analyses of leadership intentions). So what do we see when we look at the evolution of political authority through this lens?

Fig. 1

Vertical lines and shaded areas indicate, from left to right, the First World War, the beginning of the great depression, the Second World War, the beginning of African decolonization, and the end of the Cold War and the breakup of the Soviet empire. Black/grey areas at the bottom of the graph indicate the number of countries falling into Polity’s three “transitional” categories: regime transition, anarchy, and foreign occupation. You must also imagine much of the white space in the graph – representing lots of colonial possessions – to be colored dark red and green, since rule over colonial possessions was exercised through limited elite and ascriptive selection regimes, even if regimes at the imperial center were different.

A few things are worth noting. First, there is a slow but steady trend towards more large-group selection regimes – relatively competitive elections of all types, even if mixed with small-elite selection (as in competitive autocracies). These elections are not always fair [update: and suffrage is not always universal or even close to universal], but by my count nearly 70% of all regimes in 2010 involved some form of meaningful competition for executive power within large electorates, while only 30% or so did in 1900:

Fig. 2

Large-group competitive selection is now the norm, not the exception, a change that happened over the course of a century but begins in the 19^th century. But though the overall trend is clear, the proportion of large-group competitive selection regimes fluctuates quite a bit, apparently in response to major political events: the two world wars (one can identify them just by looking at the spikes in regime collapses), the beginning of decolonization, the end of the Cold War and the collapse of the Soviet Union. Interestingly, the great depression barely rates a blip – it hardly affects the trends in executive recruitment patterns. And the current trend towards more large-group selection regimes starts in the mid 1970s (the oil shocks? The exhaustion of the appeal of single-party regimes?), though it appears to accelerate by 1989. Oddly, the 70s are also the great age of coups (“self-selection” regimes – lots of these emerged with decolonization) as well as the apogee of limited-elite selection regimes (single-party regimes, especially). The data thus seem to point to a future where most regimes recruit their leaders through electoral competition (not necessarily fair!) appealing to large groups, but there is still a substantial minority of ascriptive recruitment regimes (monarchies, basically) and limited-elite selection regimes; it’s as if the only long-term stable equilibria are either competitive elections or full-blown monarchies.

Moreover, the phenomenon is pretty much global: competitive regimes with elections that appeal to large electorates are now found in every continent:

Fig. 3

(Includes both competitive electoral regimes and competitive autocracies [mixed large group/small group selection]; the picture is not substantially different with only full competitive electoral regimes included). But it was never just a European phenomenon: competitive regimes (even if not always very stable and electorally fair) are first of all an American phenomenon (and I don’t mean North American). Most large-group electoral selection regimes in the 19^th century were in the Americas (the USA, Canada, Peru, Colombia, Chile, Costa Rica, Honduras, Bolivia, Argentina, Guatemala), though we even find some in Africa (Liberia for a time, which of course was in part an American import). Genuine electoral competition for power with an appeal to large electorates is a New World invention, but not a specifically North American one, even if the North American version of the experiment proved relatively more stable than many of the South and Central American versions. To be sure, the fact that a regime is competitive insofar as executive recruitment requires an appeal to a large electorate does not mean that it is a democracy in the full sense of the term (however you want to define it); many of these competitive regimes included important restrictions on suffrage (slaves and women needed not apply), and elections were not always fair or fully free. But all of the regimes in figure 3 are fundamentally different in kind, at least with respect to the mechanism of executive selection, from regimes where leaders are selected either by ascription or by informal competition within a small elite.

Here’s a more fine-grained picture of the distribution of such regimes since 1950:

Fig. 4. Competitive regimes per region.

(As I mentioned earlier, coverage of Oceania is pretty sparse in the dataset, so you might as well ignore the Melanesia, Polynesia, and Micronesia cells). As we can see, these regimes are now common basically everywhere; the only laggards are Central and Western Asia, where large-group selection competitive regimes (let's not speak of democracies, however) are still less than 50% of the total.

Non-electoral regimes – monarchies, single-party regimes, etc. are now almost exclusively found in Asia and Africa, though they used to be pretty evenly distributed throughout the world:

Fig. 5

Interestingly, the only real shocks to the distribution of these regimes seem to have been decolonization in the 1960s and the end of the Cold War in 1989. It is as if new countries generally end up with limited selection or unregulated recruitment regimes, and it takes a while for them to move either toward large-group selection or full monarchy. 

We can also look at this history in combination with the history of economic development. Using data from the Penn World Tables (caveat: some countries have no income data, and what data exists only goes back to 1950 for most countries), we can see that the rise in competitive selection regimes is visible in every income quantile except the highest:

Fig. 6

The proportion of competitive regimes in the richest quantile seems to be declining over time (and stabilizing in the third quantile) at about 80% of all regimes, as stable oil monarchies and other limited-elite selection groups rise to the top of the income distribution. As Przeworski has argued, there seem to be diminishing returns to conflict over executive selection mechanisms in rich countries, so all regimes should be relatively stable at high levels of income per capita. To be sure, full electorally competitive regimes (with “free and fair” elections) are less common in poorer countries, but even if we restrict ourselves to these regimes, we get basically the same picture:

Fig. 7

We can actually investigate this further by looking at the “transition matrix” of regimes over this period of time per income quantile. We basically look at what the regime is like at time t, and then what it is like at time t+1, and make a matrix, where each cell represents the percentage (number) of cases over the period in question where a mechanism of political selection in the rows changed to one in the columns (so the diagonal represents stability):

(Click here for a full spreadsheet version). Here we see that mechanisms of executive recruitment in countries in the poorest quantile remained stable about 90% of the years in question (i.e., on average they switched to another selection mechanism about once every ten years); by contrast, at the richest quantile, only self-selection regimes were stable less than 95% of the time. Competitive electoral regimes in the richest quantile were stable basically 100% of the time; but all other regime categories were also stable, and hereditary monarchy was basically just as stable as democracy at this level of income (more than 99% of the time). Even state breakdown appears stable in the richest quantile; those puzzling 13 years of stable state breakdown in the second table represent Lebanon, which appears to have maintained a relatively large income per capita during the years of civil war (though note economic data is likely to have been spotty and unreliable during that time, so take that factoid with a large scoop of salt).

Moreover, rich countries were likely to transition to competitive regimes if they transitioned at all; and the regimes most likely to transition where the self-selection regimes, whereas in the poorest quantile we find transitions to a wider variety of other regimes.

So where does that leave us? More in part II (looking at the forms of political competition over this period), but basically I think what we see is the political manifestation of the long process of global economic change since 1800 (the industrial era). We start with ascriptive selection and limited elite selection regimes everywhere in the world (or at least in zones of state power), but economic change slowly alters the basis of political power in them (more so in certain kinds of economies than others). These changes render norms of political selection unstable, and more particularly they make it possible for elites (new and old) to “defect” from previous norms of selection by appealing to larger groups in the selection process (this, generally, means elections). And global political shocks (the world wars, the breakdown of empires) can push the process in particular directions at least for a time – e.g., towards single party regimes in the wake of WWII and decolonization, or towards electorally competitive regimes after 1989. But after countries reach a certain level of income, political competition over executive recruitment is less useful; so all regimes eventually stabilize. 

Code necessary for replicating the graphs in this post, plus further graphs and ideas for analysis, in my repository here. (Still pretty rough, though). You will need to download the Polity IV and Penn World Table datasets directly.

[Update 21/2/2010: added a few small clarificatory remarks about how comeptitive elctoral regimes are not necessarily democratic in the full sense of the term]

The Half-life of Leaders and the Half-life of Regimes

Thinking back on the last couple of posts, a couple of questions arise naturally. First, there is the question of the survival of regimes in general, not just democracy: if most democracies die within 15 years or so, what is the median duration (the “half-life,” if you will: the time it takes for half of them to be gone) of other regimes? And second, there is the question of the relationship between the half-life of regimes and the half-life of leaders: do regimes whose leaders tend to have longer half-lives also have longer half-lives? My interest in these questions stems from my current research on the question of legitimacy: my sense is that legitimacy matters much less than people usually think to the survival of large-scale patterns of political power and authority, so I’m interested in trying to figure out if there are systematic differences in survival between more and less “legitimate” regimes and other political structures. So this is another exploratory post, with lots of graphs.

How do we measure the duration of non-democratic regimes relative to democratic regimes? Though democratic regimes are not always straightforward to identify, non-democratic regimes come in a much wider variety of forms – from hereditary, absolute monarchies to single party regimes and multiparty hybrids, and some of these forms shade gradually into one another over the course of many years. (For a sense of this variety, consider the differences between Mexico before the 1990s under the PRI, whose presidents succeeded each other with clockwork regularity every six years and a lively opposition existed but could never win the presidency, North Korea today, where opposition is non-existent and succession is controlled by a tiny clique, and Mubarak’s Egypt.) To get a handle on this question, I’m going to use the Polity IV dataset, which codes “authority characteristics” in all independent countries (with population greater than 500,000 people) from 1800 to 2010. (I’ve been convinced by Jay Ulfelder’s work that the DD dataset I used in my earlier post is not appropriate to study comparative regime survival due to the way it codes certain democracies where alternation in power has not occurred as dictatorships, which systematically biases the survival estimates of democracies upwards).

The Polity dataset is fairly rich. Most researchers seem to use only the composite indexes of democracy and dictatorship it offers, but these indexes, while useful, do not have a strong theoretical motivation, as Cheibub, Gandhi, and Vreeland argue here. For my purposes, it is best to use the dataset to extract those authority characteristics of political regimes it purports to measure: the mechanisms of executive recruitment, the type of political competition, and the degree of executive constraint. Mechanisms of executive recruitment include hereditary selection, hybrid forms combining hereditary and electoral mechanisms, selection by small elites, rigged elections, irregular forms of seizing power, and competitive elections; types of political competition range from the repressed (all opposition banned, as in North Korea) to the open (typical of thriving democracies); and executive constraints range from unlimited to “parity” with the legislature. (See the Polity IV codebook for a full discussion). In theory, the dataset distinguishes eight kinds of executive recruitment mechanisms, ten types of political competition, and seven degrees of executive constraint, plus three different kinds of “interruption” (including breakdowns of state authority, loss of independence, and foreign invasion and occupation), leading to a possible 563 possible patterns of political authority, but these dimensions are all highly correlated (over .99); indeed, only 212 combinations of executive recruitment, political competition, and executive constraint actually appear in the date, most of them only once and for short periods of time, and it is obvious that some combinations do not even make sense. (And those that do make sense do not always capture all the information we would normally want about a political regime: Polity has no good measure for the extent of suffrage in competitive regimes, for example). But the dataset helpfully indicates how long each of these patterns last, so we can attempt a first cut at the question of the half life of regimes using a Kaplan-Meier graph:

The half-life of an “authority pattern” – a combination of an executive recruitment mechanism, a type of political competition, and a specific form of executive constraint – is 6.6 years, though the tail of the distribution is very long: some of them have lasted for upwards of a century. Switzerland, for example, has had the same authority pattern for 162 years, and Afghanistan retained the same authority pattern from 1800 to 1935 (a hereditary monarchy). As it happens, social and political life comes to be mostly structured in most places by the long-lasting patterns, but most patterns of authority do not last that long. Incidentally, at this level of abstraction there are no great regional differences in the half-lives of authority patterns, though it does seem as if authority patterns last slightly longer in Europe and the Americas than in Africa and Asia:

Yet an “authority pattern” is too amorphous a unit of analysis. We might get a better handle on the question of comparative regime survival by looking specifically at the mechanism of executive selection, since the manner in which the chief power in the state is selected is normally thought to be quite important and to have far-reaching consequences: whether supreme power is achievable by hereditary succession only or through designation within a closed elite or via competitive elections or some other means seems to have important consequences.

Of all the mechanisms of executive selection identified in the Polity IV dataset, only one, “Competitive Elections,” is unambiguously democratic by most people’s lights. Though within the dataset the fact that a regime has competitive elections is no guarantee that it will also have universal suffrage, for the most part “competitive elections” identifies most countries that most people think are democratic. We can thus calculate the duration of all periods of “competitive elections” and compare them to the duration of all “non-democratic” periods – those periods where executive selection happened through some other means. The details are somewhat tricky (see the code), but here are the results:

Some notes. As we might have expected from the discussion in the previous post, full hereditary monarchies (Russia under the Tsars, Saudi Arabia, Iran under the Shah, Portugal and Romania in the 19^th century, Nepal in the 19^th century, among others; there are 65 episodes in 40 countries in the dataset) have the longest half-lives (nearly 32 years; this increases if we collapse the two hereditary monarchy categories. Note these are not “constitutional” monarchies like the British one). But competitive electoral regimes are no slouches, with a half-life of about 17 years (in keeping with Jay’s numbers in this post, though he uses a different dataset), and as time goes on their survival rates seem to converge with those of monarchies. Similarly, “limited elite selection regimes” (e.g., single party-communist regimes, where a narrow clique selects the leader without open competition) have a half-life comparable to that of democracies, but as time goes on they tend to break down more; their survival rates seem to diverge from those of competitive electoral and monarchical regimes. Low survival rates are found especially among political forms that appear to have internal tensions, such as competitive authoritarian regimes, where elections exist and are contested by an opposition, but it is very hard for the opposition to attain real power (e.g., Zimbabwe today). I confess I don’t really understand Polity’s “Executive-guided transition” category, but it’s obviously a regime that is turning into something else (the Pinochet regime in Chile after the 1980 referendum but before the return of competitive elections counts, for example), and “ascription plus election” includes regimes where the monarch retains some real power but the legislature and other executive offices are no longer under its thumb  (only a few are recorded in the data, including Belgium in the late 19^th century and Nepal in the 1980s and 90s); it makes sense that such regimes, halfway between “real” monarchies and purely constitutional monarchies like the British, should have short half-lives as the conflict plays out and either turn into competitive electoral regimes or into more absolute monarchies.

It is also interesting to compare the relative survival rates of competitive electoral patterns of authority vis a vis periods where selection happens by non-competitive electoral means (regardless of whether the selection means stay the same):

Though the difference seems to narrow as time passes, the half-life of non-democracy since the 19^th century has been a bit longer than the half-life of competitive electoral regimes (23 vs. 17 years). In sum, political regimes do not last much more than a generation.

(For those still following, the regional breakdown indicates that competitive electoral periods have had the longest half-lives in Europe and the Americas, whereas non-democracy has had the longest half-lives in Africa and Asia; no special surprises there, though I am not sure about the reason).  

How does this relate to the half-life of leaders? For that, we turn to the ARCHIGOS dataset by Goemans, Gleditsch, and Chiozza, which contains information about the entry and exit date of almost all political leaders of independent countries in the period 1840-2010. It’s a fantastic resource – more than 3000 leader episodes, and information on their manner of exit and entry. And the conclusion one must draw from examining it is that power is extremely hard to hold on to; a ruler’s hold on power seems to decay in an exponential manner (note I haven’t checked that the decay really is exponential in the technical sense, though I'm thinking of doing that). Over this vast span of time, covering all kinds of political regimes, the half-life of leaders is only about 2 years, or a third of the median authority pattern, as we might have expected from the previous post (though the half-life of leaders is even smaller here):

Yet of course it is the people who beat the odds – those who last much longer than the average leader – the ones who shape social and political life. (There’s an endless parade of mediocrities in the dataset, two-bit prime ministers gone after a few months of ineffectual dabbling and the like).

(But don’t some leaders come back to power after losing it? In fact, the vast majority of leaders only attain power once, and never return to power, though about 100 did manage the feat three or more times. In fact, practice does not help; survival in power only appears to decrease the more previous times the leader had been in power, though note that the uncertainty of the estimates also increases, and one might expect that age would take its toll too).

We are now in a position to extend the analysis in the post below by merging the Archigos and the Polity dataset to calculate the survival curves for leaders conditional on the pattern of executive recruitment. Though I would take these curves with a grain of salt, here are the results:

As expected from the previous post, it’s good to be king – the half-life of absolute kings is about 12 years (and it’s almost always king: there are only 41 female leaders in a 3000 case dataset). Interestingly, a similar result for the half-lives of Chinese emperors is reported here (10 years: Khmaladze,  Brownrigg, and Haywood 2010, ungated) as well as for the half-lives of Roman emperors (11 years: Khmaladze,  Brownrigg, and Haywood 2007, ungated). There is something about the deep structure of monarchies in many different periods and societies, it seems, that points to a half-life in power of about 10-13 years for monarchs. 

More generally, authoritarianism pays in terms of leader tenure, despite the fact that non-competitive regimes do not always last longer than competitive ones. The highest half-lives of leaders beyond monarchs are found in limited elite selection regimes, executive-guided transitions (where non-democratic leaders are changing the rules), and competitive authoritarian regimes; but democracies are more lasting than most of these regimes (except for monarchies; see above).

Another way of looking at this is to calculate what we might call the “personalization quotient” of a regime: divide the half-life of the leader (for a given regime) by the half-life of the regime to get an idea of the percentage of the regime half-life that a leader is expected to last. So a monarch is expected to last about 37% of the half-life his regime (31.86 / 12); this is the most intensely personalized of regimes, as one might have expected given that it is devoted to the maintenance of a family line. The next most personalized regimes are competitive authoritarian regimes (28%), “self-selection” regimes (15%), limited elite selection regimes (16%), and “executive-guided transitions” (40%; this is pretty much by definition, however, so I don't make much of them). A competitive electoral regime has a personalization quotient of 8% - an expected leader half-life of about 2, divided by an expected regime duration of about 17 years. From the point of view of such a leader, it pays to try to move towards a competitive authoritarian regime, and it pays for the leader of a limited elite selection regime to move towards a formal hereditary monarchy (as is happening, in a sense, in North Korea right now, and almost happened in Egypt and Libya). 

But are authoritarian regimes more risky, so that leaders will try to hang on to power more? We can also look at that using the archigos dataset. Though leaders in non-democratic regimes have a slightly higher risk of leaving office with their heads on pitchforks or hanging from lampposts, the vast majority leave by "regular" procedures.  

More, perhaps, could be said. I’ve been wondering, for example, about whether there is a relationship between the breakdown of particular regimes and the tenure of leaders, though I’m not sure how to go about tackling that question. From the point of view of the study of legitimacy, however, what strikes me is the general fragility of patterns of authority and rule: few patterns of authority are expected have half-lives that exceed a single generation, and most don’t last nearly as long, regardless of their “legitimation formula” – heredity, competitive elections, ideology, whatever. Of course, some beat the odds, especially some competitive regimes and some monarchies, and these shape history. But the historical evidence suggests that they are in a sense the exception rather than the rule.

Code necessary for replicating the graphs in this post, plus further ideas for analysis, here and here. You will need to download the Polity IV and ARCHIGOS datasets directly, and this file of codes from my repository.

[Update: fixed some typos,  9 Feb 2012]

Comparative Political Leader Survival, 1946-2008

After playing around with Jay Ulfelder's data on the survival of democracy in the previous post, it occurred to me that I have not seen survival estimates for leaders in different kinds of regimes like the ones he discusses for democracies. So, in the spirit of exploratory data analysis, here are some graphs using data from the DD dataset of political regimes by Cheibub, Gandhi, and Vreeland, which provides information about regime type, effective heads of government, and leadership tenure for most countries in the world for the period 1946-2008. (Fuller data and methods note at the end of the post).

First, let's look at a simple estimate of leader survival for all (effective) political leaders in all regimes in the post WWII era:

The figure shows an estimate of the proportion of leaders who are expected to still be in power after n years. So, for example, after four years in power, less than half of all leaders are expected to still be in power, and after 20 years less than 10% of all leaders are expected to still be in power; the majority of all leaders last less than 4 years in power, and the vast majority less than 5. [Update: of course, some of these leaders come back to power after a shorter or longer period out of power.] This may be easier to see if we draw the plot on a logarithmic scale:

This looks like a classic "long tail" distribution of a kind often produced by "rich get richer" processes: most leaders don't last in power very long, but those who beat the odds can do very well indeed, as power feeds on itself and leaders become increasingly difficult to dislodge. (I won't say anything about power laws for fear of attracting the ire of the statistical gods). 

Nevertheless, democratic leaders and non-democratic leaders aren't equally successful at hanging on to power:

While the median democratic leader can expect less than 3 years in power, the median autocrat can expect a bit less than 7. And the gap widens with time: less than 8% of all democratic leaders can expect to hang on to power for more than 10 years, but more than 40% of autocrats do, and no democratic leader in the sample has lasted more than 25 years in power (Lynden Pindling of the Bahamas and Eric Williams of Trinidad and Tobago; your mileage may vary as to how democratic you think they were, but that's how they are coded), whereas nearly 20% of autocrats do. This may seem obvious (after all, autocrats typically impose larger barriers to political competition than democratic leaders, and ordinary people face larger obstacles in trying to get rid of them) but it also presents a bit of a puzzle, for democracies are supposed to be more responsive to popular wishes and more legitimate, and dictators are always at risk of being overthrown by their close associates. (For one influential explanation of the observed pattern of survival by Bueno de Mesquita, Smith, Siverson, and Morrow, see here and here). The greater legitimacy of democratic leaders, and their closer connection to popular opinion (to whatever degree: let's not exaggerate, either), does not seem to translate into a surer hold on power. 

Not all autocrats do equally well; absolute monarchs are especially successful at holding on to power:

Though the uncertainty of the survival estimate is larger for monarchs than for other regimes (there are just fewer monarchs in the sample) their advantage is large enough to be noticeable above the noise: nearly 60% of all monarchs can expect to last 20 or more years in power, while only 20% of other autocratic rulers can expect to survive that long, and less than 1% of democratic leaders can hope for such a career. This is another reason to think the Middle Eastern monarchs are probably safer from being overthrown than the leaders of the "republican" regimes, as Victor Menaldo has recently argued. His argument points to specific features of the political culture of these monarchies that enable elites to better monitor and discipline leaders; but other things may be going on as well (monarchs elsewhere in the world also appear to have done well, so whatever enables monarchs in the Middle East to survive appears to also work elsewhere, though admittedly most of the world's absolute monarchs since 1946 have been concentrated in the Middle East). It is also interesting to note that military and civilian dictators do not differ (much) in terms of their survival expectations (the estimates fall within each other's 95% confidence intervals), despite theoretical and empirical work that suggests that military regimes are less stable than civilian dictatorships. (Of course, this could be due to any number of things, including problems with the coding of the data and the fact that the stability of regimes is a different thing from the stability of any given leader's grip on power).

I was also curious to see whether the survival of leaders differs across regions of the world. And at least for non-democratic leaders, that seems to be the case:

There's a lot of uncertainty in these estimates (and I could have made a mistake), but in general it seems to be the case that autocratic leaders have had less success hanging on to power in Latin America, despite the USA's not always benevolent influence in the region. That was surprising to me, so perhaps someone will tell me why this is wrong. By contrast, democratic leaders all have very similar survival expectations all over the world; no evidence of "regional" effects seems evident:

Now that I've mentioned the USA's influence, we might as well look into whether autocrats (or democratic leaders) have had more trouble hanging on to power during or after the cold war. Surprisingly, it seems they have not: leaders in both regimes had the same survival expectations in both periods. But this was tricky to figure out how to calculate, and it is the most likely spot where I might have made a mistake (see sources and methods note below):

Sources and methods. A full description of the DD dataset can be found here, including the criteria it uses for categorizing regimes as democratic or non-democratic and a general defense of its methodological approach. (It used to be possible to download it as well from that page, but the form no longer seems to be working. I've animated the dataset here.) These criteria have been criticized for a variety of reasons, but in general DD does not suffer from worse problems than many of the other common datasets of political regimes (like Polity IV or Freedom House). It is possible that some of the coding decisions they make might influence the estimates of survival presented above, e.g. because they err on the side of classifying some regimes as dictatorships that could have been considered democratic (when there has been no alternation in power). This would tend to bias downward the survival estimates of democratic leaders. At any rate, DD includes information about leaders and their tenure, which is missing in other datasets and makes the data-wrangling easier, though this information is not always complete (there is sometimes more than one leader in a year for a given country, a fact that the dataset must omit, given its country-year resolution) and is not quite in the right format for survival analysis. I thus had to reshape  it (R code and a general description of the process; rank amateurism on display). I created three data files: one for the plots of survival for all leaders and leaders by democracy/non-democracy (ddsurvival.csv); one for the plots of survival by autocratic regime type (ddsurvival2.csv); and one for the plots of survival during and after the cold war (ddcoldwar.csv). (R code for generating all plots is here). These files treat the leader spell as a case; "right censoring" occurs when the leader dies or if the leader is still in power by 2008 (see the DD codebook; the files use a variable called ecens2). Since DD does not distinguish between deaths by natural causes and political assassinations or death in revolution, this introduces a certain amount of bias; in theory, "political" deaths should not result in"censoring" of the data. I should note that the plot of survival by autocratic regime  type does not take into account some cases where "left censoring" occurs (i.e., when a regime starts before 1946), though the number of cases where that is a problem is very small. Finally, there are a small number of repeated cases in ddsurvival.csv and ddcoldwar.csv due to problems guessing the right "entry date" for the leader; these must introduce some small amount of error, though I couldn't possibly say how much or in what direction the bias would work.

[Update, 1/31/2012: Fixed minor typos]
[Update, 1/02/2012: Changed location of code and data files]