There Are No “Best Practices” for Democratic Transitions

I’ve read two pieces in the past two days that have tried to draw lessons from one or more cases about how policy-makers and practitioners can improve the odds that ongoing or future democratic transitions will succeed by following certain rules or formulas. They’ve got my hackles up, so figured I’d use the blog to think through why.

The first of the two pieces was a post by Daniel Brumberg on Foreign Policy‘s Middle East Channel blog entitled “Will Egypt’s Agony Save the Arab Spring?” In that post, Brumberg looks to Egypt’s failure and “the ups and downs of political change in the wider Arab world” to derive six “lessons or rules” for leaders in other transitional cases. I won’t recapitulate Brumberg’s lessons here, but what caught my eye was the frequent use of prescriptive language, like “must be” and “should,” and the related emphasis on the “will and capacity of rival opposition leaders” as the crucial explanatory variable.

The second piece came in this morning’s New York Times, which included an op-ed by Jonathan Tepperman, managing editor of Foreign Affairs, entitled “Can Egypt Learn from Thailand?” As Tepperman notes, Thailand has a long history of military coups, and politics has been sharply polarized there for years, but it’s still managed to make it through a rough patch that began in the mid-2000s with just the one coup in 2006 and no civil war between rival national factions. How?

The formula turns out to be deceptively simple: provide decent, clean governance, compromise with your enemies and focus on the economy.

This approach is common in the field of comparative democratization, and I’ve even done a bit of it myself.  I think scholars who want to make their work on democratization useful to policy-makers and other practitioners often feel compelled to go beyond description and explanation into prescription, and these lists of “best practices” are a familiar and accessible form in which to deliver this kind of advice. In the business world, the archetype is the white paper based on case studies of a one or a few successful firms or entrepreneurs: look what Google or Facebook or Chipotle did and do it, too. In comparative democratization, we often get studies that find things that happened in successful cases but not in failed ones (or vice versa) and then advise practitioners to manufacture the good ones (e.g., pacts, fast economic growth) and avoid the bad (e.g., corruption, repression).

Unfortunately, I think these “best practices” pieces almost invariably succumb to what Nassim Taleb calls the narrative fallacy, as described here by Daniel Kahneman (p. 199):

Narrative fallacies arise inevitably from our continuous attempt to make sense of the world. The explanatory stories that people find compelling are simple; are concrete rather than abstract; assign a larger role to talent, stupidity, and intentions than to luck; and focus on a few striking events that happened rather than on the countless events that failed to happen.

The narrative fallacy is intertwined with outcome bias. Per Kahneman (p. 203),

We are prone to blame decision makers for good decisions that worked out badly and to give them too little credit for successful moves that appear obvious only after the fact… Actions that seem prudent in foresight can look irresponsibly negligent in hindsight [and vice versa].

When I read Tupperman’s “deceptively simple” formula for the survival of democracy and absence of civil war in Thailand, I wondered how confident he was seven or five or two years ago that Yingluck Shinawatra was doing the right things, and that they weren’t going to blow up in her and everyone else’s faces. I also wonder how realistic he thinks it would have been for Morsi and co. to have “provide[d] decent, clean governance” and “focus[ed] on the economy” in ways that would have worked and wouldn’t have sparked backlashes or fresh problems of their own.

Brumberg’s essay gets a little more distance from outcome bias than Tepperman’s does, but I think it still greatly overstates the power of agency and isn’t sufficiently sympathetic to the complexity of the politics within and between relevant organizations in transitional periods.

In Egypt, for example, it’s tempting to pin all the blame for the exclusion of political rivals from President Morsi’s cabinet, the failure to overhaul the country’s police and security forces, and the broader failure “to forge a common vision of political community” (Brumberg’s words) on the personal shortcomings of Morsi and Egypt’s civilian political leaders, but we have to wonder: given the context, who would have chosen differently, and how likely is it that those choices would have produced very different outcomes? Egypt’s economy is suffering from serious structural problems that will probably take many years to untangle, and anyone who thinks he or she knows how to quickly fix those problems is either delusional or works at the IMF. Presidents almost never include opposition leaders in their cabinets; would doing so in Egypt really have catalyzed consensus, or would it just have led to a wave of frustrated resignations a few months down the road? Attempting to overhaul state security forces might have helped avert a coup and prevent the mass killing we’re seeing now, but it might also have provoked a backlash that would have lured the military back out of the barracks even sooner. And in how many countries in the world do political rivals have a “common vision of political community”? We sure don’t in the United States, and I’m hard pressed to think of how any set of politicians here could manufacture one. So why should I expect politicians in Egypt or Tunisia or Libya to be able to pull this off?

Instead of advice, I’ll close with an observation: many of the supposed failures of leadership we often see in cases where coups or rebellions led new democracies back to authoritarian rule or even state collapse are, in fact, inherent to the politics of democratic transitions. The profound economic problems that often help create openings for democratization don’t disappear just because elected officials start trying harder. The distrust between political factions that haven’t yet been given any reason to believe their rivals won’t usurp power at the first chance they get isn’t something that good intentions can easily overcome. As much as I might want to glean a set of “best practices” from the many cases I’ve studied, the single generalization I feel most comfortable making is that the forces which finally tip some cases toward democratic consolidation remain a mystery, and until we understand them better, we can’t pretend to know how to control them.

N.B. For a lengthy exposition of the opposing view on this topic, read Giuseppe Di Palma’s To Craft Democracies. For Di Palma, “Democratization is ultimately a matter of political crafting,” and “democracies can be made (or unmade) in the act of making them.”

Leave a comment


  1. Oral Hazard

     /  August 23, 2013

    Your reference to Google and Facebook case studies reminded me of yesterday’s headline from “The Onion”: “Economists Advise Nation’s Poor To Invent The Next Facebook”

    The formula for creating a lasting electoral democracy spanning 250 years following liberation from colonial rule really is deceptively simple:

    1) Establish independence from colonial power through armed struggle;

    2) Draft a consensus post-revolution constitution that fails;

    3) Draft a second consensus constitution that holds up a bit better, in part by leaving a number of factionaly points of contention vague or unaddressed;

    4) Survive a retaliatory war waged by the miffed former mother country;

    5) Have enough territory for expansion and natural resources to support enormous population growth;
    5a) Have a populace with a cultural tradition of high literacy;
    5b) Handle indigenous people’s resistance to said expansion/exploitation of natural resources with indifference/extreme prejudice;

    6) Avoid secession of half of your country by successfully fighting a 4 year civil war in which the side representing unification prevails in spite of amazing early military incompetence;

    7) Adapt social structures and legal/political framework as needed to adjust to radical changes in demographic concerns, such as slave emancipation, women’s suffrage, influx of poor immigrants, historically unprecedented technology disruptions;

    8) Adapt and adjust to exogenous disruptions to economy and security as needed, all the while accommodating and reflecting diverse domestic perspectives on matters such as religion, ethnicity, gender, and social justice.


  2. Rex Brynen

     /  August 23, 2013

    Unless you’re going to go so far as to argue that agency doesn’t matter (and I know you wouldn’t), then strategic political choices made from a range of available options matter, even when those options are constrained by structural conditions. The question then becomes “does comparative analysis give us any basis for determining what choices might result in the smoothest transition?” Here one might argue that contextual make cases so different that comparisons are impossible–but I can’t imagine any large-n quantitative analyst ever arguing that 😉

    In turn, if (1) agency and choices to matter in some way (even if marginally so–and I think effects are more than marginal), (2) comparative analysis holds some value (even if constrained by contextual differences), then I would argue that there IS some utility it trying to identify best practices–with an underlying recognition (to use all of the usual clichés) that there exist no slam-dunks, silver bullet fixes, or cookie-cutter solutions.

    • You’re right that I don’t mean to attribute everything to structure. Choices do affect outcomes. What I’m rejecting is the idea that we can identify specific choices or even broad strategies that will consistently produce similar results across diverse contexts. Pretty much all of the versions of these “best practices” of democratization I’ve seen have confused outcomes with strategies (“Forge pacts!”) or are so banal that they’re useless (“Compromise! Grow the economy!”). I think that’s because in the practice of politics, the specifics matter greatly. How you do these things is the hard problem, and there aren’t generic solutions.

      The other big problem I didn’t touch on in the post is that these lists and formulas are often predicated on the assumption that practitioners want democracy more than anything else, or at least pretty badly. Advice isn’t very useful if the end it’s meant to produce isn’t the one its intended audience is seeking, and I think that’s often a problem in transitional cases.

      In short, I agree that choices affect the outcomes of transitions, but for advice about those choices to be useful, it has to be based on an appreciation of the specific circumstances and objectives its audiences are confronting, and I don’t see how any generic set of “best practices” can do that.

      • Oral Hazard

         /  August 23, 2013

        Somewhere between horror and mischievous amusement that your post, which basically advocates employing something approaching rigor in using comparative politics to derive prescriptive suggestions across contexts, would be in the slightest bit controversial.

        It would be funny to look at much of the field as a veritable nudist colony of naked emperors if it weren’t for the policy implications of the writings of “highly credentialed” people.

  3. Grant

     /  August 23, 2013

    You might say something about crafting institutions that will be the way to resolve disputes that arise, but it’s been shown that even in the U.S. those can’t be guaranteed to work in times of serious partisanship (see the Civil War which happened at a time of both courts and elected officials).

    • Yes, and the deeper problem with thinking of institutional design as a solution to these troubles is that the design and production of those institutions is endogenous to the politics they’re supposed to “fix.” People have to sit down and do the designing and come to an agreement on what to implement and the do the implementing. So who gets to do the first part, and what are the decision rules? Then who are the agents in the second part, and how do you solve the principle-agent problems involved? In many cases, disagreements on the answers to those questions will reflect the very factional politics they’re supposed to fix, so we wind up right back at more parochial questions about how those factions can find enough common ground and trust to agree on something.

  1. Friday’s Reading List | Smoke & Stir
  2. Bosnia 2013 – a tale of two pictures | TerraNullius
  3. Most Popular Posts of 2013 | Dart-Throwing Chimp
  4. Reactions to Reflections on the Arab Uprisings | Dart-Throwing Chimp

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

  • Author

  • Follow me on Twitter

  • Follow Dart-Throwing Chimp on
  • Enter your email address to follow this blog and receive notifications of new posts by email.

    Join 13,612 other followers

  • Archives

%d bloggers like this: