Chris Blattman

Corruption, revisted

One of the topics I’m most notorious for blogging about is the topic I know the least about: corruption. Basically, I’ve said it’s a second order issue, and a Western fetish. Here is a full slate of posts.

Matthew Stephenson, Harvard Law professor, has a new anti-corruption blog

As for citizens (especially poor citizens) in developing countries, if we want to know what they think about corruption, we could ask them–as indeed we have (see, for example, herehereherehereherehere, and here). And they consistently rank corruption as among their most significant concerns.

…I’ll throw out my own hypothesis about why so many academics in the blogosphere are drawn to the anticorruption-is-a-Western-obsession-that-doesn’t-matter-much-for-development canard: academics (and I speak as a member of the tribe) enjoy feeling like iconoclasts willing to speak uncomfortable truths to power. And in the development field, a certain type of academic particularly enjoys attacking anything that the major institutions (World Bank, U.S. government, OECD, etc.) seem to be for.

There are fair points throughout. I’ll make two in response.

  1. Second order issues are still important issues. The problem with most development policy is that everything is important. Everything matters for development. Setting priorities is hard. I haven’t seen the evidence that says corruption is one of the top 3 things the World Bank President or head of USAID should care about. Citizen surveys are useful, but not fully persuasive.
  2. Corruption strikes me as a symptom of deeper governance problem. Many if not most anti-corruption policies strike me as treating symptoms not the disease. Acemoglu and Robinson put it well: “corruption is a way for many economists and policymakers to talk about bad political outcomes without talking about politics.” I think we should be thinking about what builds bureaucratic capacity, how to restrain over-centralized regimes, and the political incentives leaders have to use patronage and repression to control their societies. Transparency might be part of that. And less corruption would be an important outcome. maybe. It’s not clear to me that either ought to be front and center policy aims.

41 Responses

  1. Africans describe corruption as they are eating the money. I am surprised you think it is a western fetish. They also call the corrupt by another name – thieves. It is certainly not a western fetish. They hang people in China for it. Although I don’t condone that level of punishment for corruption. I think that it is clear it is hardly a Western fetish. It is against the law in most of the world.
    As for your point on bad governance and A and R’s point about bad politics being causes or more important, I think people just say they are thieves eating the money.

  2. My impression is that corruption – as a point of citizens’ outrage – reflects their sense of injustice as to how resources are distributed in their society. Currently CEO pay captures a lot of attention in North America but is it really the cause of unemployment or stagnant wages or an effect of a particular system that is not delivering broad gains?

  3. Chris,

    Thanks for the response. Much as I took issue with your earlier posts on corruption, I find myself largely in agreement with the two posts you make here.

    On your first point: I wholeheartedly agree that citizen surveys, though useful, have their limitations. I invoked them principally as a response to the sweeping statements I’ve seen in some posts suggesting that most people in the developing world don’t care that much about corruption. (I see this more in posts others than yours, but I think your “Anglo-American fetish” language does imply that view.) I also agree with your point that there are many important issues, and that setting priorities is hard — though I would also argue that corruption is one of those issues that is bound up with a lot of the other issues you and others think are important (property rights, health, etc.), which is why I do think it’s a significant across-the-board issue. But I entirely agree that the need to prioritize requires us to have these exchanges.

    Your second point is well-taken. It’s possible that corruption is associated with, but not the main cause of, broader governance problems. Some fairly good research tries to disentangle these questions, but I agree it’s hard, and indeed I view this point as the most plausible rebuttal to the evidence that corruption is very bad for a range of development outcomes. But, still, three quick points in response: (1) plausible as the response is, I do think there’s fairly good evidence that corruption qua corruption is often a significant problem; (2) if the policy responses to corruption and to the deeper governance problems (the “root causes”) generally converge (on things like more accountability, more bureaucratic capacity, etc.), then we might ultimately end up in more or less the same place, and the question whether the problem is “corruption” or “bad governance” starts to become semantic rather than substantive; (3) I actually don’t think I agree with Acemoglu & Robinson that corruption is a way for people to talk about bad political outcomes without talking about politics — it seems to me the truth is closer to the reverse: in many circles there is (or has historically been) a reluctance to talk about corruption precisely because it seems to political (and accusatory), and so people often substitute euphemisms like “governance”.

    One more thing: I’m not sure if your post included a direct link to my original commentary. For readers interested in checking it out, that link is:

Why We Fight - Book Cover
Subscribe to Blog