Chris Blattman

Local politics a tough nut to crack

Donors push “community driven development” programs largely to strengthen local institutional capacity, democracy, and inclusiveness. (Sometimes overlooking the fact that these three goals are not necessarily in harmony.)

It’s nice to see these claims being put to ambitious test. Casey, Glennerster and Miguel evaluate one such CDD program in Sierra Leone:

We find positive impacts on the establishment of local development committees, local public goods provision, links between communities and local government officials, household economic welfare, and village-level market activity. However, we do not find any program impacts on community social norms, the role of women and youths in local affairs, more egalitarian decision making or the capacity for collective action beyond the immediate project sphere.

Overall, these findings suggest that community driven development programs and related donor projects may leave communities materially better off but may be less effective in fundamentally transforming local institutions or power dynamics.

This matches my intuition. The CDD-can-change-local-politics view has always struck me as the victory of rhetoric over reality, with a naive view of institutional and political change. There are perfectly good grounds for CDD programs in terms of economic impacts and fairness, plus (the possibility of) better decision-making.

There’s a dozen or more CDD evaluations completed or underway, including an upcoming book from Mansuri and Rao (see their 2004 critical review here). I don’t follow these debates closely enough to know the emerging conclusions. Reader insights welcome.

4 Responses

  1. Interesting questions and good investigations by my friends Kate, Rachel and Ted, whom I worked with in Sierra Leone. I’m also eager awaiting for Biju and Ghazala’s to unveil their meta study.

    I’m afraid that those of us in the tranches, if we read papers at all, may jump straight to the conclusion section of the paper (or a quote on your blog) and start quoting “conclusions”. E.g., many have quoted Olken’s “100% audit is better than participatory monitoring” “conclusion.” It may well be true for societies where wearing a badge or showing an official letterhead scares village elites more than (often unorganized) local protests. A repeat study of the same question in a diff setting may not be publishable, esp. if it confirms the conclusion of the ground-breaking paper. But we need more studies and also to understand why things work or not in different settings.

    The paper’s claim that “The evaluation conclusively rejects the idea that CDD is an effective method to initiate social change or fundamentally alter local hierarchies and decision-making processes” runs the risk of being “very quotable.” The value of this kind of context-rich research may not be to “conclusively reject” or “conclusively support” but rather to shake people out of their convictions and motivate us all to enquire more and figure out why local power structure is resilient to social engineering.

    On a separate point: Does India’s panchayat strengthening package (election of village governments, with 1/3 or 1/2 reservation for women, and block grants to village gov) count as CDD? It has the same flavor of social engineering as far as elections and grants are concerned, but do not include the “facilitation” aspect. Formal social engineering confers more power than an externally introduced CDD program. An interesting question is whether “facilitation” affects local politics in socially divisive settings like Bihar, where I work now.

  2. Perhaps not CDD directly, but I’ve done a good bit of work that looks at local social structure, development and adaptation. One pub ( examines the conservative tendencies of livelihoods, and a chapter in my book (out in February – shows how road construction triggered huge income shifts in two small villages, but no real changes in social structure, including gender roles.

Comments are closed.

Why We Fight - Book Cover
Subscribe to Blog