Chris Blattman

Search
Close this search box.

IPA’s weekly links

Dress like Abhijit
"Can you keep your medicaid in Arkansas 'game'"

Guest post by Jeff Mosenkis of Innovations for Poverty Action.

  • There’s a new evaluation out of the Northern Ghana site of the famous expensive Millennium Villages project most associated with Jeff Sachs. I’m not an expert, but as I understand it, the theory is that an intensive big fix (building new institutions like hospitals and many other things at once) could fix the interdependent problems of poor areas.
    The thing is that Sachs insisted he knew it would work, and it didn’t need an independent evaluation, in fact threatening people who criticized the project’s inadequate evaluation, and yelling at a reporter (disclosure: she’s now my colleague) who asked for the full budget numbers.
    My understanding is that in addition to the outside funding Sachs brought, it required local government funding and free labor, so also came at a cost to the local region.
    DFID, the UK aid agency, disagreed and did insist on this evaluation, which finds “small or null” effects (I couldn’t find an ungated version but here’s the DFID report version)
  • It’s job market season, lots of good job market papers on the Development Impact Blog. In addition, two JMPs/JMC’s which caught my attention:
    • Moffii Odunowo has a couple really interesting papers, one thinking about the longer lasting impacts of women’s education in Nigeria using natural geographical variation in a 1970’s-era educational reform. Every additional year of school they got increased:
      the probability of children completing primary school by 22 percent, and attending secondary school by 29 percent. I find that the effects are particularly pronounced for girls.
    • One of her other papers also looks at the effects of exposure to Boko Haram attacks on children’s cognition:
      Children exposed to terror attacks are 0.35 SDs shorter and lag in cognition by 0.18 SDs. The deficits are largest in children exposed to violence at younger ages. Mediation analysis shows that 6% of the effect on height is mediated by nutrition and parental investment can explain 14% of the effect on psychological development.
      According to her website, she’s got an RCT and several other interesting projects in the works, which I’m looking forward to seeing.
    • Wayne Sandholtz blogs about his JMP, looking at electoral consequences of an existing Liberian school reform. As it happens, the program was being RCTed, and he ran a parallel study on voting and opinions. He found that parents of kids in the new schools responded to how well the school was doing with their voting, rewarding the incumbent if the new school was performing well, and punishing if it wasn’t.
  • Relevant for current U.S. announcements that the administration will take away SNAP food benefits for 700,000 poor adults, and school lunches from nearly a million children (some can reapply), is this brief from Ideas42 on work requirements for benefits. The evidence showed they don’t help and just put up barriers (see the “game” on page 14 to see if you’d be able to keep your Medicaid benefits with Arkansas’ barriers). I’ve never seen any evidence that punishing the poor out of poverty works. In fact, one of the themes of Abhijit Banerjee’s Nobel lecture was that the poor aren’t poor because they’re lazy. His donation to the Nobel Prize museum was bags sewn by women in Ghana from a study with my IPA colleagues showing that poor women who got more cash aid became more, not less, productive at work. Also see this part of his Nobel lecture, where he discusses findings from 7 countries where social welfare programs didn’t reduce work.
  • Rachel Strohm has a nice piece out about what people mean when they say being in “the field.” She concludes that it’s shorthand for geographic and power dynamics. I’ve noticed the same thing, and started just saying where I mean (the name of the place or “the study locations”)
  • Kenya’s new railway from the port has increased costs by nearly 50% for some shippers, due to the high fees needed to repay the Chinese backers of the project. Shippers are being required to use the new railway.
  • The Nobel lectures by Banerjee, Duflo, and Kremer were all very watchable, but my favorite part is probably the fashion articles from India about the traditional clothing Banerjee and Duflo wore to the award. I’m guessing it’s the first fashion article about MIT economists.