Chris Blattman

Comment of the week: Should I give my salary to the poor?

After mentioning my upcoming field projects and travel, Lukas comments:

Dear Chris,
Are you sure you really add more value, compared to the scenario of just distributing your wage in Liberia? I’m aware that it’s not a fair comparison, but maybe something people in development should think about.

I love it. That is indeed the benchmark.

Probably the relevant comparison is not my salary (which the university pays, and where the counterfactual is not less poverty but rather more arcane research). We should be thinking about the cost of the research–which happens to be much higher than my salary.

The answer depends on whether the research serves a public good, and makes future aid more effective. My research “fails” the test if just one of three things is true:

  1. My research sucks
  2. I don’t publicize my results, or
  3. I do 1 and 2 but no one in the aid community reads it

Most researchers don’t have the incentives or skills for 2, and do a miserable job.

We all try to do 1, and it’s a mixed bag. There is a ton of crud research in development.

We can’t control 3, and that worries me at least as much as 1 and 2.

As it happens, Jeannie and I give 15% of our earnings away, so you might infer from that my self-rated answers to 1 to 3.

51 Responses

  1. What a cool question from a Lukas!

    Also so cool that you give 15%. Members of Giving What We Can give 10% of their income to the most cost-effective charities:

    @Michael, their website also has quiiite a lot of methodology on how to evaluate a charity.

  2. The corporate sector recognizes the importance and spends millions on research when it comes to increasing efficiency and getting product into the hands of consumers. There’s no reason why it can’t do the same for international development. It just has to be useful and the right research. Not research for research’s sake.

    P.S. Would love it if you could write about what to do when you want to get a PhD but your research interests are so interdisciplinary that they don’t fit neatly into one program!

  3. I’d love to hear how an economist who seeks to help the poor goes about deciding which charities to donate to

  4. Nice points all around. Some thoughts:

    @Vladimir: I think you’re right about the venture capital approach, but this hardly forgives the academics who don’t do the minimum to publicize.

    @Julie: I suppose I was lumping relevant research with “good” research, but finer distinctions are fair. We ought not to penalize basic research, however, which includes a lot of stuff that will look arcane. In the long run this can be more lasting and influential. If every cancer researcher tested drugs and no one mapped the genome, we’d be behind.

    @Eli: I’m pretty convinced I’ve had the most impact in the world by employing enumerators.

    @Arielle: If I thought the aid community read the very best of 1 and 2 I would find it easier to agree. People are busy and, frankly, don’t spend as much time learning and challenging themselves as they want and ought to.

    @Ron: A sabbatical in Africa is overdue. Co-locating with a spouse is likely to crowd out the Liberias in favor of the more routine places, but we will see.

  5. Better yet. Take a sabbatical by joining as a visiting faculty at the University of Liberia, advise or mentor doctoral students or junior faculty. Direct or co-direct doctoral research projects, and above help theses institutions establish world class data repositories of primary data to help guide a range of long-term follow-up for future researchers and scholars based at these schools.

    Bottomline- genuinely strengthening tertiary education would boost your impact. And above all avoid the “capacity building” claptrap so cavorts by international development set.

  6. this list seems a bit flawed. #3 is, at least in part, determined by your success at #2. If no one in the aid community reads about your results, then you probably didn’t publicize them enough/ in a convincing way/ to the right people.

  7. Note that the benefits of a research project include not only the knowledge generated, but also the benefit to the respondents and research staff. A savings account experiment provides knowledge, savings accounts for the poor, and jobs for local field officers. The question stands, but the bar seems a little easier to get over in that case.

  8. It seems like #4 is missing: My research is relevant, meaning it answers an important/relevant (not just an academically-interesting) question. It seems to me that there’s an awful lot of research out there that fails the #4 criterion.

  9. Ha, I appreciate the good answer!

    For what it’s worth, I think you are adding a good amount of value. And this blog certainly helps with 2. and 3.


  10. I would suggest that research funding should be regarded like venture capital – the vast majority of it produces absolutely no useful result for broader society or even the narrower goal of creating knowledge , with the rare successes producing spectacular gains. The test then should not be so much the individual research projects but the “portfolio” which I guess would be a particular tranche of research money or long run activities of particular funding agencies.

  11. Yep, key question. A softer option is going on holiday to developing countries and spending in locally-owned outfits. For me, that skews the payments to mountain biking and windsurfing companies, but it’s a start and lots of fun! Next step is to suggest to my boss that giving me longer holidays would increase my impact.

Comments are closed.

Why We Fight - Book Cover
Subscribe to Blog