4 thoughts on “Does foreign aid work?”

  1. Posted 08/10/2014 at 10:55 | Permalink

    A rather partial review of the literature. For example, that meta analysis you cite has been critiqued here:


    and I could cite dozens of papers that find a small positive impact upon growth. For example one study finds “on average a sustained inflow of $25 aid per capita is expected to increase the rate of economic growth by around half a percentage point, reduce poverty by around 6.5 percentage points, raise investment by around 1.5 percentage points in GDP, augment average schooling by 0.4 years, boost life expectancy by 1.3 years and reduce infant mortality by 7 in every 1000 births.” (http://www.wider.unu.edu/publications/working-papers/2011/en_GB/wp2011-044/)

    And we should expect to find only a small positive impact upon growth, because standard economics says welfare maximization implies most aid ought to be used to fund consumption, and the impact of invested aid on growth rates will be small. Most researchers conclude the balance of evidence is for a small positive effect on growth, not zero.

    There is also lots of research – cross country regressions and impact evaluations of the RCT type – showing positive effects on health outcomes. The impact on poverty itself is under researched, but if you wanted to you have could have found papers that find a positive impact on poverty, instead you choose just one paper that finds a negative impact on inequality in a small sample middle income countries. That is consistent with the idea that aid largely funds infrastructure etc. in middle income countries and stimulates growth and growth disproportionately benefits the better off. Not much aid in MICs is targeted at poverty alleviation.

  2. Posted 08/10/2014 at 20:07 | Permalink

    I am fully aware that Mekasha and Tarp have criticized the meta analysis cited above. However, for the sake of impartiality, you should have noted that Doucouliagos and Paldam have replied: https://ideas.repec.org/p/dkn/econwp/eco_2012_4.html They show that Mekasha and Tarp make a number of mistakes, and in particular that they ignore the publication bias in the literature. Not doing so is what drives Tarp’s claims. Tarp’s critique of Nowak-Lehman et al. (2012) was even more embarrasing.

  3. Posted 09/10/2014 at 10:27 | Permalink

    yes you’re right Doucouliagos and Paldman did come back, although I do not share your assessment of who came out on top in that argument. And as for saying the article that pointed out Nowak Lehman took logs of zeros and negatives is “embarrassing”, well …

    and “for the sake of impartiality” is rich coming from you!

  4. Posted 09/10/2014 at 12:54 | Permalink

    I’m not going to judge, but have a look at the reply from Nowak-Lehman et al. They make it quite clear that Tarp et al. reported tests that went their way and did not report equally important tests that did not.
    About impartiality: I don’t think any of us would believe research into the effects of drugs if the researchers were paid by the pharma industry. Then why do so many people put so much faith in research done by people paid by the donor community?

Comments are closed.