It is currently fashionable to say that poor countries are caught in a poverty trap; and that a big push involving aid and investment would lift them onto a virtuous circle of faster growth and rising prosperity.
So said Jeff Sachs in his report to the UN Secretary General on meeting the Millennium Development Goals, and so said Tony Blair’s Africa Commission in the build up to the G8 summit at Gleneagles.
It is an attractive argument: it suggests that if we can organise sufficient resource for a large increase in aid, we can shift the poorest countries out of poverty once and for all. This was the original argument for the introduction of foreign aid in the 1950s, and some version of this story has come in and out of fashion periodically in the last fifty years.
There is just one problem: it does not seem to be true. As my colleague William Easterly has shown in a new working paper, we do not see evidence in the growth data for countries being caught in a poverty trap from which some escape:
- the countries that start off poor do not appear to have significantly lower growth rates subsequently, and growth rates in poor countries are not close to zero
- there are few examples of a country achieving "take off" in which there is a marked acceleration of growth (Japan being the notable exception, and perhaps some of the other Asian tigers).
It does not follow from these results that aid is not effective in supporting growth and reducing poverty. There is good evidence that aid does lead to growth and poverty reduction (see my collection of articles here). But it does weaken the argument for "one big heave" of the kind envisaged by Gordon Brown’s proposed International Finance Facility, and supported by the Commission for Africa.
Jeff Sachs has done a tremendous job in drawing the world’s attention to the need to do more to achieve the Millennium Development Goals. He is right that more aid would make a positive contribution to reducing poverty. But it is unrealistic to expect that a even a large injection of aid is going to lead to a step change in the rate of economic development of poor countries. As another of my colleagues, Michael Clemens, showed last year, it takes a hundred years to raise school enrolment rates, not just fifteen years as envisaged by the Millennium Development Goals.
This is a call for realisim, not pessimism. It is a call for more action: the sooner we get on with it, the better. But we should be realistic about the speed with which we can expect to see change, and we should not set ourselves targets and expectations that we have no hope of meeting.
6 responses to “One last heave against poverty?”
Fascinating paper. Might find that turning up elsewhere soon.
What I’m most fascinated by is why Easterly completely ignored the evidence produced by Michael Clemens (who is cited as a commenter on the paper, and who also works for the CGD) on the positive effect of aid on growth? That research directly contradicts his assertion that “more sophisticated testing that would control for selection effects and reverse causality fails to find a robust effect of aid on growth”. He doesn’t even mention the Clemens paper – instead, he cites Rajan and Subramanian’s 2005 IMF paper, which you (Owen) have fairly comprehensively critiqued.
I’ve not finished reading Easterly’s paper yet (I am on holidays, after all), but he seems to be taking a deliberately crude approach to the debate. He assumes that the poverty trap theory assumes that poor countries must experience zero per cent growth. Does Sachs really say this, though? And his tests seem to feature comparing a lot of averages of averages, again in stark contrast to the more sophisticated approach adopted by Clemens et al.
Owen replies: Jim – he doesn’t ignore Michael’s paper; but he doesn’t agree. I am sure, however, that Michael agrees with Bill E. that the “one big heave” theory is too optimistic and risks setting us up for failure. It will be a long, slow haul. I think Michael would be with me in taking the view that aid is, on average, effective at raising growth rates, and that we should not base the argument for it on “one last heave”.
I also noted that he didn’t deal with the paper Jim mentions.
I’m a little surprised that Owen doesn’t point to what I think is the crux of the paper:
“The paper instead finds support for democratic institutions and economic freedom
as determinants of growth that explain the occasions under which poor countries grow more
slowly than rich countries.”
He actually uses the Freedom House and Fraser Institute (are they the same as the Heritage ones or different?) measurements to come to this conclusion.
Which is very much what I’ve been saying this past year…yes, yes, I know I stumbled upon it, no evidence of course…and certainly Jim has been shouting at me for doing so.
Tim
I don’t think anyone seriously disagrees that support for democractic institutions and economic freedom are very important determinants of economic growth. (At any rate, I don’t; and nor do my CGD colleagues.)
What we might quibble about is:
a. is aid much more effective if there is good governance? (My view: evidence suggests that aid is a bit more effective, but only modestly so, and certainly not enough to base an entire aid allocation strategy)
b. can governance be improved by withholding aid from countries that do not perform? (My view: no. There is no evidence that aid conditionality is effective in improving governance).
c. if a country does not have good governance, but aid could nonethless be effective in raising growth in that country, should still we give aid? (my answer: yes – there is no good reason to withhold it).
Owen
“he doesn’t ignore Michael’s paper; but he doesn’t agree”
I have to disagree with this. If he mentioned the paper (or any of the other research that finds a positive effect on growth) and explained why he doesn’t agree with it, that would be something. But he doesn’t even mention it. In fact, he gives the impression of looking in vain for research that robustly supports a significant positive effect of aid on growth: the Clemens paper is exactly that, but he can’t bring himself to even mention it.
“I am sure, however, that Michael agrees with Bill E. that the “one big heave” theory is too optimistic and risks setting us up for failure.”
That’s my worry too, and it’s a perfectly valid point. But I think he misrepresents the state of the evidence on the impact of aid, and I’m certain this paper will be lept upon by people who oppose the idea of foreign aid per se. I’m sure he’s well aware of that possibility (it’s happened in the past after all), which makes the glaring ommissions in his paper all the more disturbing.
Owen replies: You are quite right. I had misremembered – I thought Bill Easterley had mentioned Michael’s paper but he doesn’t.
UNDP Human Development Report.
As reported in the Guardian today the UNDP has released this year’s Human Development Report. Things are not getting better everywhere and more should be done. That basic analysis seems unobjectionable. There is one slight problem, however, if this Bill