Looked at over the fifty-year span since the publication of W.W. Rostow’s The Stages of Economic Growth: A Non-Communist Manifesto, development economics as a field looks far more like literary criticism than like those natural sciences it emulates.
As we contemplate the series of enthusiasms that have characterized different moments in development theory and policy, it looks an awful lot like movements from romantic to symbolist to modernist to beatnik poetry. We may not yet have seen beatnik development economics, but we can still hope!
The difference between poets and economists, however, is that for poets, as for literary critics, there are rivalries and certainly individual claims to preeminence, but as a general rule, there is an acceptance that there are many ways to write a great poem, just as there are many enlightening ways to read any great poem. Bound as it is to the model of the natural sciences, economics cannot accept that there might be two incommensurable but equallyvaluable ways of explaining a given group of data points.
That is Mike McGovern, one of my favorite antropologists, reflecting on a year spent among the development economists in Oxford. More specifically, he is reviewing Paul Collier’s books in the latest Perspectives on Politics.
An excellent read, with many pointed lessons for the field.
At the same time, I think he takes a few works as representative of the whole. I think he’d get a very different feeling from reading Banerjee and Duflo’s Growth through the Lens of Development Economics or their new book, Poor Economics. Or, getting macro, Dani Rodrik’s latest book on globalization.
One area where I agree: economists often take their models too seriously, and too far. Unfortunately, no one else takes them seriously enough.
In social science, models are like maps; they are useful precisely because they don’t explain the world exactly as it is, in all its gory detail. Economic theory and statistical evidence doesn’t try to fit every case, but rather find systematic tendencies. We go wrong to ignore these regularities, but we also go wrong to ignore the other forces at work–especially the ones not so easily modeled with the mathematical tools at hand. I think that might be a better starting point for a critique of the discipline.
But that is a blog post for another day. Perhaps a post-tenure day.
Meanwhile, see Mike’s other work here, or his excellent new book, Making War in Cote d’Ivoire. My copy just shipped, and is awaiting me at home.
13 Responses
I see a few themes in the piece, all but one of which I agree with:
1. The difficulty of establishing causality, with Collier often ignoring these difficulties and sometimes even making the classic mistake of confusing correlation with causation – Agreed. But Collier’s books aren’t very representative of the field here; most economists’ work, even popular development economics books, are much better about this. But before we give ourselves too much of a pat on our backs, on to theme 2:
2. Economists kidding themselves when they see themselves as just “sticking to the facts” and avoiding subjectivity. Their unconsciousness of the subjectivity of their inferences often makes their results *more*, not less, off-base than if they were to engage directly and consciously in more clearly subjective methodologies, like those of anthropologists. – Agreed. Props to McGovern on calling Collier out on his sleights of hand here.
3. A call for humility and a greater hesitance to make policy prescriptions – Agreed. Collier’s prescription for a contract between rich countries and troubled countries to keep coups d’état in check sounds like the musings of a twelve-year-old who just got done playing a game of Risk. Of course, that’s also the charm of it, and perhaps it’s worth trying. But how in God’s name could anyone, much less an academic, ever offer a policy prescription like that without the slightest bit of sheepishness?
4. An ode to ethnography — Disagree with him here. Ethnographic work is wonderful and important to development economics, but not for the reasons McGovern argues. Its relevance to development is that it’s micro-focused, detail-oriented, and it forces researchers to get off their asses, look around, and talk to the people they’re studying. It’s not wonderful and important to development because them thar Africans are really different from us whities. Well, they are, but not in any way that has anything to do with development economics. Ethnographic research is just as useful to public policy construction in the U.S. as it is in Côte d’Ivoire. The paragraph on “the political economy of a cocoa plantation frontier zone, the history and demographics of long-term internal colonization of the region, and the political entrepreneurship of specific individuals who capitalized on these structural possibilities in order to benefit from a situation of neither peace nor war” is spot-on. But the paragraph after that, which concludes, “Such ‘irrational choice’ models must account for the presence and significance of actors’ desires for respect, honor, adulation, and revenge” — no. All those emotions are important when writing a biography (or, well, ethnography), not so much when doing economics. Honor and revenge play big roles for a lot of people in a lot of wars, but they’re never the limiting reagent, so to speak. Ivorians aren’t any more honorable or vengeful than Guineans; no need to include those in a model explaining the different outcomes between the two.
In fairness Chris- this is primarily a review of Paul Collier’s books.
And no review of Bottom Billion should avoid mentioning some of pretty cringeworthy ‘inside the mind of the rebel’ stuff that Collier wrote.
And I think Mike McGovern is right that economists would do well to be more humble.
If models are like maps- then to extend your analogy I would say that when I’m trying to find a place, certainly I look at a map. But I also look where I walk, and if the map is leading me off a cliff (*cough* IMF structural adjustment policies *cough*)- then I might just ignore the map.
This is not quite enough:
Economic theory and statistical evidence doesn’t try to fit every case, but rather find systematic tendencies. We go wrong to ignore these regularities, but we also go wrong to ignore the other forces at work–
Here’s the thing. Economics studies Human Action (like von Mises did). When econ statistics allows many people to change their behavior to make more money (i.e. buy internet stocks, ’95-2000; or US houses, ’95-2005), the people DO change their behavior.
But this change in behavior invalidates whatever the “economic law” that was being followed.
This critique of economics is always valid for the relevant portions, those that allow decision makers to profit/ benefit in some way. Where nobody can use the economic law to make money, and thus there is little incentive to change behavior, the behavior won’t change.
David is wrong here: Economists have not shown themselves to be reliable in getting an economy to grow, or in eliminating poverty.
In America, poverty is eliminated in people who follow 3 steps:
a) graduate from High School (knowing HS stuff, reading etc., not just paper),
b) keep the same job for over a year,
c) avoid having children out of wedlock.
There’s virtually no poor people in America who follow those three rules. The problem is that politicians and voters want economists to find some way of eliminating poverty without big changing of the behavior of the poor people. In that, economists will fail.
Also, science fails when people want a perpetual motion machine.
But most folk know that perpetual motion machines won’t work, and most politicians won’t even promise such hooey.
Yet most International Aid donors continue to fund, and many politicians continue to ask for, ways of reducing poverty that avoid changing the way the poor people live and act.
It is pretty tautological to say, “People who reliably find work and have few childcare obligations are often not poor.”
I mean . . . yeah.
McGovern raises the issue of whether homogeneity really leads to greater governability. It would seem intuitive to me, but a little while back I read Chris Coyne citing some papers saying there’s little correlation. I provide some quotes and cites from his book here.
“They actually spent millions a year on this. ”
They should spend tens of billions. That is the problem. It’s actually worse that they’re aware enough to know it’s an issue, but don’t just wave the money wand at it.
What I find odd is that arguments about economics seem to be focus on whether economic methods are valid or not. “Rigorous” is a popular word in such discourse.
But in the natural sciences, the status of which economics appear to aspire to, validity, and more important social authority, is based not on method, or rigour, but on results.
Economists have not shown themselves to be reliable in getting an economy to grow, or in eliminating poverty. Physicists, on the other hand, have proven themselves extremely reliable in providing the means for packing transistors onto a chip, plotting a trajectory to the moon, or landing an artillery shell in an enemy’s lap.
That’s why the knowledge physicists produce is taken seriously.
Donors should apply to economics the same professional criterion they apply to doctors and engineers, tax administrators and computer programmers. Before being allowed to work with the most vulnerable population, every discipline must first demonstrate that it is able to deliver results.
Off the cuff thought: Agree with both ends of the argument, really. Economics should face the fact it’s a social science, not a natural one and not a purely mathematical one. When it doesn’t do that, then I tend to fall on the McGovern side of the argument. Having said that, Chris, I like your proposed starting point. It acknowledges how economics as a social science, one that employs statistics and math, is so useful.
“Surely neither group would recognize themselves in my description, but perhaps it says something about a kind of mutual attraction that dares not speak its own name.”
“If we as economists were serious about this, the first thing we’d have the IMF do is fund statistical agencies in every country.”
Actually they do – as do most multilateral development banks (World Bank, African Development Bank, etc.). They actually spent millions a year on this. Read some of their annual reports. IMF, in fact, has training programs where they bring officials from statistical bureaus in Africa for courses in DC.
I’d love to see an anthropologist’s take on Poor Economics. Ed Carr had a good debate emerging from a geographer’s viewpoint on Poor Economics and RCTs (http://bitly.com/j6vltL) and this review (http://bit.ly/lNHqOy) actually said “randomised trials have turned economists into first-rate anthropologists”. I’m not sure this is true; some of the randomistas’ work seems to have been coupled with good qualitative research (I’m aware of an ongoing evaluation in Mali where this seems the case) but other parts of it seem to fall into Collier’s ‘correlation + supposition based on a couple of conversations’ trap. We should also remember that part of the problem is the simplification inherent in books that are trying to summarise years of work into an easy read for a general audience.
My copy of Mike’s book on Cote d’Ivoire has also shipped. Have you read it? If so, it’d be interesting to read your thoughts. If you haven’t read it, I’d be interested to know how you come to describe it as ‘excellent’. I hope indeed it is.
I guess. If we as economists were serious about this, the first thing we’d have the IMF do is fund statistical agencies in every country. It feels like the combination of donor politics and unvoiced assumptions leaves development as largely nonscientific. It’s not that we take our models too seriously or not seriously enough — it’s that we don’t seem to be very excited about the data. Or making sure that the data are representative.