MethodLogical is now at

MethodLogical is now at

Due to some persistent technical issues we've been having with Blogger, we're now posting at Please update your RSS feeds, etc. For the time being, our new posts will automatically be mirrored here, but you'll have to visit the new site to comment.

Monday, February 7, 2011

Answers to Questions No One Asked Me

Owen Abroad and Chris Blattman both have recent posts responding to overlapping sets of questions about development posed by a journalist writing a series of articles on development. The questions are really broad - I'd say they're really The Questions in development research - but Chris and Owen provide concrete and well-reasoned answers and both their posts are worth reading. Owen also got a bunch of other great answers in his comment section.

Despite my lofty status as a second-year Ph.D. student I was not contacted for my take on these questions. But I've never let not being asked keep me from volunteering an opinion before, and I'm not going to start now. I'm hoping ML's other contributors will chime in with their own responses or at least criticize mine.

1. It is realistic to think that poverty can one day end?

Yes, without a doubt.We've made considerable progress toward that goal already - look at the massive improvements that have happened recently in China, across Eastern Asia, and increasingly in India. I see nothing to indicate that we can't reach a point where everybody has at least the standard of living of the global middle class. It's natural to fear that poverty is a zero-sum game, but the rise of living standards in places like South Korea has helped other people around the world, not hurt them.

Both Chris and Owen make concessions toward a relative definition of poverty, rather than an absolute one. They're correct that relative poverty is always going to exist: random variation, be it genetic, environmental, or otherwise, means some people will be better off than others no matter how high incomes become. But I'm not willing to admit that relative inequality constitutes something we should call poverty. Absolute income deprivation is causally linked to all kinds of bad stuff, from the incidence of civil wars to malnutrition and low life expectancy. As I've already discussed with Andrew on this blog, I'm not convinced that inequality makes any difference at all.

I think a numerical example might be helpful here. Imagine that in Society A there are only two people: myself (living on a $20,000/year annual stipend) and an average Malawian (living off $800/year at purchasing power parity, or 1/25th what I do). Now Imagine Society B contains just myself and Bill Gates, who earns roughly $200 million/year, which is 10,000 times my income. By any reasonable measure of inequality Society B is more unequal than Society A - but which is more concerning? Bill Gates' living standard is much closer to mine than mine is to a typical Malawian's. I worry a lot more about helping the billions of people living on just a few dollars a day than I do about the fact that Bill Gates could, in principle, buy millions of BMWs to drive around when I can't even afford one. I think it's counterproductive to talk about poverty as a relative concept: it means the goalposts are always shifting and can confound our attempts to

2. What are the best global solutions?

Blattman points out that industry is really the difference between rich and poor nations and doubts that de-worming can really be transformative, but openly admits that we're not sure how exactly industry develops in a poor country. I would argue that health interventions can be transformative - if you take Bleakley's hookworm paper seriously, the implication is that Rockefeller's campaign to eradicate that parasite was directly responsible for big income gains in the American South. Eppig's work on disease burden and IQ points toward the same possibility: fix disease and you'll boost economic development. My undergraduate advisor, Roger Noll, oversimplified it beautifully, saying "we know the solution to economic development. It's 19th-century public health." That's not the only thing but I do think truly solid public health - clean water and malaria eradication in particular - would go a long way toward spurring industrialization where it is lagging.  Deworm the World is an excellent start.

Exposure to disease leads to lower IQ - and worse economic outcomes

The other big solution I'd push was one mentioned by Barder: migration. Immigration is the single most powerful development tool we have. A Haitian can octuple her income by moving to the US. People can increase their incomes drastically by moving to a developed country and when they do they tend to send remittances back to help their families. Opposition to immigration runs deep; almost by its nature it is driven by a nationalism that values foreigners less than one's countrymen. But policymakers have a lot of wiggle room to let in more immigrants, and they should use all of it.

3. How urgent is it to act?

Very urgent. No matter what your policy goal, tackling poverty is the best solution in the long run and we need to start now. To take my favorite example, rising incomes lead to falling family sizes, and this effect is already forecast to put a stop to global overpopulation. Or another: carbon intensity is a falling function of per-person income. We're not going to stop global warming without increasing incomes. These effects will only kick in over the very long run so we need to start now.

At the same time, urgency frequently gets in the way of clear reasoning when it comes to doing good in the world. It's the role of development economists to say "not so fast" and point out what doesn't work and what we don't know, and hopefully to impart healthy skepticism to other people involved in development.We need to move quickly, but first to gain knowledge and only second to act.

 4. Do you believe there is hope for the future?

Of course. Otherwise I wouldn't be trying. But the more I learn, the more I see that no problem in development has a simple solution


  1. Could you explain why you think inequality doesn't matter? I can't help but feel you're relying on a very narrow definition of what it means for something to matter.

  2. I'm not sure I can do much better than I did already with the Bill Gates example, honestly. The gaping inequality of Society B seems of trivial importance compared with the absolute deprivation prevalent in Society A.

    Here's another look at it: you can choose to place value on whatever you like, so if you dislike inequality I can't argue with you. but when I'm making my own social welfare evaluations I try to base them on what matters to the individuals in a society.

    That can happen in two ways. First, inequality can per se be something that many people dislike. This requires an awareness of inequality, which is plausible for Brazil (where David Lam says everybody knows what the Gini is and how it's changing) but less so for the US (where my classmates in econ grad school don't know the Gini, and where massive income inequality is presented as a shocking fact). In general I don't buy that people are aware of aggregate social inequality so I have trouble believing that it bothers them.

    Alternatively inequality can affect other things that matter, like health. The studies showing this kind of thing are frankly plagued with errors. Tellingly, in the four weeks on inequality in my development class we haven't looked at a single paper claiming it causally affects any economic, social or health outcome. This year I've read dozens of papers on the links between income levels and health outcomes, education, lifespan, civil wars, HIV infection, IQ, you name it.

    Even if I think inequality is plausibly a bad thing in and of itself I'd also need to be convinced that the effect is large. If we focus on inequality that means less focus on other things we know to be important. For example, Lam proved that an unambiguous improvement in education (a rise in the average and a drop in inequality) can lead to more inequality in wages, albeit with a higher mean for everyone. This was likely going on in Brazil in the 1980s. Looking at that tradeoff, I'll take the improvement in schooling and in average income and accept the cost of more income inequality.


Note: Only a member of this blog may post a comment.