Chris Blattman

Close this search box.

Two views on fighting world poverty




Not everyone like my idea for fighting world poverty.

Bill Gates wants to solve African poverty by helping more poor people raise chickens. So recently I wrote Gates an open letter in Vox asking him two things. First, consider cash, because chickens are probably not that great a gift to the poor. Second, we really don’t know who is right. But an awful lot of people would be better off if we invested in a little research to find out. I think $15 million would be enough to run a horse race between different ways of giving poor people stuff to help them raise their earnings: cash, chickens, training, and combinations thereof.

Yesterday Lant Pritchett expressed his bewilderment on the CGD blog. You should read the full post, but let me summarize in a few sentences.

First, Lant thinks that focusing on the very poorest is arbitrary. Official development goals and lots of programs target the people living under $1 or $2 a day, but people living under $10 a day are also pretty badly off.

Second, and more important, if we really wanted to alleviate poverty we’d focus on economic growth, and changing the systemic problems that keep countries poor. Anti-poverty programs like mine are band-aids.

I think Lant’s right and he’s wrong. We have to focus on the big picture and growth as a society, but I think there’s a strong argument for directly tackling the worst poverty now. Especially because we know how to do that pretty well. And we could do it even better pretty easily. More so than figuring out the secret to growth. I have a hard time imagining the $15 million research project that would affect country growth rates one bit.

Let me explain.

First, I agree with Lant that we can’t lose sight of the important goal: doing whatever possible to get economic growth rates up and steady in the poorest countries. But then, I’d argue that development economics is focusing on the wrong things.

The problem in the poorest countries today is first and foremost politics.  More people should be trying to understand what is up with Sudan or Nigeria or Afghanistan, and how societies like these can constrain their leaders and states; how states can build better bureaucracies, collect more revenues, and shape society; and how the risk of coups and conflicts and crises can be averted. And how US and European policy is partly to blame.

Currently that’s a niche area (to say the least). Lant has written about this topic more than most in articles I love and teach. And indeed I think the question of getting politics right is so important I teach a whole course on it.

I’m glad rich countries spend money on this. The UK’s development agency, DFID, is one of the few donors supporting this work on a large scale. I wish they would do more of it.

But that gets me to my second point: I would have a hard time convincing myself, let alone Bill Gates, that this is the first and best way to spend money.

This is partly because I’m not even sure money is the missing ingredient. I don’t know what the secret sauce is to be honest. I’ve seen DFID pour millions upon millions into research on the questions Lant likes. I’ve enjoyed reading many of the papers and reports. Others weren’t worth the paper they were printed on. But did any of them, even the best, ever have a direct impact on policy? I’m not sure anyone knows.

In spite of this, I’m glad DFID and others invest in it. But the best use of Bill Gates’ money? I doubt it.

The reason I like the research I propose is simple: I can see exactly how it will be used and how it will change life for a large number of people in a short period of time.

International organizations and governments give poor people a lot of a lot of expensive stuff to help them be less poor. Billions a year. Most of that money is wasted, because they give the wrong stuff. I think that could change.

In fact I’ve seen it change. In the last five years the cash and other program evaluation evidence has had a big impact on these “give stuff” programs, pushing them to be more efficient and better at helping some of the poorest. I think a couple of monumental studies would be an even bigger push for more governments and organizations to change. That is, I am betting that a $15 million study would change how hundreds of billions are spent. If I’m right, the marginal return to this investment is pretty good. I don’t know many better examples.

Does this kind of policy focus too much on people who are below an arbitrary threshold like a dollar a day? Maybe. If it does, I don’t have a big problem with that, because under a dollar a day is a really miserable place to be. It’s probably the place where an extra dollar of income per day will have the biggest impact on things like child heath and mortality.

Also, my impression is that getting someone from $1 to $2 a day is easier then getting them from $10 to 11.

The little utilitarian in me feels good about this.

It’s a moot point. The fact is, these kinds of “give stuff” programs go to the $1 a day and $10 a day earners alike. In the Syria region, the research has pushed organizations to give ATM cards and cash to millions of refugees. In sub-Saharan Africa it’s pushed governments to give cash out to huge swathes of the population to stimulate local development, including the “middle class” who seem to do quite well with some extra cash. And my hope is that in Latin America maybe the evidence would push governments away at least a little from ubiquitous middle class training programs that don’t work very well.

But just because I think the best use of Gates’ $15 million is my project, doesn’t mean I think all or most of development aid should go to poverty alleviation. Not surprisingly, the answer is we need to focus more on growth and direct poverty alleviation. I think Lant is focused on the totals we spend: the world spends too much on the worst forms of poverty now and not enough on growth and ending poverty in the future. That might be right. But it’s an argument that has to be made partly on faith, because it is very, very difficult to connect the salary of a growth economist to somebody’s life being better off in 40 years. I think growth economics needed a little competition and has gotten it, and will hopefully get better and rise again.

17 Responses

  1. The example of microcredit comes to mind. While microcredit is not exactly giving cash out to people, rather giving people loans, there are some important lessons that can be taken from it. Organizations like Grameen Bank in Bangladesh have indeed helped some people out of poverty (some of whom were under $1 in income), but there seems to be apparent misuse of the money lent as well. There are examples of families simply taking the microcredit money from one organization, living on it, and burning through without ever moving towards economic growth.
    It would certainly be interesting to see if those unwanted dynamics occur in the scenario of cash handouts.

  2. The world would undoubtedly benefit from a sharper grasp of the political reforms that lead to widespread growth and poverty alleviation, and it’s great that DFID invests in it. But even with a fantastic empirical literature to guide them, most of us (especially Mr. Pritchett) would stop short of saying that the Gates Foundation ought to start trying to enact political reform in foreign countries. We like the framework of household-level capital transfers in part because it’s a form of altruism that doesn’t involve men like Bill Gates taking on powerful roles in nations and communities to which he doesn’t belong.

  3. @chris Is there any difference between Bill Gates’ money and money printed domestically? Put another way, isn’t the cash problem a failure of the models, tools and instruments of the central bank to transmit monetary policy where it is most needed? As for the chickens, the plan is an ill advised intrusion into the normal operation of the market

  4. @Michael Strong is so right on the need to deregulate — far more important than corrupt local politicians collecting more revenue.

    Your idea of comparing “cash” gifts, to “chicken” gifts is great. I actually suspect that when a poor community gets the one-time chicken gifts, the economic growth in that community will increase slightly faster, due to more successful chicken entrepreneurs learning to be successful in that specific community.

    I also suspect that $15 million spent on helping the “most successful” of the poor business owners in a poor community would lead to higher economic growth.

    I call this “trickle up” econ growth, and is based on the idea that in any group of poor folk, there are some harder working, some smarter, and some luckier than others. When these less poor folk get assistance, the harder & smarter workers provide good examples for others to emulate.

    When (if?) the process allows the best chicken farmer owners to hire the less smart, less lucky, less hard working, both the owners and the hired workers are better off.

    With multiple trials of such development aid, it might even be found out how many new successful small business owners are needed to support a community tipping point.

    My own plan for $15million in a village would be to make the money available for loans to new businesses, as signed up by new business partners. Imagine 15,000 adults each being given a $100 investment book of ten $10 vouchers, and various folks presenting their business plans to their friends and families and getting support for starting a new business, whether raising chickens or goats, or building housing, or furniture; or getting a tractor to timeshare among farmers. Based on the individuals being invested in.

    Yes, this sends the aid to Investment rather than Consumption, among poor folk who are often desperate for more consumption; but growth depends on investment, and positive returns on investment.

  5. +1 for the politics first direction.

    As much as poverty reduction is a noble pursuit when fear of explosions remains a rational self preservation mechanic poverty is going to sit on the back burner.

    Almost 15 years ago I printed a map of the world and started sticking pins in to flag conflicts and how intense they were. What struck me was that I kept adding pins and never taking them away. Off hand I can’t think of any conflict that can really be said to have been resolved since that time, the closest being the FARC *skirmish* in Columbia.

    It seems almost Orwellian that since, well for sake of homage, 1984 the conflicts of the world have been basically static and without victors (ok, probably 1990 is better but forgive me the chance to make things look nice). So why is this?

    Conflict resolution really strikes me as a priority for both research and action but, as John Kerry can probably firmly attest, really doesn’t get the sexy coverage it should.

  6. I like the fact that Blattman at least acknowledges that growth would be better if we knew how to do it (that is in contrast to his belief that we know how to tackle the worst poverty “pretty well”) . That said, a few complaints: 1) . As with most of the rest of development economics, he seems to ignore special economic zones as a development strategy. A decade ago Bill Easterly acknowledged that the field of development economics had ignored zones – and that it had been a mistake to do so. Now we have Lotta Moberg and Mark Lutter with recent GMU dissertations on zones, but the field as a whole has been remarkably neglectful of the most potent development strategy of the past fifty years. WTF? 2). Blattman is legitimately concerned with the challenges of violence and weak institutions in many developing nations. That said, I see no emphasis on the simple fact of excessive regulation in developing nations, including peaceful, stable poor nations. Perhaps he feels that that issue is banal, with the Doing Business rankings and all, but somehow Bill Gates, Peter Singer, and almost everyone outside of development econ. seems to be unaware of this issue. Meanwhile, Naomi Klein and the entire academic industry vituperating about neoliberalism claim that capitalism makes the rich richer and the poor poorer. Many people still believe that “neoliberalism” resulted in an “unregulated” third world in which corporations exploit the poor. While China, where average urban wages are up 6x in the past twenty years, had a relatively unregulated labor market, most developing nations heavily regulate both labor and business. Most of these regulations are inconsistently enforced, but if they were it would wreak havoc in these countries. The resulting uncertainty and arbitrary enforcement is the worst of both worlds: a terrible environment for business AND no protections for the weak. Have we truly optimized communication on these issues? 3). Moreover, this is an issue on which broader awareness could have an impact. There is a literature on ways in which the Doing Business index has resulted in reforms. My wife, an African entrepreneur and speaker, regularly speaks to audiences in Africa, Europe, and the U.S. on this issue and in every conference, she is the ONLY one bringing up the issue. Often African ministers and NGO leaders are completely unaware of these issues. If the information is esoteric, how can it have an influence? 4). Bob Haywood, the former ED of the World Economic Processing Zones Association, and who was personally involved in the development of dozens of successful zones around the world, made the case that zones were a solution to North’s “natural state” problem. In short, most zones are supported by local elites – but not the leading oligarchs, but rather the second or third tier elites, or young brothers or cousins of first tier elites. In essence, connected individuals who are not at the center of the rent-seeking system. The zones are typically designed so that they provide improved business opportunities for these advocates within the zones while not threatening the entire rent-seeking structure of the society. As the zones grow, they then create a constituency and a demonstration project for economic liberalism. Haywood makes the case that zones have led to broader liberalization in Taiwan, South Korea, Mexico, China, Mauritius, and Ireland. 5). In many ways, these criticisms are true of much of the profession and Blattman is simply responding to standard professional norms. But Pritchett’s frank comments are all the more valuable precisely because the professional norms seem to limit the boundaries of which issues are highlighted in the profession. Within the profession, renegades such as P.T. Bauer and Bill Easterly have probably done more for the global poor than did the mainstream of the profession. To take a different kind of renegade, Gary Becker expressed the belief that Milton Friedman deserved more credit for bringing hundreds of millions out of poverty than any other individual. Outside of academia, the SEZ industry has done much more for the global poor than did all of development economics, which for decades was hostile to zones when it wasn’t ignoring them.

  7. Fascinating discussion. I think almost nobody argues against the notion of power relations as being central to the problem, and politics as a critical element to the solution. At the same time, however, many support the moral imperative of simultaneously implementing solid community-level interventions to help those who are vulnerable.

    In relation to your $15 million research project: Why don’t you connect it to the announced GiveDirectly experiment of giving cash (a grant of approximately $30 per month) to 6,000 people in some communities of Kenya for an extended period and having J-PAL testing the impacts (grant vs non-grant villages)?

    Berk Ozler (WB) had suggested that this type of experiments should extend its focus (in terms of outcomes being measured) and include a comparison against other places where other cost-equivalent programs are being implemented such as BRAC’s Graduation Model or those long-term models implemented by well-known International NGOs. What about the Sach’s MVP? I think at some point he even invited the Gates Foundation to independently supervise a full evaluation.

    I am saying that, of course, you need adequate resources for your proposed research but, perhaps equally or even more important, is the need to gather support and openness of other organizations implementing poverty reduction programs. By having each organization run (with their own funds) its own program in its assigned areas, there are incentives for everyone to do their best to maximize the effects on the wellbeing of people (health, education, economic development, the capacity of communities to organize and act, etc.).

    There is an issue of trust and political will in all this.

  8. The world would undoubtedly benefit from a sharper grasp of the political reforms that lead to widespread growth and poverty alleviation, and it’s great that DFID invests in it. But even with a fantastic empirical literature to guide them, most of us (especially Mr. Pritchett) would stop short of saying that the Gates Foundation ought to start trying to enact political reform in foreign countries. We like the framework of household-level capital transfers in part because it’s a form of altruism that doesn’t involve men like Bill Gates taking on powerful roles in nations and communities to which he doesn’t belong.

    How Bill Gates ought to spend his money is a question worth answering, even if it’s not the top priority of the global community or one Mr. Pritchett finds interesting.

    The distinction between the projects of altruism and growth seem salient to everyone here. It’s a constant topic on this blog And from his CGD essay: “… there are two distinct goals one could pursue: ‘fighting world poverty’ and ‘eliminating destitution.'”

    Both goals need researchers. The one is undoubtedly more important than the other. If we agree on that, the residual disagreements seem somewhat second order.

  9. Great post as usual. I wish, I wish, I wish there would be some RCTs being done on the poverty alleviation programmes being undertaken right now in China. From village relocation to cash transfers to agricultural modernisation to e-commerce – it’s all being done right now for the last set of 45 million people in rural poverty… that would provide some pretty decent data I expect…

  10. Fantastic post, as usual.
    Just as an aside, however, I think we might be underestimating the returns to such ‘chickenVcash’ research a bit in terms of affecting overall economic growth. A good example would be the research being done in the areas of human capital which is a key driver of long run econ growth. Yes, weak political institutions in a country might impede any reform in, say, the education sector, but what about developing countries whose institutions are just starting to improve? Wouldn’t they be eager to learn from the ongoing research that what kind of interventions would be best to improve learning outcomes? Wouldn’t such research be then able to help those countries speed up their ascent up the economic ladder?

  11. To your first point Chris; Lant and others at the research group I was a part of at the University of Manchester are focusing on this topic. Kunal Sen, Eric Werker and Lant have a book which is due to be published later in 2017 about how Political Settlements affect growth patterns within a country so it makes your interesting and in my opinion correct point moot. You two are looking at these issues along similar lines and don’t even know!