Abhijeet Banerjee and Esther Duflo are busy doing shows/interviews marketing their new book- Poor Economics.
David Leonhardt interviewed the duo recently. South Sudan is going to be the newest country and Leonhardt asks the duo what would they advise the new government to improve the lives of their citizens:
Ms. Duflo: In just a few minutes, we could not cover very detailed ground. So we’ll have to focus on the basics. First of all, I would try to convince them that a key priority would be to invest enough money and talent in running good quality social services for the poor, including free access to good schools, preventive medical care, and hospitals. This may not seem like rocket science: but these are basic human priorities, and these are also domains where some things are known about what may work.
Second, I think I would try to convince them to run anti-poverty policy in a more intelligent way than what we see in most countries. In particular, I may try to encourage them not to listen too much to the elevator pitches of all the other experts, and stake their entire policy course on the basis of those… Of course they’ll have to start somewhere, and there is a body of knowledge available to choose policies that are likely to work. But they will still have a lot more to learn about the best ways to achieve their objectives. So I would like to advise them to always keep some margin to experiment, in order to find the best programs to reach those goals.
Mr. Banerjee: Since they will no doubt want more specific suggestions, here are two policies that I think every poor country should implement. A small universal cash grant to everyone over 12, based on biometric identification. This guarantees that no one has to face the humiliation of being totally indigent, and from our evidence, makes people more productive as well. Making it universal is important, so that they do not attempt to identify the poor (which is very difficult to do effectively in poor countries).
Second, a free universal health insurance policy that covers catastrophic health events, which allows people to go to private or public hospitals. Catastrophic health shocks do enormous damage to families both economically and otherwise, and are easy to insure, because nobody gets them on purpose. On the other hand, insurance policies that only treat certain catastrophic illnesses are hard to comprehend, especially of you are illiterate and unused to the legalistic nature of exclusions etc. Therefore people do not value them as much as they should which makes it hard for markets to supply them. This is an obvious thing for governments to take on.
Chris Blattman dissents and says these programs might work in 2021 not now:
I’m going to be provocative and say: these would be my last priorities for the new government. If Leonhardt had asked, “How best to relieve poverty today?” Banerjee and Duflo might be exactly right–there are no better poverty experts alive. But South Sudan is no Uganda or Bihar. It is an entirely different animal. I would follow their recommendations, but in 2021 rather than 2011.
Today, South Sudan is a state in name only. The long term welfare of its citizens means sustained stability and security and order. Without it, all the anti-poverty impacts, no matter how great, will evaporate in months.
In fact, huge and expensive anti-poverty programs could be counter-productive. Trying to build a 21st century welfare state (or even a 19th century welfare state) in a new and fragile nation, with virtually no legislative or bureaucratic capacity, may be a burden too great. Ignore for a moment whether vast aid flows would distort or corrupt, since those aid flows will happen anyways. I think an anti-poverty push would be a distraction, possibly an existential one.
States, like people, have attention problems, only more extreme. The new government may only accomplish one or two big things in their first five years. If, fifty years hence, we want the poor of South Sudan to prosper, paradoxically the last thing we need to do is push for the Millennium Development Goals today.
He says he knows little of politics of Sudan but still his suggestions would be:
1. Build compacts, possibly unequal and unsavory ones, with warlords and other big men, giving them a stake in continued peace, even if it means they control crucial ministries or development organs.
2. But for goodness sake try not to give up the ministries or development organs. There are non-pecuniary ways to buy people off. And spread it out so you get petty barons rather than oligarchs. They’ll be easier to deal with in 20 years when you have the strength.
3. Next, give every incentive for elites, especially the ones apt to war, to invest in fixed assets whose value depends on stability and growth. Make them entrepreneurs. Oil rigs don’t count. Property in Juba does. So do plantations and small factories, even if they need subsidies to operate at first. This is hard, and will require attention and dedication.
4. Aim for minimal corruption in twenty years, not two.
5. Create a minimally competent police force, one that is less criminal than the criminals. And a court system, with particular attention to the places where ethnic groups repeatedly clash over land or rights or respect. Target programs to these hotspots to buy some measure of content.
6. Train and educate the military like the bejeezus, and at all costs do not let it slip into factions.
7. Roads, roads, roads. Not only are they good for growth, they are good for exerting state control and building a sense of nation.
When these are done, Blattman says he would aim for economic growth.
You may argue: it’s not a zero sum game, we could do both. I say you’re partly right: we can do a little poverty reduction, but it’s a zero-point-one sum game, and there are some hard trade-offs to be made this decade.
You may argue: but relieving poverty reduces the incentives for people to revolt! I say balderdash. This was a plausible but naive theory of conflict that has turned out to be mostly wrong. Poverty is a third or fourth or fifth order factor in a decision to revolt.
You may argue many more things, and I am eager to hear them. Dissents from my dissent?
Well Banerjee and Duflo comment towards the end:
We feel that you slightly over-reach when you imply, without having read the book, that our emphasis on redistributive policies is based on not realizing that politics is primary. We have an entire chapter arguing why we think that the primacy of politics is way over-sold. It is easy to say that they should get buy in from warlords. But who do you think will take on that job (another warlord? why would he be trusted?). Our recommendations were based on a simple idea, very much rooted in politics: An effectively implemented redistributive policy is a very good way to give a new state a clearer identity in the minds of the voters. This can create ownership and start a virtuous cycle where the majority has a stake in fighting against the take over the state by one group. It also creates a basis for developing state capacity; focusing on doing one or two things well is the best way to give the state’s agents credibility and build their skills.
It is interesting that you chose to remove a little part of what Esther said: “Second, I think I would try to convince them to run anti-poverty policy in a more intelligent way than what we see in most countries. In particular, I may try to encourage them not to listen too much to the elevator pitches of all the other experts, and stake their entire policy course on the basis of those…”
We think the key is that these are problems that will probably not be addressed in the format of a blog (which is why we wrote a book…) so we will leave it at that. Hopefully this will be enough to give you and your readers the desire to find out where our ideas come from.
Great discussion this.
Well, there is little doubt that both politics and contexts would matter. Without knowing the details, one cannot really say what would work. Irony is even if you know the situation, it is difficult to know what would work. Like Blattman says roads is what matters. No doubt they do. But this experiment in Malawi shows that without a subsidized motorised transport people still use either cycle or walk. So not much gains from good roads.
Development is very tricky. Knowing these intricacies, I would have wished Banerjee-Duflo to skip the question and instead say we need to do experiments in the region to figure that out.