Is the World Bank Still Relevant? - Shanta Devarajan

Summary

The CGD Podcast: "Is the World Bank Still Relevant?" — Shanta Devarajan

Date: April 14, 2015
Host: Rajesh Merchandani (Center for Global Development)
Guest: Shanta Devarajan (Chief Economist, Middle East and North Africa, World Bank)

---

Episode Overview

In this episode, Rajesh Merchandani sits down with Shanta Devarajan during the World Bank Spring Meetings to probe the question: Is the World Bank still relevant after 75 years? The conversation explores how the Bank's traditional lending model is holding up in a rapidly changing world, why knowledge transfer may be more important than finance, challenges posed by fragile states and persistent poverty, and the evolving nature of development aid—touching on global public goods and the need for institutional evolution.

---

Key Discussion Points & Insights

1. The World Bank’s Enduring Relevance: Knowledge vs. Lending

• Devarajan asserts that the World Bank remains highly relevant, primarily as the “greatest repository of development knowledge in the world” ([01:09]). The power of accumulated experience and knowledge transfer is positioned as central to the Bank’s ongoing impact.
• He acknowledges the Bank’s origins as a lender but notes how private capital flows now dwarf Bank lending (“capital flows of the order of $400 billion a year, private capital flows…our lending per se as a fraction is pretty small” — [02:00]).

2. The Knowledge Edge and Welcoming Competition

• Devarajan highlights the positive aspect of countries accessing private finance, viewing it as a sign of policy improvements and ultimately a success for the Bank ([02:53]).
• He articulates a vision where the Bank’s financial arm “should declare victory when it runs out of business,” but maintains that “knowledge assistance will always be necessary” ([02:53]).

3. Graduation from Lending & Persistent Poverty

• The Bank’s model fosters the graduation of countries from concessional to market loans as their economies grow, but Merchandani challenges that this doesn’t necessarily mean poverty is eliminated within those countries ([03:33]).
• Devarajan agrees, emphasizing that future poverty will be “highly concentrated…in the fragile states” and within “pockets” inside graduated countries ([04:13]). He argues that ending poverty by 2030, one of the Bank’s twin goals, will require focusing on these new challenges.

4. India as a Case Study: Money vs. Incentives in Education

• The conversation spotlights India: although still home to hundreds of millions living below the poverty line, India borrows relatively little from the Bank compared to its own budget allocations ([05:22]).
• The Bank’s value-add lies in international learnings—implementing randomized controlled trials and knowledge sharing:

  "Education is not a question of money, it is a question of incentives. And we are trying to strengthen those incentives." — Shanta Devarajan ([05:56])

5. Skepticism About Knowledge Impact

• Merchandani plays “devil’s advocate,” raising whether the knowledge is really making a difference given persistent problems in countries like India ([06:36]).
• Devarajan responds with humility, arguing development is “a constant process of learning and learning from your mistakes” and that the Bank enables cross-border learning ([07:04]):

  “There’s no silver bullet to development… individual countries don’t always want to be responsible for their own mistakes. But we can observe the terrain… and tell another country, don’t try this. It didn’t work here.” — Shanta Devarajan ([07:04])

6. Tackling Fragile States

• Devarajan stresses that in fragile or post-conflict countries, Bank financing remains vital due to the lack of private or bilateral capital ([08:46]).
• He suggests fragile states are “qualitatively different from non-fragile states,” being stuck in low trust/low service equilibria ([09:10]):

  “…the citizens have lost trust in the state and you are stuck in this low level equilibrium…” ([09:10])

• To spur real change, disruptive interventions and special funds (“like a venture capital fund”) are advocated. The Bank is already directing more resources to fragile states, with early signs of success ([10:40]).

7. Should the Bank Lend to Sub-National Entities?

• Bank lending is predominantly to sovereign governments for creditworthiness and AAA rating stability ([12:04]). Subnational projects usually require national guarantees ([12:38]), but the Bank often works directly with state-level authorities in project design and implementation.

8. Income Measures and Inequality Challenges

• Repayment calculations and concessionality are based on national income per capita, but growing inequality within countries raises questions about this metric’s adequacy ([13:30]).
• Devarajan notes classification and concessionality are more about creditworthiness than income distribution ([13:50]). He uses Gabon as an example—a middle-income, resource-rich country with deep poverty and poor health outcomes but adequate revenue to repay loans ([15:44]).

9. Global & Regional Public Goods: Institutional Constraints

• The Bank’s country-based model doesn’t always fit well with issues like climate change, water basins, or other global public goods ([17:24]).
• Some innovations are in place, such as funding regional projects when multiple countries coordinate ([18:40]), but overall, these mechanisms remain “underutilized” ([19:31]).
• Devarajan argues for the creation of a new “window” to directly funding global public goods ([20:05]).

10. The Future of the World Bank

• Looking ahead 25 years, Devarajan envisions the Bank moving “in these two directions”: a shift away from country-focused lending, toward tackling regional/global public goods and targeting pockets of poverty at the subnational level ([20:33]).

  “…we might see [the Bank’s offices] in Patna, in the capital of Bihar, or in Kano, Nigeria, rather than in Abuja. Because that’s where, if that’s where the problem is, that’s where the World bank should be.” — Shanta Devarajan ([20:33])

• Merchandani concludes that true success would be if the Bank rendered itself obsolete ([21:26]), to which Devarajan replies with a memorable anecdote:

  “…we should convert the World Bank building into a museum of poverty. So people never forget what it was like.” — Shanta Devarajan ([21:58]) “I hope entrance would be free.” — Rajesh Merchandani ([22:02])
  “It’s $1.25.” — Shanta Devarajan ([22:05])

---

Notable Quotes & Memorable Moments

• “Our financing and ARM should declare victory when it runs out of business.” — Shanta Devarajan ([02:53])
• “Development is not a silver bullet. There’s no silver bullet to development. It is a constant process of learning and learning from your mistakes.” — Shanta Devarajan ([07:04])
• “Fragile states might be qualitatively different from non-fragile states…” — Shanta Devarajan ([09:10])
• “If Gabon wants to borrow from us, it has to be a project where we can make the case that’s going to benefit the bottom 40%…” ([16:29])
• “I wouldn’t be surprised if we start seeing, instead of having our country office in New Delhi, we might see it in Patna, in the capital of Bihar, or in Kano, Nigeria, rather than in Abuja…” ([20:33])
• On the prospect of ending poverty and the World Bank’s legacy:

  “We should convert the World Bank building into a museum of poverty.” — Shanta Devarajan ([21:58])

---

Important Timestamps

| Timestamp | Segment | |-----------|-----------------------------------------------------------------| | 01:09 | World Bank’s knowledge repository described | | 02:00 | Scale of current capital flows, challenge to Bank’s lending | | 02:53 | Knowledge assistance is the future | | 04:13 | Poverty’s shifting geography | | 05:56 | Bank’s role in India: knowledge over funds | | 07:04 | Is Bank knowledge really making a difference? | | 09:10 | Unique challenges of fragile states | | 12:04 | Why Bank lends to sovereigns, not subnational entities | | 13:50 | Debating per capita vs. median income for concessionality | | 15:44 | Gabon: resource-rich but poor; the dilemma of classification | | 17:24 | Global public goods and Bank constraints | | 20:33 | The World Bank’s evolving future: regional and subnational role | | 21:58 | “Museum of poverty” vision |

---

Conclusion

This episode presents a candid, insightful examination of the World Bank’s legacy and future. As the development ecosystem grows more complex—with poverty increasingly found in fragile nations and inequality remaining stubborn—the Bank’s greatest resource may lie not in its loans, but in its ability to transfer experiences and support innovation worldwide. The conversation concludes with a hopeful vision: that someday, the World Bank may have succeeded so fully in its mission, its building might serve not as a workplace, but as a museum chronicling the now-solved problem of global poverty.

Transcript

[A]

Foreign.

[B]

This is Rajesh Merchandani with the latest edition of the CGD podcast. Now, anyone who works in Washington in anything that's a sort of globally focused role knows that this is not a good week to take annual leave, as I found out in the three months that I've been here. That's because this is the week that the World bank spring meetings take place, when the city is full of important, very senior finance and economics ministers and their representatives from pretty much every country in the world. To coincide with the spring meetings, CGD and my colleagues here at cgd, Scott Morris and Madeleine Gleave, have just published a new paper called the World bank at 75. It's looking ahead to 2019, when the bank will mark three quarters of a century since it was originally 70 set up. My guest to talk about this from the World bank is Shanta Devarajan, the chief economist for the Middle east and North Africa region. Shanta, great to see you. Thank you very much for joining us.

[A]

Thanks for having me.

[B]

Rajesh, my question to you is, after 75 years, is the World bank still relevant?

[A]

Of course the World bank is still relevant. This is the greatest repository of development knowledge in the world. We have almost 75 years of experience working in almost every developing country and in almost every sector. And that, as we have learned, and CGD is the prime example of this, that development happens through knowledge. We have accumulated and are using that knowledge around the world. And that will continue to be relevant as the development agenda moves forward.

[B]

Okay, but the World bank wasn't set up to be just a repository of knowledge. It's a lending organization. Its core model has been remarkably resilient in in all these years. You lend to countries and repayments are based on national income per capita. Is that model still relevant?

[A]

Well, that model is a model for financing the knowledge work. So the capital flows are important. They're huge, off the order of 30, 40 billion dollars a year. But you could also make the case that when the World bank started, it was the only show in town. There was very little capital flows at that time. Today, capital flows of the order of $400 billion a year, private capital flows. And so our lending per se as a fraction of the global capital flows is pretty small.

[B]

But that's what I mean. There's much more competition for the World bank in terms of developing countries getting finances, whether it's from mobilizing their own tax revenue, from remittances, which are huge, to other multilateral bank which are kind of as we used to say in journalism, eating your lunch.

[A]

All the more reason why it's the knowledge that's going to make the difference in development. And we welcome the fact that developing countries are able to attract private capital flows because that's a sign that they're actually improved their policy framework to the point where private firms, banks, want to lend to them. That's the goal of the World Bank. If you like, our financing and ARM should declare victory when it runs out of business. You want these countries to graduate from financing, but that knowledge assistance will always be necessary just as much as they're necessary in developed countries.

[B]

I mean, in terms of graduating, as you say, that's the term you give when countries sort of move up the scale from concessional lending to hard loans to not having the need for loans. Our analysis shows that by 2019, there'll be far fewer countries who are World bank borrowers. At the moment, I think, you know, it's more than 100 countries in the world, more than half the population of the world. By 2019, it probably won't be half the population of the world in countries that are World bank borrowers. But more importantly, that doesn't mean that poverty is necessarily being eradicated. Although According to the MDGs, we're on our way. It just means that poverty is in different places.

[A]

Well, there are several points to that. One is it does mean that poverty will be more highly concentrated, for instance, within countries or in countries. It'll be concentrated in the fragile states, the thing that we were talking about earlier this morning. But that again means that if you want to end poverty by 2030, as one of our twin goals claims, we need to double our efforts in figuring out what it's going to take to end poverty in fragile states. So we want the entire international community to be able to give us the support to help these fragile states. Similarly, the countries that have graduated, so called, supposedly graduated, still have, as you suggested, pockets of poverty.

[B]

Poverty.

[A]

And those pockets are not small. I mean, India still has 300 million people living on $1.25 a day.

[B]

If India stops being a borrower, then how are you going to help eradicate poverty when there's 400 million people living below the poverty line in India?

[A]

But India stopping being a borrower doesn't mean that we don't work on poverty in India. In fact, the amount that India borrows from the World bank is tiny compared to the amount of private capital that India is able to attract, not to mention their own budget. I mean, we've been working, for instance, in education in India. I think we've done about $3 billion of education loans. India's own budget for education is 300 billion. So we're a drop in the bucket in terms of financing.

[B]

So what are you bringing to that then?

[A]

We are bringing international experience on how to manage education programs, how to incentivize teachers. We have been doing randomized control trials about different ways of incentivizing teachers across India, including contract teachers, including school lunch programs, including these programs of trying to give teachers a bonus if they show up for work and supporting, for instance, transparency of information. What we find is that education is not a question of money, it is a question of incentives. And we are trying to strengthen those incentives.

[B]

Now a skeptic. And far be it from me to play that role, but seeing as there's no one other in the room, I will might say that if in 75 years of, say, working with India, you haven't managed to eradicate poverty in India and you haven't managed to radically change the education system in India, then the knowledge might not actually be worth a great deal. Maybe you haven't got the right knowledge.

[A]

It's a fair question. But much of the knowledge is accumulated based on experience about what works and what doesn't work. I think it's fair to say that in the old days, we thought development was just a financial problem. So we would throw money at the problem, so we would build a school. And we spent a lot of money building schools. And the Indian government spent a lot of money building school, their own money building schools. And then you find that the teachers were absent 25% of the time. Now that's a finding that came out of a World bank research project that it was in 2003. Right now we're trying to say, well, with that finding, how do we change the way we do business? Development is not a silver bullet. There's no silver bullet to development. It is a constant process of learning and learning from your mistakes. And that's, I think, another role that an international organization can play because individual countries don't always want to be responsible for their own mistakes. But we can observe the terrain, see what doesn't work in one country and tell another country, don't try this. It didn't work. Here.

[B]

Let's go back to something you were talking about before. Fragile states, something that you will have a lot of experience of with your remit, particularly Middle east and North Africa. The trend is that as fewer countries are poor in the world, poverty will remain concentrated in fragile and conflict and post conflict states. So what do you think organizations like the World bank can bring to that to help those countries more in the future?

[A]

Well, for one thing, these are the countries for which World bank money is actually extraordinarily important because they're not getting private capital flows, they're not even getting much bilateral flows. So a multilateral organization that can diversify its risk can be a major source of financing for these countries. But the other is that again from our research and our experience, what we're learning is that fragile states might be qualitatively different from non fragile states. So it is not the case that simply they're just weaker at whatever they do. But there may be a situation in many of these countries where the state has failed, has broken down so badly that the citizens have lost trust in the state and you are stuck in this low level equilibrium where citizens don't trust the state and the state doesn't deliver any services to the citizens, which reinforces their mistrust or their lack of trust in the state. And there is, once you recognize that we have to then think of if you like disruptive interventions that help these countries break out of this low level equilibrium. We can't just treat them as on a line with other countries, only worse. They are qualitatively different. And many of us have called for, and the bank has actually responded by saying that we should actually be concentrating additional resources on these countries rather than our usual method because we allocate the concessional resources according to, among other things, the quality of governance in a country. So these states actually, under normal circumstances would get very little money because the quality of governance is very poor. But because of that, the quality of governance doesn't improve. So we're actually been advocating that they should have a special fund that is, it's a little bit like a venture capital fund. You're taking a risk on Guinea Bissau and saying if we actually support Guinea Bissau, knowing that the place has had multiple coups and civil wars and everything else, but unless we actually show a big show of support, it may never get out of this equilibrium. And I think this is working. I mean, we've actually, the IDA deputies have agreed to having a special fund for fragile states. And just recently you're seeing countries like Guinea Bissau coming out of this trap quite rapidly.

[B]

Let's think about countries moving out of kind of poverty again and say countries that move to sort of becoming middle income status. We talked about earlier that the World Bank's kind of core model is that you only lend to states and repayments are based on national income per capita. When poverty is focused in pockets within what might be middle income countries. Is it, is it the right idea to only lend to states? Should you think about lending to sub national regions or regions even?

[A]

The only reason why we lend to states is because we are a sovereign creditor and it's only states that can guarantee to pay us back. That's the only reason. If there was some way in which a region or a sub national entity could be certified that they will actually pay back so that the World bank can maintain its AAA rating, we would be happy to do so. But unfortunately the world is organized in such a way that we have to lend to states, but that actually you.

[B]

Can'T lend to like a region of a state and have that backed up by the state.

[A]

Well, see, but that's effectively what we do. In other words, we have lent, for instance in India. We call it a state loan actually. But what, what really happens is that the money goes to the federal government, which in turn on lends it to the state government. But the guarantee is provided by the federal government. But all of our negotiations, all of our discussions, all of our policy discussions are with the state government. There's hardly any discussion really in Delhi, when we were doing a state loan to Andhra Pradesh because of the kind.

[B]

Of guarantees you have to go through the state.

[A]

That's right. That's the only thing. But the rest of it is all with the state.

[B]

Okay. The other part of the model is, is national income per capita as the basis for repayment as countries?

[A]

You mean the basis for the concessionality of the loan? Is that what you mean?

[B]

I think so, yes. Yeah, but as if, as we're seeing that inequality within countries is growing, if you use national income per capita as a measure, then that's not necessarily representative of what the majority of people in that country are living on. Would it not be better to use say the median income?

[A]

Oh well, you know, the decision about what country to lend to, you know, what degree of concessionality to give a country is purely based on credit worthiness, which is simply whether the country is creditworthy or not, which it doesn't really depend that much on the difference between the median income and the mean income. It's simply whether the policy framework in the country is such that they can actually raise the money to pay you back.

[B]

Except that the repayments then might be hurting the poor in that country more.

[A]

No, no, but that's a function of the Policy framework, Does the country actually have a tax system that taxes the rich rather than the poor? So I don't think it's a matter of the median versus the mean. It's really a question of whether the country has the capacity to raise the tax revenue to pay back its loans. Now I think maybe what you're referring to more is this question that there are countries. And it's really, again, not to do with the creditworthiness, but simply the classification we classify countries. Low income, middle income. Now it's lower middle income, upper middle income. And there are some questions about that because there are some countries that become, say, particularly resource rich countries, mineral exporters which become middle income with lots of poverty. Take a country like Gabon. Gabon actually has a per capita income of $10,000, but it has the second or third lowest child immunization rate in Africa and very high levels of poverty and very high levels of child mortality. So, but Gabon is a middle income country. In fact, it's an upper middle income.

[B]

Country, which means there's a lot of.

[A]

Inequality within the country, huge inequality. But that's simply, I think it's simply a way of classifying countries. At the same time, frankly, Gabon is quite credit worthy because all of that oil money is concentrated in the state and so they don't really even need to tax people in order to pay back their loans. So I don't worry that much. I mean, the classification sometimes finds people. People have a lot of debate about it, including people at cgd, about whether something should be lower middle income or upper middle income. But functionally I don't think it makes much of a difference.

[B]

But a country like Gabon, then if you're happy to lend to it, then there's no incentive for that country to reduce its levels of inequality.

[A]

Oh no, but when we lend, we only lend for projects that we think will actually reduce inequality. So I mean, that's why we, that's the fact that we have these twin goals, you know, the ending poverty by 2030 and boosting shared prosperity, which is defined as the income of the bottom 40%. So if Gabon wants to borrow from us, it has to be a project where we can make the case that's going to benefit the bottom 40% or that's going to do something about poverty. And, and we have cases where we turn down projects. I don't want to mention the country, but our country wanted us to build an airport in a luxury seaside resort. And the nice thing was we could just say no because we have these goals. We're not in the business of financing European tourists to come to this country.

[B]

While we're scrutinizing the core model which talked about country lending and national income per capita, what about global public goods? That is a hugely important frontier. The environment, trans boundary agreements for rivers, ocean pollution, taxation, is it financial flows? These are things that require international cooperation and that the bank could really lead on, yet it's constrained by lending only to countries.

[A]

Absolutely, that's a very good point. I mean the bank is lending, I mean the bank is leading on some aspects of it. For instance, on the knowledge side we do a lot of work on climate change. For instance, we have a vice president in charge of climate change. He's also the special envoy on climate change. And we've done quite a lot of work on the adaptation side as advice to countries of how to avoid adapt to climate change. And even on lending we are not entirely constrained because we have something called regional projects. And for low income countries, a regional project requires I think more than two countries to agree on a regional public good, not necessarily a global public good like a river basin that you mentioned, where the countries, if they agree on a project that benefits both of them and it's a common project, so it's a common framework, then only a third of the financing of that project comes out of their allocation, their IDA allocation. So 2/3 of the project is financed out of a separate pool. And we've done that now with transport projects in Africa, with a trade project, even a science and technology education project in West Africa, Africa and things like that. But I have to say unfortunately our regional project window, I mean we put aside quite a lot of money for it is still underutilized.

[B]

But that's also the criteria for that lending is regionality, isn't it? It's not necessarily climate change mitigation or transparency.

[A]

But the reason is that mechanism or that modality for that to work you need to get all the countries together to agree on that project. Now you can do that with three countries on the Zambezi Basin or something like that. Try to do that with 190 countries on climate change. They are trying to do that. It's very difficult.

[B]

We've seen the results.

[A]

So I think we, and I would say that it would be nice if there were another window where the World Bank's finances could be used for global public goods. But up to now such a window has not been developed.

[B]

That's interesting. So in that case, what do you think the World bank in 25 more years is going to look like the World bank at 100.

[A]

I think it actually will go in these two directions that we've talked about that the country focus will become less and less important. And there'll be two other foci. One is regional and global public goods because that's not what the private sector is going to deliver on. And the other is sub national public goods. There'll be pockets of poverty in the country and I think the World bank has to focus on that. And I wouldn't be surprised if we start seeing, you know, instead of having our country office in New Delhi, we might see it in Patna, in the capital of Bihar, or in Kano, Nigeria, rather than in Abuja. Because that's where, if that's where the problem is, that's where the World bank should be.

[B]

Well, Shanta Devarajan, in some ways the World bank at 100 shouldn't really exist if everything goes to plan. That would be the goal really, to put yourselves out of business. We all want to do that.

[A]

Absolutely. And in fact I actually had this conversation with a former boss who's an architect and this will be relevant in a minute where we were idly speculating what would we do if we actually eliminated poverty. And he suggested something which still stick with me. He said we should convert the World bank building into a museum of poverty.

[B]

Wow.

[A]

So people never forget what it was like.

[B]

I hope entrance would be free.

[A]

It's $1.25.

[B]

Okay. On that note, Shanta Devarajan, really interesting to talk to you. Thank you very much indeed.

[A]

Thank you.

[B]

And don't forget, you can find out more about all of CGD's work on our website, www.cgdev.org. i'm Rajesh Merchandani and please do join me for the next podcast for from cgd, the Centre for Global Development.