International Monetary Fund

Share This

Still the Bottom Billion

Finance & Development

PDF version

Glenn Gottselig interviews Oxford economist Paul Collier

Listen to the podcast

F&D: In a recent presentation to IMF economists, you spoke about the macroeconomics of the bottom billion. What do you see as the macroeconomic challenges that these countries have in common?
I think the countries of the bottom billion, the low-income countries, are distinctive not just in terms of having on average fewer good policies than the middle-income countries; they’ve got different problems. So good policies in those environments would just look different from good policies in middle-income environments, and that’s not sufficiently recognized. Let’s start with what the key differences are between a low-income economy and a middle-income economy. Out of those differences will emerge different strategies and different responses.

The overarching difference is that these countries are desperately capital scarce. The implication of that is that they need to go through a prolonged phase of high investment. For the moment, in Africa the average investment rate to GDP is less than 20 percent, whereas to catch up, to converge with other economies, it needs to be over 30 percent. So they must move from under 20 to over 30. That’s a big change.

The Bottom Billion

F&D: How can we do that?
It means an agenda of raising the capacity to invest productively. I call that a phase of investing in investing. It is something that has partly a macroeconomic agenda, but also a microeconomic agenda. If we just say it’s hopeless, the country doesn’t have a capacity to invest, it drives them into what I call the economics of Polonius: “Neither a borrower nor a lender be.” The economics of Polonius was always mocked. In Hamlet, Shakespeare set Polonius up as a pompous fool, basically. For low-income countries, neither a borrower nor a lender be would be ruinous, because they could never finance the move to investment rates above 30 percent.

Of course, in moving up investment rates and borrowing, we don’t want to repeat the debt crisis, so investment has to be done much better than it is at the moment. That is the strategy for investing in investing, building the capacity to make good investments. And it is something that the Fund can’t do on its own. It’s largely a microeconomic agenda, and so the macro depends upon the micro. The IMF needs to work with the agency responsible for that micro agenda, which is largely the World Bank. Of course, in principle the IMF does this, but in practice not enough. The two institutions need to work on a common core agenda of investing in investing, which is something that might take about three years working with governments to get the capacity for good investment decisively raised.

So capital scarcity is the overarching defining feature of low-income countries, but it’s not the only distinctive feature. Typically, they are resource rich, and this raises issues also of savings and investment. Countries are depleting their natural assets as they extract resources, and because commodity prices are very volatile, the revenue stream is very unpredictable. Such circumstances call for savings and investment strategies that are distinctive to low-income countries.

The objective of reducing absolute poverty—the only objective for the bottom billion—is one from which we’ve been diverging for 40 years from the rest of mankind. The primary focus must be on convergence. They’ve got to catch up; that means they must grow faster than other developing countries. And for that they need investment rates that are at least commensurate with the successful developing countries.

A third feature that makes these low-income countries distinctive is that they need to live down the past. The past has usually been rough, and so these countries often lack good reputations, especially with investors. Given this situation, they need commitment technologies—by which I mean some mechanism through which these countries commit beforehand to certain actions to be taken later and, thus, can build credibility. Part of the IMF’s core business is providing commitment technologies via conditions inherent in lending programs.

Finally, there is typically low capacity to actually implement public expenditure to adequate standards of honesty and efficiency. So building systems that actually achieve that is part of the agenda for both the Fund and the World Bank in the low-income countries.

F&D: As governments of advanced economies around the world shift their spending priorities to deal with their slowing economies, how important is it that aid continues to flow to developing countries?
Now is the time for public resource transfers to the poorest societies, because private resource transfers are falling. Private resource transfers have been in two forms, remittances and private investment, and both are falling fast. For the low-income countries, this crisis was not of their making, but they’re suffering from it. This was the sort of situation for which the public agencies for development were created. Looking back to the late 1940s, it was a somewhat analogous time, where there was no hope that private capital would rebuild Europe, and so public institutions were created to channel public money into that task. And now it’s a different set of countries that need help and international public efforts. That’s the right solution for the present time.

F&D: In recent months the G-20 has taken on a more prominent role as a forum for the major governments. Do you see this as a positive development for the bottom billion?
Very much so. What has happened over the years is that the group of countries that are credibly part of the solution to international problems has expanded enormously. Sixty years ago it was all down to the United States. Then Europe came from being a problem to being part of the solution. And now a whole new class of countries, like Brazil, China, South Africa, are part of the solution. I’ve just been working in Haiti, where 9,000 Brazilian peacekeeping troops have kept order and peace for the last five years. That’s a huge contribution by Brazil to the poorest country in the Western Hemisphere.

The G-20 recognizes, rather belatedly, the reality that we’ve thankfully moved from the G-1, which was true 60 years ago, to the G-8, and now the G-20. That’s something to celebrate, and it’s something that institutional architecture has also been late to recognize.

F&D: What’s the IMF’s role in helping low-income countries, in your view?
Good policies in low-income countries are not going to look the same as good policies in developed countries, and so there is no model up there in the sky called “soundness” to which we all aspire, with the stratospheric clouds being the G-8 and the very low clouds being the low-income countries. That’s not our world.

I think that there are three different roles for the IMF. First, for governments of low-income countries, the Fund is a source of money. Second, the Fund provides a commitment framework for donors through its programs. And the third role, which I think is the most important, is one of providing a conceptual and coordination framework to assist the many different players in the low-income development field, including various agencies and the different governments.

But my larger point is that the right macro answers depend on resolving the micro and institutional issues. The right macro answers, taking the micro and institutional as given—which is what the IMF has been doing—are the wrong macro answers for development.

The implication of that is that the IMF cannot walk away from the micro and institutional agendas. The IMF obviously cannot do everything, but it has to learn how to merge its low-income work with the other agencies that cover the micro and institutional angles. You need joint teams, codirected around common directives, particularly with the World Bank.

F&D: What are some of the obstacles to productive high levels of investment in the public sector?
One is the identification of projects and another is the implementation of projects. We know that the identification of projects is very badly done. There is, of course, an economic technology for identifying projects, called cost-benefit analysis, but I often wonder whether that’s a realistic approach and whether there is a more effective shortcut.

It’s not even always a desirable approach, because cost-benefit analysis of projects works on a discrete, piecemeal basis. Yet the business we’re in with these low-income countries is the business of trying to turn them into middle-income countries, and we try to do that quite quickly. And the piecemeal approach, looking at one project at a time, doesn’t really capture all of the effects that are external to a single project.

Take something like the big arterial transport links that would get landlocked Africa better stitched to the coast. In 40 years, those transport links have not been done. Those investments are not being made. Why? Because even when we get to cost-benefit analysis, and usually we don’t, but even when we do, they fail the cost-benefit tests, which are done nation by nation and take all of the other parts of the economy as given. I think it makes more sense to take a leap and ask what is typical of public infrastructure in middle-income countries, which is where we have got to get to. And given we are so far off sensible public investment planning at the moment, maybe we need this type of a shortcut.

F&D: What determines success or failure?
I’ve just been analyzing the data set of all World Bank projects, thousands of them, to see what’s the difference between success and failure. In particular, I looked at postconflict environments. And the answer is that supervision is much more important in determining the difference between success and failure.

Now admittedly, postconflict environments are at the extreme end of low-income countries. After a conflict, the private sector has retreated during conflict away from anything that’s formal. During conflict the state is predatory, and so the private sector learns to escape; it informalizes. Postconflict reconstruction is partly about coaxing the private sector back into formal structures, and if you hit the formal sector with high taxation because you’re trying to build revenue too fast, you retard that more fundamental process of rebuilding the postconflict economy.

In a postconflict environment, there are some issues common to both the public and private sectors, and a key one is the high cost of capital goods in low-income countries. These capital goods include structures produced by the construction sector. In low-income countries with low investment rates, the construction sector is small, and again, postconflict is an extreme example. During conflict the construction sector withers away because nobody’s doing construction. The society is focused on destruction. And so, once conflict ends, you inherit a tiny construction sector. But what you desperately need in a postconflict environment is reconstruction. And so the intense demand for reconstruction collides with a tiny construction sector, and what you get is a very steep supply curve in the construction sector.

This is microeconomics with macro implications, because what it means is that even if you spend a lot on investment, public or private, you don’t buy very much. Your spending gets dissipated in marching up that steep supply curve. And so a policy priority in low-income countries is to flatten that construction supply curve.

F&D: How do you do it?
Again, it’s coming down from macro toward micro issues and involves looking at the chain of production in the construction sector. Often in these environments there are legal bottlenecks that prevent access to land for construction. There are bottlenecks in material imports—cement is a classic bottleneck. There are bottlenecks in skills—a minimum level of construction skills is needed, and that means investing in the education capacity that builds those skills. And finally, you need organizations—firms that specialize in construction.

Typically, there is somewhat of a bypass of the domestic construction sector by bringing in foreign construction firms, and that’s throwing the baby out with the bathwater because potentially the construction sector can generate a lot of employment in these economies; in postconflict situations, that’s enormously valuable. In technical terms, the shadow wage of young men in postconflict environments is negative. It’s worth spending money employing them even if they were to do nothing. But actually you can get them productively employed in the construction sector.

F&D: What about resource-rich developing countries?
This is the area that I’ve been working on probably the most for the last few months. Resource-rich countries are distinctive in that natural assets pose problems of depletion and problems of price shocks. So depletion and volatility go hand in hand with resource riches, and each has implications. The depletion of natural assets obviously has implications for savings. As you run down one set of assets, you need to build up some offsetting asset. Maybe not one for one, but you certainly need to build up, and that implies that for the resource-rich countries, the savings rates need to be even higher than for the other low-income countries.

I’ve suggested that the typical low-income country should be investing something like 30 percent of GDP. And for low-income countries that are depleting natural assets, it should be higher than that. In what form do these countries save? Low-income countries are not Norway. They are not capital abundant. They don’t have a lot of capital per worker. On the contrary, they have the lowest capital per worker on Earth, and so evidently they need to use the savings toward the capital stock in the country. We need a phase of investing in investing, and this goes back to my earlier point. An investing-in-investing phase is even more important in the resource-rich low-income countries.

Now, how to manage volatility? Well, the standard approach is liquid savings, but I have come to the conclusion, fairly reluctantly, that the resource-rich low-income countries have no choice but to take the volatility within the real economy rather than trying to smooth it through sovereign liquidity funds. I don’t want to take that to the extreme because that would eliminate entirely some role for liquidity funds, but my point is that the objective should be much more modest than actually stabilizing public spending.

If we accept that public spending is going to be volatile, into what part of public spending, investment, or consumption should the volatility be channeled? And here the modern economics textbooks instruct us that it is a bad idea to have volatility in public consumption; there is such a thing as habit formation, so that bringing down public consumption is socially costly once habits are formed.

Volatility in investment is not as bad, because you can have quite big fluctuations in investment, and they translate into only very tiny fluctuations in the capital stock. They stabilize public consumption, and in so doing they pretty closely stabilize the capital stock, but investment is left to fluctuate.

Now, the low-income countries still must do this without messing up the rate of return on investment, and that complicates the investing-in-investing agenda because, for the resource-rich countries, it means you’ve got to have a capacity to change the investment rate. But remember, they will be fluctuating around a high rate of investment, say 35 percent. Maybe at the most extreme, they will be moving between 45 percent and 25 percent. That 20 percentage point swing is less disastrous than moving between 19 percent and minus 1.

F&D: How badly has the global economic crisis affected low-income countries?
The main impact is on the public sector through drops in commodity prices and corresponding drops in revenue. And in my own very recent work, the severity of a falling commodity price is, through its implications for GDP, dependent upon prior structural policies. So it is not just a matter of how a country responds after the fact—if the country is shock prone, it can design policies so that it reduces the macroeconomic consequences of commodity price falls. The micro agenda that seems to work there involves freeing up firms to be able to enter quickly and exit quickly. Using the database from the World Bank’s Doing Business surveys, this is what we find. Where countries have easy entry and exit for firms, the consequence of a falling commodity price is much smaller for GDP.

The money that the G-20 has come up with, to be channeled through the IMF, is ostensibly, I suppose, for balance of payments support. I think the balance of payments support should be directed toward supporting the public sector increasing its fiscal deficit. Then the question is, what does that imply for different parts of public expenditure?

If investment were optimal, then, as I alluded to earlier, these countries would take the shock by letting investment fall. Unfortunately, investment in the low-income countries isn’t anywhere near optimal. It is far too low, and so they are in a dilemma. The strategy of taking the shock on by reducing public investment takes us in precisely the wrong direction for the longer term. Where public recurrent expenditure has recently risen, then it seems obvious that it should be brought down again before habits are formed. So before countries get used to these higher levels of public consumption, public consumption should be brought down. Public investment should be protected.

More radically, at the same time as increasing the fiscal deficit, I think there’s a case for shifting the composition of public spending quite sharply from consumption to investment. This goes back to the commitment problem. These are low-credibility environments, and in low-credibility environments, raising the fiscal deficit while cutting investment is easily construed as a signal of populism. The right response is to counter the potentially damaging signal of an increase in the fiscal deficit with a robust signal that this is a government that is trying to protect the future by increasing investment and lowering public consumption.

Finally, a negative shock is often an opportunity for major policy change. Crisis is opportunity. To my mind the central opportunity that has to be seized is for governments to adopt this investing-in-investing concept and get serious about trying to raise the capacity to invest. The crisis is an opportunity to refocus policy priorities toward this building of the long-term capacity for investment.

Paul Collier is Professor of Economics at Oxford University and Director of the Center for the Study of African Economies. Glenn Gottselig is on the staff of F&D.

Finance & Development Logo

Write to us

F&D welcomes comments and brief letters, a selection of which are posted under Letters to the Editor. Letters may be edited. Please send your letters to

Free Email Notification

Receive emails when we post new items of interest to you.
Subscribe or Modify your profile