Transcript

November 27, 2007
Washington, DC

SPEAKERS

Patricia Alonso-Gamo, IMF Policy Development and Review Department
Vasuki Shastry, IMF External Relations Department
Tilla McAntony, IMF External Relations Department
Roger Nord, IMF African Department
Soren Ambrose, Action Aid consultant
Eric Gutierrez, Action Aid South Africa
Specioza Kiwanuka, Action Aid Uganda
Paula Mendonca, Action Aid Mozambique

PRESENTATION

Mr. McAntony: Thank you for joining us today for this conference call with Action Aid representatives in Africa to discuss the IMF's Policy Support Instrument (PSI) and a recent paper on the topic prepared by Action Aid. Joining us today from the IMF is Mr. Roger Nord, from the African Department. And then we have Patricia Alonso-Gamo. She is with the Policy Development and Review Department. Roger, why don't you begin, and then Patricia will take over.

Mr. Nord: Thanks: Tilla. My name is Roger Nord. I am Assistant Director in the African Department, I'm also the Mission Chief for Tanzania, which is one of the countries that has a program under the PSI, and hence, I read your paper with great interest. I'm going to speak for about three, four minutes about the context, the macroeconomic situation in Africa and in the PSI countries in particular, and give you my perspective on how I think the PSI has helped these countries in particular. And then Patricia will talk about the PSI as an IMF instrument, how it contrasts with the PRGF, and how we see it going forward as you probably know the IMF's recent Regional Economic Outlook for Africa underlines very strong performance over the last few years and a positive outlook. Economic growth has averaged 6% over the last five years, which is double the level of the previous five years. Inflation remains in single digits, and external debt, which had averaged 45% of GDP in the late 1990's, has fallen to 11%. This provides many African countries with tremendous opportunities to reduce poverty and improve living standards for all. Indeed, many countries are using this fiscal space to raise spending, notably in the social sectors and on public infrastructure. Now, as we work with countries in Africa, I think there are three issues that I think I would emphasize in particular:

First, it's important that this fiscal space be used efficiently to provide better health and education and to address inadequacies in infrastructure. And that means a big emphasis on public financial management-getting more bang for the buck, if you will.

Second, there's a need for more efficient financial systems; in other words, better access to financial systems to foster private investment, in particular in the rural areas. But not only there.

And third, there's a need for countries to reduce the cost of doing business by lifting regulatory obstacles to get the private sector to invest. In the long run, growth and employment are not going to come from more foreign aid, they are going to come from more private investment.

These policies can only be successful if they are embedded in strong domestic reform programs. That brings me to the PSI. Including Senegal, for which a PSI was approved just a few weeks ago, there are now six such programs in Africa. And in all cases, the PSI is based on a well articulated national strategy; for example, the Nigerian Needs Program, or the MKUKUTA in Tanzania.

The targets of these programs are very much the countries' own. Inflation objectives, for example, are set by governments that are well aware of the social cost of high inflation. The decision to refrain from domestic borrowing in Tanzania, for example, reflects the government's preference to avoid an expensive source of financing and to rely instead on domestic revenue and concessional foreign assistance.

The choice of IMF members for a PSI reflects a commitment to sound policies. And signaling such a commitment can be valuable. It can be valuable to external audiences-donors and investors-to indicate the quality of the policy framework; also to domestic audiences, whether politicians or domestic markets. So it's best to think of the PSI as a way to credibly commit to good policies, similarly, say, to an inflation target or budgetary objective. Such public and transparent commitment has tangible economic benefits in setting priorities and expectations.

I would emphasize three ways in which the PSI does that. First, it has helped countries develop a macroeconomic policy framework and a structural reform agenda. It has served the authorities well as an internal discipline as policy issues have emerged, and it has provided a framework for targeted technical assistance and support from the international community, and increasingly from private investors.

Finally, let me say that these efforts are bearing fruit. Growth in PSI countries has exceeded the average in Sub-Saharan Africa. Public spending has risen by more, and poverty-with all the measurement caveats-is falling faster in PSI countries. Let me stop there and let Patricia talk a little bit about the PSI as an instrument for the IMF.

Ms. Alonso-Gamo: First of all, let me start by saying that we welcome the chance to engage with Action Aid on the PSI in low-income countries. And to tell you that after three years of experience with the PSI, the Fund will conduct a review of the instrument, probably next year.

First, let me tell you a little bit, about how the PSI was conceived. We did a survey with member countries, including those with programs under the Poverty Reduction and Growth Facility (PRGF), and with donors. Donors were relatively happy with the situation. But a number of countries-what we call mature stabilizers-said, "We want to graduate from the PRGF, but we want to have the framework that can signal very strong policies" They didn't want to have what other countries had at the time: intensified surveillance, or fund-monitored programs, etc. They said, "We want something stronger that is endorsed by the Fund's Executive Board, and we can tell the world that we have good policies."

So it was very much demand driven. And the characteristics of the PSI were very much tailored to what the countries wanted. They wanted something that was rooted in the Poverty Reduction Strategy Papers (PRSP) or in the national process. And they wanted to set their own targets, and they wanted something structured. But at the same time, because they were mature stabilizers, the type of conditions that they would have would be different because they were at a different level of development in their reform.

But they definitely wanted something that was endorsed by the world and that would signal very strong policies. Our initial proposal was to have reviews not at regular intervals like in the PRGF. But it was the countries themselves at the Board that asked to have this regular review process. The main point is that the reviews of the PSI are more nuanced. It's not just to reach the conclusion of the review. One important element is that there's a nuanced assessment as to what are the things that are going well and not so well because donors can then make up their own minds as to whether they're going to disperse aid or not. A lot of donors do not have this automatic link to the conclusion of the review, but they want to have the good information to make up their own minds as to whether they want to proceed with disbursements or not. That was the main objective.

Now, if I turn the page to what the PSIs have actually done, as Roger says, there are now six countries that have PSIs. One of them is already on its second PSI: Uganda. Nigeria, which is coming to the end of its first PSI, is considering a second one. In that case, it's very clear that there's no donor money needed. There's no requirement, no debt relief needed. It's just that they think that this framework will work for them.

And if I looked at comparing PSI program averages to the PRGFs that the countries had in the past, we can see that PSIs generally program higher levels of growth than in the previous PRGF, with the exception of Mozambique, which is programmed at 7% relative to 7.2%. All of them have growth objectives that exceed 6%. And spending and overall deficits excluding grants were programmed to arrive in five out of the six countries. And current account deficits were programmed to widen in all cases, except Nigeria where the sharp rise in oil prices has increased the current account surplus. But if you look at the non-oil current account, the deficit has widened considerably. Again, what matters is absorption of fiscal space.

Inflation targets, as Roger said, are set by the countries themselves, but although they tended to be in the 5% to 10% range, in some cases, the out-turns have exceeded 10%. But we don't target inflation, per se. What we target is a level of spending, which as I said, has widened. The inflation target is set by the countries, and we just build around that. The only two countries that had relatively low levels of inflation programmed were Cape Verde and Senegal, which have exchange rates that are pegged to the euro, which has been appreciating.

Uganda's initial PSI target is an average level of inflation of 4% and is revised upward to 5% in the second PSI so, again, all the inflation targets and out-turns have been in the 5% to 10% range, which is what most research says is a good level.

By the way, I want to clarify one point that's being made in many reports, which I think is a bit misleading. That point is the assertion that increasing inflation gives you more fiscal space. That is a misapprehension. What increasing inflation gives you is the illusion of more fiscal space: you think you have more money to spend. But unless there are more resources available, there just are no more resources there. You don't create more fiscal space unless there is an increase in the deficit of the current account, which means that more foreign resources are coming in, or there are other things that are not being spent; for instance, debt relief. Under those circumstances, you have those resources to spend on something else. So what we have to focus is on the real fiscal space, which means actually having more resources to deliver.

Mr. McAntony: I think we can start the discussion now. I think it'd be useful if you could identify yourself, and we can just start the discussion now.

Mr. Gutierrez: This is Eric Gutierrez from Johannesburg. Just a few basic questions on countries that get a PSI. Are they required to or do they have to make something - changes on their...growth strategies, and are these changes discussed with the IMF? Just some basic questions on how the whole process is implemented.

Ms. Alonso-Gamo: The starting point is a country's PRSP strategy. And then there is a discussion to see the kind of macro framework that can built in terms of setting the priorities and fulfilling the expenditure requirements. It's always clear that there are different paths and different scenarios that can be followed in a PRSP, and one has to look at the amount of resources that are available from donors and otherwise to try and choose a path that will do that. But, basically, it's rooted in the national strategy, and the PSI is built around that.

If the national strategy didn't make sense, then we wouldn't take it to the Board to be supported by a PSI. But the countries that have come requesting one are countries that are, as Roger said, mature stabilizers. They have a good track record. They have very good policies, and they want us to prove to the world and to investors that they have very good policies.

Mr. Nord: Typically, the national frameworks are very much a bottom up, grassroots effort. I'll give you the example of the MKUKUTA in Tanzania, which is the Swahili abbreviation for the PRSP. This starts very much at the local level, and the MKUKUTA is discussed over many, many months throughout Tanzania, and in the end it is crystallized into a massive national document with ownership well disbursed across Tanzanian society. The MKUKUTA sets itself targets for education, targets for infrastructure and a macroeconomic framework. And the PSI supports this national strategy and, given the IMF's expertise, it supports in particular the macroeconomic framework. But that macroeconomic framework is consistent with the objectives that the authorities have set themselves in terms of their national development.

Ms. Alonso-Gamo: If I may add, PSIs are reviewed every six months, and they're pretty flexible. So, for instance, if we looked at the Nigeria PSI, the original objective was to have spending at around 31% of GDP. By the end of the third or the fourth review, we were at 34% of GDP. The national program identified good spending projects that the Nigerians wanted to do. And the PSI was adapted to accommodate those projects.

Mr. Gutierrez: Following this line of discussion, what may be raised by a specialist in Uganda is the Bonna Bagagawale scheme in Uganda, which is a national priority. It's a domestic priority, and then it's based on analysis that some support needs to be given because there have been failures in providing credit to small farmers. So do you have any sort of particular response on this point, which was raised in our paper?

Mr. Nord: We looked at that, and, frankly, we're a little bit puzzled by the statement that that the plan was not incorporated in the PSI because, in fact, the program is very much a national initiative that goes well beyond the financial sector. Bonna Bagagawale, which in Swahili stands for "prosperity for all," is a comprehensive program to improve livelihoods in rural communities for the provision of extension services, infrastructure- for example, water-for production inputs such as fertilizers and access the credit. Frankly, the IMF has never challenged this. On the contrary, it's very much part of Uganda's national development strategy, which is supported by the PSI. So we're not quite sure where this particular allegation comes from, but not from the Ugandan government. That, I can assure you.

Access to finance is an important issue in Uganda. It's not easy, as in many African countries with a large part of the population living in rural areas. This is a real challenge. One way that the Ugandans have tried to address this challenge is through expensive microfinance and savings and credit cooperative initiatives. I think that's working relatively well. It's very much donor-supported. The IMF's involvement is to ensure that the sector is well regulated and that costs are reflected in the budget to the extent that these costs are borne by the government. But that's been the extent of our involvement.

Mr. Gutierrez: Roger was mentioning earlier that one way of expanding fiscal space is to have good public financial management, which means basically making government expenditures more efficient. But I just would like to get a quick review of what you think would be sources of fiscal space that countries should take into consideration when making those national plans. How do we get money to finance the programs and the targets that they're trying to achieve?

Mr. Nord: This will of course differ from country to country, but let me give you the example of Tanzania, which I think is probably representative of the way other countries would look at it. Over the past five years or so, Tanzania has been able to raise public spending quite significantly, by about 6% of GDP or so. And that was done in two ways. First, Tanzania receives significantly more foreign assistance today than it did five years ago. And in particular, the proportion of foreign assistance that comes as budget support early in the year, which the Tanzanian government can then allocate to its own spending priorities, has risen quite significantly and is now about 70% of the foreign assistance that Tanzania receives.

That flexibility has allowed Tanzania to channel the foreign assistance to its own priorities, which are education, health and infrastructure. For example, the current budget for 2007/2008 doubles spending on infrastructure, which they've seen as a national priority. So that's been one way that they've expanded fiscal space.

The second way in which Tanzania has expanded fiscal space is by targeting better revenue administration. Through quite significant technical assistance-both from the IMF and from others-and the streamlining of tax administration and the combating of tax fraud, they've been able to raise tax revenue by about 3% of GDP over the last couple of years without changing tax rates. That, of course, has given them more room for discretionary spending in the areas of their priorities.

This is the example of how Tanzania has done it. What's been important, as I mentioned earlier, is that they've made a commitment to avoid financing deficits through domestic borrowing, which has been very expensive and which would add a lot of interest costs to their budget. That, together with debt relief from donors, has lowered debt service tremendously in their budget. And that, again, has created fiscal space. Patricia, maybe you would like to add some cross-country perspective.

Ms. Alonso-Gamo: One thing that puzzled me was why the report by Action Aid didn't speak more about Nigeria, because Nigeria was our first PSI-and one of the key ones. They're a pioneer, if you wish. In fact, the numbers for Nigeria are quite striking because, as I said, priority spending in Nigeria is not only very high and growing, but it's actually higher in absolute terms than in any of the other countries. We're talking now about spending equivalent to 34% of GDP, which is a very large number. And if you consider the GDP of Nigeria, you're talking a very large number, indeed. Also, it has accommodated a great deal of externally financed projects in Nigeria, and key sectors that were for development. So we're a bit puzzled why this cross country analysis of the PSI doesn't dwell in the Nigeria case, which is very typical. It's also interesting that Nigeria is the one that has inflation at 10%. But that was something that puzzled me when I was looking at this study, and I think that the study really should include Nigeria case.

Mr. Ambrose: May I respond to that?

Ms. Alonso-Gamo: Yes.

Mr. Ambrose: This is Soren Ambrose. It seems to me that it's clear that Nigeria is an exceptional case in every way. Even though it was the first PSI country, it is the only one of the PSI countries not to have had a PRGF immediately before, so there was no basis for comparison between PSI experience and PRGF experience.

Ms. Alonso-Gamo: Actually, there was a basis for comparison because Nigeria had enhanced surveillance, i.e. a program that was monitored by the Fund for about two years prior to the PSI. This program wasn't approved by the Executive Board and endorsed by the Board, but it went to the Board, and it was published as such. So you had a perfect point of reference in the staff-monitored program.

Mr. Ambrose: Well, the -

Ms. Alonso-Gamo: I think if you have only give PSIs-Senegal was only approved last month-I don't think you can take 20% out of the sample and say it's an exceptional case. Each one of them in its particular way is an exceptional case.

Mr. Ambrose: Well, yes, of course, but this is an exceptional case in more ways that. We were trying to compare it to the PRGF specifically. But in addition to that, Nigeria has oil income that these other countries don't have, and all the press reports and everything else we heard. But it's simply the path that Nigeria was taking to the PSI was substantially different than the other countries were taking, which is to say that it was doing this specifically in order to qualify for payers club debt relief.

Ms. Alonso-Gamo: But as I mentioned earlier, they're negotiating a second one. They don't need the debt relief now anymore. So it's clear that it's completely unrelated to the debt relief. Mozambique has oil too.

Mr. Ambrose: Well, it's hardly comparable. Come on.

Ms. Alonso-Gamo: No, the point is that I think that when you do a serious review of some things, you can't exclude things from the sample. I think you have to look at all of them....

Mr. Ambrose: Well, I don't think that was an arbitrary exclusion. I do think that we were trying to do a comparison between the PRGF and the PSI. And the fact is that this was not a Board-approved PRGF document, so the basis for comparison was not the same.

Ms. Alonso-Gamo: I beg to disagree, but anyway, when we do the PSI review next year, we will use the full sample. There are a number of points that I think that we have to make, in fact, two corrections to things in the report. For instance, you say on page seven in Box One that PRSPs are subject to approval by the Fund. In fact, that is not true. The PRSPs are country-owned documents. In 2005, we in fact eliminated the requirement that the Fund and the Bank jointly assess the PRSPs. The board does not approve the PRSP's. Then if you turn to page nine, on Kenya and the price of negotiating a PSI, that also is not correct. There has been some interest but there are no plans for talks right now.

Mr. Ambrose: That was based on conversations with Kenyan Finance Ministry officials, which we later found out did not go forward.

Ms. Alonso-Gamo: The PSI is not surveillance. The report says on page 21 that it's surveillance. It is not. It's actually under a different article. Article nine is simply the form of technical assistance. The report on page six says that it's only a handful of countries will benefit from debt relief. In fact, it's 43 countries, of which 24 have gone through already.

Then the report says on page nine that conditions have been added at the time of the review. This is actually quite normal, because if you're looking at the three-year program, you don't want to micromanage three years down the road what may or may not be important at that point. And what you do is you re-evaluate at each point.

And as I said, they're fully country owned, and in many cases-in fact, for instance, in the case of Nigeria-we're there at the insistence of the authorities. But they insisted that they have to be part of the program, because they wanted to send a signal domestically. So even if all the conditions are there because they're part of the nationally owned program, in some cases we would have been happy even to have fewer conditions..

Mr. Ambrose: I don't think that we meant to argue that the process of adding conditions at the time of reviews was in any way abnormal. It was just an explanation of how we were doing the analysis. We were explaining that the original PRGF or PSI document that one would look at would not be a complete listing of conditions. One would have to look at the subsequent reviews in order to understand the full breadth of the program at any given time. So if we implied that this was somehow exceptional, I apologize for that.

Mr. Gutierrez: I think the main point really that we're trying to get in this paper is that there are indeed enlargements you have acknowledged and demands on the resources. So far, based on the research that we've done and the interaction that we've had with national governments, it's really a tight situation. As we've said in our title, we can understand belt-tightening when there are crises taking place, but as these countries mature, would those belt-tightening measures be loosened? So my question is more a general one on policy of fiscal space.

Ms. Alonso-Gamo: Again, I'm a bit a puzzled, because looking at the numbers for all the PSIs, spending has increased in all of them except in Uganda. It remains around 21%, 22% GDP in Uganda, and that was mainly because the country was going to have higher revenue-not because of lower spending. And in all the others, spending went up. So I'm a bit puzzled when you talk about belt-tightening, because I'm not quite sure what you mean.

Mr. Nord: I think what we've seen in most of the PSI countries, in fact, is that there has been an increase in spending, as Patricia says. And there's been an increase in spending for a variety of reasons. Debt relief is one of them. But the other reasons, as I mentioned before, are more mobilization of domestic revenue and more foreign assistance. So the belt-tightening metaphor doesn't really apply in the countries that have a PSI at the moment.

What may be useful, if you haven't done so, is to look at the Fund's last regional economic outlook for Africa. (See Economic Outlook for Africa). (http://www.imf.org/external/pubs/ft/reo/2007/AFR/ENG/sreo1007.htm).

In fact, it has a chapter on fiscal space with a couple of case studies, including PSI countries such as Tanzania, but also some non-PSI countries such as Ghana. And I think what it shows is that creating fiscal space is important, but using fiscal space appropriately, if anything, might be even more important. The two things together will provide for more effective public spending, which is very much a priority for many of the African countries' PSIs and other African countries that we deal with.

Ms. Alonso-Gamo: I want to go back to what I think is a bit of the red herring on inflation and fiscal space. Again, real fiscal space would provide more resources to spend on priority spending, on infrastructure, on antipoverty measures. That's real fiscal space. That can only come because you either have more resources that are coming from abroad-more aid, more loans or more investment. Or because you have to spend less money on debt servicing. That's where fiscal space is coming from. If you look at the spending level, you have to think that not only is the spending higher, but they're not spending so much money on debt service. So comparatively speaking, they have more resources for priority spending.

But when you say having a higher inflation level is going to help the fiscal space, it really doesn't. There's a lot of independent research that says that inflation of between 5% and 10% is fine. But beyond 10%, you are buying the illusion of fiscal space. You're not getting more resources. You're getting more money, but chasing the same amounts of resources. That's what inflation is. It's more money chasing the same goods, and therefore, prices go up. It's not a coincidence that we now have a higher period of consistent high growth in sub-Saharan Africa accompanied by the lowest level of inflation that sub-Saharan Africa has experienced. I think that when we talk about fiscal space-and we all want increased fiscal space-we have to focus on the issue of do we have more resources going to the priority spending? That's what we should be focusing on, because inflation is a red herring. It doesn't provide more resources for priorities.

Mr. Ambrose: I think that's a distortion of our argument. We are not arguing for inflation. That would be silly. It's, I think, a red herring to suggest that we are. What we are arguing is that the measures taken in order to meet or to anticipate trying to meet ultra-low inflation targets inhibit the spending on social services that would be beneficial to a growing economy. In fact, when you choose ultra-low inflation targets, you have the effect of handcuffing the decisions made by policymakers when they are allocating for wages and other components of the budget. So it is not inflation itself that we're arguing for. It is the process of how the budgeting is done and what the targets chosen do to the policymakers as they do the budgeting process.

Ms. Alonso-Gamo: Okay, two things. One is, if you have targets that are compatible with high inflation, you are going to get the illusion that you're spending more. You're not going to get....

Mr. Ambrose: Again, it's a red herring to say that we're arguing for high inflation. We are arguing for lucent inflation targets. We're talking a matter of a few percentage points here or there. We are not even arguing for a blank check of any sort. We're not saying don't have an inflation target. We're not saying have high inflation. We are saying, at times when a country is in a growth cycle, does it make sense to make one of the key targets go down, in the case of Mozambique, for example, from 7% to 6% for an inflation target when the country could go on an expansionary route that would begin to address some of the problems that the country has had in terms of number of teachers in the schools.

Mr. Nord: I think Mozambique's actually a good example to raise because right now it is an example where the inflation target was in fact revised. In the recently concluded first review of the PSI, the inflation target for 2007 was revised upwards to 8% to reflect precisely higher anticipated inflation due to higher food prices and oil prices.

These inflation objectives are not IMF conditionality, by the way; these inflation objectives are set by the countries themselves. These inflation objectives are not rigid and do not force countries into a straitjacket. On the contrary, they are objectives, which help anchor monetary policy, but which take into account the realities on the ground. That's precisely what happened in Mozambique. The inflation target was 5%. It's been increased to 8% for this year. And, for example, it has not constrained education spending, which has increased in recent years. Education spending increased from 4.5% to 6.5% of GDP over the last three years. And that's not an exception. That's rather typical for the PSI countries that we deal with.

Ms. Alonso-Gamo: I don't know how you define ultra-low inflation, but out of the six PSIs, we have four PSIs that range between 5% and 10% inflation. And 7% inflation means that all prices double in 10 years. That's not necessarily low inflation.

Mr. Ambrose: I'm looking at Senegal, for example, where the target is now to get back to 2% inflation. I would call that ultra-low.

Ms. Alonso-Gamo: Okay, Senegal has a fixed exchange rate. Senegal's currency is pegged to the euro. What that means in economic terms is that you are committed to the same inflation target as the currency that you're pegged to. That's what a fixed exchange rate means, so it's not tough setting the inflation target. It's the choice of the monetary arrangement that the country has made. Cape Verde and Senegal have fixed exchange rates. If you want to break the fixed exchange rate and for them to go for 7% inflation, that's fine, but they would have to abandon their peg to the euro. I don't think it's in their interest to do so. They don't think it's in their interest to do so. It's not something that the Fund is determining in any way. It's driven by the fact that they have a fixed exchange rate to the euro. That's something that you can't argue with in the sense that they either have to get off the peg, and then they can choose any inflation rate they want. Or, if they want to stay with the peg, they have to stick with an inflation rate that is in the euro range. They don't have a choice on that.

Mr. Ambrose: Well, that's something I'll have to look at more, in terms of how the euro peg affects that, but I'm not as familiar with exchange rates, how the euro affects the CFA and so on.

Ms. Alonso-Gamo: Basically, what it means is the CFA is pegged to the euro, so if you have the CFA franc as your currency, you don't have an independent monetary policy. You cannot set your own inflation at any level you want. You have to follow the inflation rate of the currency you're pegged to. Therefore, the countries that have a fixed exchange rate don't have the freedom to go and say, I'm going to have 7%, because after a couple of years they would have to abandon the peg, with everything that it entails. So I think maybe you should look into that and qualify a little bit the views on the two countries that have a peg.

(Questioner Unclear) Roger mentioned earlier the Regional Economic Outlook, and one of the countries mentioned there was Ghana, which is thinking of financing its public investment by looking at borrowing money from commercial sources. Now, I wanted to get your views on what you think about the mature stabilizers forcing (garbled) by various things like those exchange rates or various other issues and then having to resort to borrowing to finance well designed programs. What in general would you say about that? Can government pressuring the mature stabilizers, can they borrow responsibly, and if they do so, if it's - would there be some kind of mechanisms or criteria that needs to be put in place to make sure that the value of whatever is - it may be a bit - the borrowing may be more expensive as compared to other source of fiscal space, but it might - are there criteria that we would recommend in terms of how big and stretch the value of borrowing?

Mr. Nord: After the Multilateral Debt Relief Initiative (MDRI), debt levels have come down significantly in many countries in Africa, including in PSI countries, so that does provide room for borrowing. The framework within which we would look at that and which countries themselves look at that, is a debt-sustainability framework in which borrowing is put into a multiyear projection/ That gives you a sense of whether that debt service is going to be a burden that the budget and the balance of payments can support. That's the framework within which we and the countries themselves would look at it.

Ghana made the decision to go to international markets and borrow $750 million with the main objective of scaling up spending on infrastructure. Other countries are thinking about it, but I can tell you, for example, from my Tanzania experience that the Tanzanian authorities are very wary about doing so, precisely because of the cost, as you mentioned. Tanzania gets about $1.2 billion a year in concessional assistance, so the need to borrow commercially on top of that is not immediately obvious. Every country will have to make this tradeoff. But the way to look at it is within a multiyear framework. Maybe Patricia can say a little bit more about the debt sustainability analysis that we do.

Ms. Alonso-Gamo: A lot of countries, after MDRI, have a lot of room to borrow. The debt sustainability framework is supposed to help countries analyze possible lulls and see what the situation would look like 20 years down the line, and depending on different scenarios when they take on this borrowing. The idea is to balance-to take on enough and on enough good terms-that they can fulfill their growth objectives, and at the same time avoid becoming heavily indebted again. So debt sustainability is really a tool. That was what Roger was saying. Countries didn't become heavily indebted because they borrowed too much, but rather because they spent it on projects that didn't provide the needed growth, and therefore they were unable to grow and repay. So the issue is how not to fall on that trap.

The other issue is we now have new lenders coming on the scene, like China or India or Brazil. At the same time, we also have countries that have become attractive for private lenders. It's how to arrive at the correct mix because, obviously, some of those loans may have ties in terms of promising your future oil revenue to the lender. Or they can have higher costs, like private loans. And they should have to come to the correct mix and meet your servicing space. S

(Questioner Unclear) This point on what is the money used for, I think we've raised it a number of times in the past papers that we've done. For example, what we're saying is that in education, the timeframes of investing in education sometimes go beyond the usual timeframe in cycles that the IMF is looking at. So you're looking at the more medium- to long-term returns when you spend on education, the same way that there might be other public spending programs that have the kind of different, longer-term timeframe.

One question that we'd like to ask in this regard is - do you take into account those investments because there won't be any immediate short-term gains that can be observed from such an investment.

Ms. Alonso-Gamo: Yes, we do and in two ways. First, as I said, the debt sustainability analysis has a 20-year horizon. We're very much conscious that a lot of this investment-in infrastructure or human resources, education, health-has a long-term return. The second thing is that this sustainability analysis is done jointly with the World Bank, who have the expertise to see what the impact on long-term growth is of those expenditures on infrastructure and education. So in choosing the long-term growth path, we very much rely on the analysis of the World Bank that does take into account this kind of effect.

Ms. Kiwanuka: I wanted to ask what kind of advice the government of Uganda can get from the IMF. Currently, the government is focusing on revenue collection as the best way for mobilizing domestic revenue. (garbled) But the tax base is narrow, so we don't know where the revenue will come from in the long run. So what kind of advice would the Fund offer to our government to expand its tax base?

Mr. Nord: I'm not an expert on Uganda, but I'll give you an answer based on the information that I have. Currently, we are preparing the review of the PSI. The paper should be available soon. We're going to the Board in December, and the Ugandan authorities have committed, as usual, to publish the paper, so within a couple of weeks, the latest review of the PSI will be available.

Based on the experience of the last couple of years, Uganda has been able to expand its tax base by about 1% to 1.5% of GDP over the last couple of years. That brings it currently to a level of about 14% of GDP, which is roughly comparable to Tanzania, but lower than in a number of other African countries. And I think you're right, there is still in Uganda a fairly narrow tax base.

Mr. Nord: You asked what advice would we give.

Ms. Kiwanuka Yes.

Mr. Nord: I think the advice in Uganda would be quite similar to that in other developing countries, which is that you want to address the tax base by broadening the tax net, not by increasing tax rates. You want to limit the amount of exemptions that you give, notably to foreign investors, for example, because that reduces your tax base significantly, or can reduce your tax base significantly. And you want to put in place an effective tax administration. This has made a very big difference in Tanzania, for example. The Tanzania revenue authority has succeeded in-without raising tax rates-increasing the revenue to GDP ratio by about 3% of GDP over the last couple of years. And interestingly enough, the Tanzania revenue authority is now providing technical assistance to local authorities in Tanzania rather than being provided by foreign assistance or donors. It's the Tanzanian revenue authority itself that is conducting this assistance within Tanzania. I think that's evidence of its success in recent years.

Ms. Mendonca: My question is on participation, because Mozambique just moved to the PSI. In the PRGF, you didn't have enough space for civil society to be participating in the discussion, even the Ministry of Education. And education is being seriously affected by the macroeconomic policies. I would like to know how much space we'll have in the discussions because even now when the mission came, once again civil society wasn't involved, and as far as I know the ministry as well.

Mr. Nord: At the beginning of our discussion, we talked about the PRSP, which I think in Mozambique is called the Parpa, which is the national strategy. That is the level at which participation is critical. And of course, that's very much a national process. In the countries that I'm familiar with, that national process is built up from the bottom, through the grassroots, with a lot of participation also from civil society. I hope that that's the case also in Tanzania and that you had input into the Parpa process. I think that's the critical point at which civil society can provide input.

In the end, in Mozambique, like in every country, governments make tradeoffs between education and health and infrastructure, and that's a normal process. So within every government and every cabinet of ministers, there are these tradeoffs that have to take place.

As I mentioned just now, education spending in Mozambique has increased. It has increased from 4.5% to 6.5% of GDP over the last three years. Maybe 6.5% of GDP is not enough. I'm sure the needs are enormous. And I think it's really for civil society to make those arguments powerfully to convince the Mozambique government that it needs to look at those priorities. The best influence U.S. civil society can have is in your dialog with the Mozambique government.

Ms. Mendonca: But when Parpa was designed, the microeconomic framework was established, meaning that the space for us to discuss Parpa was quite limited. I don't know if things will be changing from now, but in the past, the framework was established when civil society and even government discussed Parpa one and two.

Mr. Nord: For example, the budget deficit this year, including grants, is about 5% of GDP. It was 2.5% of GDP two years ago. So the macroeconomic framework has allowed for a larger fiscal deficit, and it has allowed for a higher level of spending, consistent with the lower debt levels that Mozambique currently faces, and consistent with more foreign assistance that Mozambique is currently getting. So it's within that envelope that priorities need to be set. And as I said before, I think U.S. civil society needs to make powerful arguments on where you think those priorities are. In the end, it's the government of Mozambique that has to make the decisions where to allocate scarce resources. That's no different in Mozambique than it is in any other country.

Mr. Gutierrez: I just wanted to clarify the point made earlier by Patricia on the signaling function in PSI. Can you explain a bit more on this separation of (garbled).

Ms. Alonso-Gamo: One important aspect, when we consulted with countries and donors, is that everybody wanted to have more nuanced signals, and that's an important element. We don't have a review. We give a nuanced assessment of the different elements that are going well or going less well. And then, it is up to each donor to make up their minds as to whether they want to go ahead or not.

That's important, for instance, because just because reforms in the energy sector are not going well, it doesn't mean that donors who are working on the health sector shouldn't continue disbursing. If you read the PSI document, there is a very strong attempt to give this very nuanced portrayal of how reforms are going in the different areas. So I think people should not see this as "thumbs up" or "thumbs down." Now, it's true that so far, all PSI reviews for all PSI countries have always been concluded, but in some cases, they've been with waivers because things that were not (garbled), and it's up to the donors. What the PSI is doing is it's giving many more donors the confidence to switch to larger but better support, as Roger was mentioning before, for some countries, which gives countries much more flexibility on how to spend that money.

In any case, we are, at the same time, lobbying the donors for longer-term and firmer commitments that are much more independent of any vagary of policy because we think that that's an important element of scaling up spending.

I would recommend that you look at our papers that we put out in July about scaling up and support for scaling up in some programs-both the ones that were written by our Fiscal Affairs Department and the ones that were written by the Policy Development and Review Department-because you will see that some programs really have switched to much more accommodating stances. The new policy is established in terms of adjusting and accommodating aid. So I think that you should see these as general trends to more accommodation. Also, there's very nuanced signaling because we think that it's important not to have ups and downs and donor differences.

Mr. Shastry: Thank you very much. We should continue this conversation. We look forward to seeing you in the next spring meeting.