Interview with Brian Levy and Mark Sundberg on Monitoring the Millennium Development Goals

The scope of coverage in the Global Monitoring Report is truly global. It is looking at performance across developing countries, which includes both low-income as well as middle-income countries, and this certainly includes war-torn countries. However, having said that, it is very difficult to get information on the performance of the war-torn countries, certainly not up-to-date information. In fact, across low-income countries, only about half are reporting regularly on the statistics needed for monitoring the MDGs. So there are major information problems, but the report does aim to be global. It also looks at performance in various dimensions of rich countries, for example whether they are meeting their obligations to provide aid, debt relief, or to provide market access. Regarding the specific question on Palestinian occupied territories, there is a publication that the bank has just released, the 2006 update of the World Development Indicators, that provides quite a comprehensive reporting on statistics, including the MDGs, and in this publication there is an entry for West Bank and Gaza. It reports on what data we have available on a comparable format, although I am not sure about how comprehensive the data is in its coverage or how up to date it is.

This is an excellent question concerning project impact evaluation. It is related to efforts by the international financial institutions and by countries themselves to reorient their management to focus on the results of project and policies interventions, to focus on outcomes rather than inputs. To do proper impact evaluation of programs in a country requires a lot of information about what is happening on the ground, it requires statistics and survey information that allows rigorous evaluation of impact. In many countries the information is quite weak and it is difficult to undertake. Where we do have information, greater effort is being put in place to better understand whether a given project or intervention is delivering on its targeted objectives. This information can be used to provide feedback to policy makers, or indeed to the institutions that help design projects or provide policy advice. This is important for learning from experience, which has not received adequate attention in the past, and which going forward requires more resources and effort – both to strengthen statistical capacity as well as to mainstream periodic impact evaluation.

This is a really interesting question on what are the most effective ways of getting resources directly to the front line where they actually can address poverty and the GMR answers this in a way that tries to distinguish among different kinds of country starting points. The questioner highlights the dilemma -- " more than twenty cents in a dollar reaches the ultimate beneficiaries after accounting for the transmission costs" -- it is a very interesting piece of work that is referenced many places and cross-referenced in the GMR on Uganda on public expenditure tracking, what resources reached the front line, and in the beginning it did show that number, but the interesting thing is that the Ugandan government then sent resources down to the local and posted in community centers that the resources and amount of resources made available and quite rapidly that the amount went from 20 cents on the dollar to 90 cents on the dollar. So this signals a number of different messages that are inside the Global Monitoring Report. The first is global transparency in terms of getting resources used effectively. The second one is when the institutional arrangement functions effectively. Then money can in fact pass through public systems and be used for poverty reduction, and indeed one of the key messages of the GMR is that strengthening public systems is key to scaling up the resource transfers because they are a cost-effective way if they work well. The third message in the GMR, which this underscores, is that in those settings where the public systems or resource transfers do not work effectively, there can be real advantage in working in bottom-up service delivery approaches, for example, providing grants directly down to the lowest tier of local government's on some sort of formula basis or activities and this is a what the World Bank supports and this is a good question indeed in terms of highlighting there are many different ways of transferring resources to the local community and which way works depends on local resources.

This is a very good question which is relevant not only to India, the specific example that Naveen raises, but also to middle income countries and indeed rich countries. Across any nation there are going to be regions that lead and lag and often times in middle income or in wealthier countries, there will be regions where there is intense poverty and poor performance that is not captured in national data that reports averages. If you take the United States, the Appalachian region is known to lag. If you look at European countries, the coal-mining regions and rust belt areas lag. In middle-income countries, Mexico or Brazil for example, the southern states of Mexico or the Northeast of Brazil have had long entrenched poverty and weaker performance on social indicators. The case of India is one that speaks to me in particular because I worked on India for some years, including on India’s poorest province of Bihar. Bihar once comprised both present day Bihar and Jharkhand, and has not experienced the national trend toward improved literacy, immunization coverage, or progress reducing poverty. For these lagging regions it is critical to focus on specific policies and interventions that will reach them, help integrate them with national markets, and overcome key constraints to development – including, as is often the case, weak governance.

One of the important themes of this year's report is how to monitor governments and this is a particularly difficult area. The report identifies governance as having three parts, ability of the bureaucracy, the quality of checks and balances institutions like the courts, media systems, the quality of independent audits, and the broadest civic participation. And the GMR identifies a set of 14 core governance indicators which can be used to monitor across these different dimensions and more broadly how countries perform with respect to governance. Some indicators will focus on corruption, others will focus on public services, other indicators will focus on vase and accountability and the quality of law. Now, with those indicators as backdrop, a key message of the VMR is that while we know what a well-functioning governance system looks like, and we know what a poorly functioning system would look like, there is no unique part of moving from a dysfunctional system to a well functioning system. Some countries improve in one dimension, others improve in checks and balances. In the long run all the different elements of the system need to be working effectively for that system to be sustainable. Now, if you take the example of a country, one can think of Uganda from the 1980s to the 1990s, or one can think of Indonesia in the 1970s and 1980s, those two examples are examples of countries which made major gains in the quality of there bureaucracies and alongside, they both showed tremendous gains in terms of the reduction of poverty, and that was one of the goals we care most about. The difficulty which becomes very evident in the example of Indonesia, by the early to mid to mid 1990s, the wheels are coming off the governance train, there is a financial crisis, there was a political crisis that followed and the re seem of president Suharto fell and I think it is clear in retrospect, using the example of Indonesia, that the bank would have done well to have paid close attention not only to bureaucratic quality but also to the trends and with respect to voice and accountability and the rule of law. I think without getting specific with respect to any other contemporaneous situations because those are judgments that aid donors in general and World Bank in particular are made by the operations staff, I think the message that is underscored in the global monitoring report, one ought to monitor for countries whose governance is stronger than others. One of the expectations is will it improve, are those expectations being melt, and in the event that the data is showing that there was an agreement that there is a need for improvement, one needs to think through its implications. One last point, as it turns out, the World Bank's own formula for allocating aid is driven very heavily by performance on governance. The series of governance, all of which will be released this July for 2005 are used in deciding on the aid allowance for country's and depending on a country's governance performance, aid allocations vary down as $3 per capita to the well governed countries, up to $15 per capita. So clearly governance is something we take seriously. It is part of the story in the GMR and it is already the case that it is shaped by good country performance.

The question about African performance can be answered on different levels and as Milton's question makes clear, performance across Africa as a region has been very sobering. Africa is not on track to meet any of the Millennium Development Goals as a region.

However, within Sub-Saharan Africa, there is quite varied performance and there have been some countries that have shown quite strong growth, with rising per capita incomes, while others have experienced stagnant or declining growth. Country performance has been quite varied. The Global Monitoring Report shows that trends in income growth in recent years have been very encouraging in Sub-Saharan Africa, and there is evidence of poverty reduction in some African countries, not only in the more rapidly growing regions of East Asia and South Asia.

There have also been several countries in Africa for whom recent survey data reveals quite remarkable progress in the human development indicators. Madagascar, since 1997, has experienced a 6 percent annual decline in child mortality. That is a very rapid rate that is very encouraging. South Africa has been able to reduce the incidence of measles by 90 percent since 2000. Mozambique is another example where the poor are clearly benefiting from human development efforts. Immunization coverage has improved by 18 percent annually between 1997 and 2003 among the poorest households in the country, which is four times as fast as the national average. So it is true that regionally Africa faces the most daunting challenges to meet the MDGs. However, it is a diverse region, and many countries are achieving remarkable progress. We need to learn more from those examples and try to replicate this positive experience elsewhere.

This is a great question. One of the really interesting paradoxes that emerged in this year's GMR -- the paradox is that not all good things come together when it comes to poverty reduction and governance. We looked at countries, the quality of policies and the extent of corruption, and while in many countries it was clear that bad policies and high corruption went together, there were a group of countries that were quite mixed, countries where, for example, you had high corruption but good policies, and other countries where you had low corruption and bad policies, and as it turns out in that group was Bangladesh, which is among the world's most corrupt but, in terms of the quality of its policies, turns out to rate relatively high.

So you look at a country like Bangladesh in terms of meeting the Millennium Development Goals, Bangladesh has done well, and in fact is one of the world's leaders in terms of change, universal education, the goal of access to water points, and the Millennium Development Goals of infant mortality, child mortality under the age of five.

Now, looking at Bangladesh as a whole, the approach from the GMR would suggest that the long- term sustainability of Bangladesh's gains will depend on a much more balanced improvement in governance, where it is not only policies that are good, but the checks and balances that limit corruption increasingly are strengthened over time, but it is one of the important and complex messages of the GMR that policy-makers are going to have to come to grips with, that in the short term good things and bad things sometimes come together.

This question raises a very serious problem that many African countries face with inadequate skilled labor to expand public services, not only in health, but in education, in strengthening infrastructure and delivery of water and sanitation. All of these require skilled labor, as well as in government, public administration and management. It is estimated that to meet the millennium targets in health, there will need to be roughly a tripling of health personnel, skilled labor in the health industry, in Africa, over the next ten years. Where will that labor come from, and in particular, how will it be retained if working conditions and remuneration are weak and out-migration has been such a problem. This was one issue raised in last year’s Global Monitoring Report which discusses the need to both improve incentives to keep skilled labor in place, and measures to halt immigration.

One example the GMR provides is of Malawi, I believe, where of the 600 doctors that have been trained since independence, only 50 are estimated to remain in the country. This is a very grave issue. Reducing or halting out-migration and improving incentives must be part of the answer. Looking forward, there will need to be an increase the training capacity to expand the supply of skilled labor which requires educational strategies that deal not only with MDG of 100 percent primary completion, but also secondary and tertiary education.

One of the key messages of this year's report that I would like to highlight since it is relevant to this issue is the quality of external aid. The majority of aid is provided in forms that are not flexible enough to be used to meet the needs of salaries and other recurrent costs required to expand public service delivery. There is an important agenda in improving aid quality; making it more predictable, allocating it where it is needed and will be well used, and providing aid in forms which can be used help expand core public services, in order to pay for salaries of health care workers and teachers, and for other recurrent costs. That must be part of the developed countries' and the international financial institutions' agenda moving forward.

This question is one of a number of questions which asked about the role, potential role, of civil society organizations in relation to achieving the MDGs, and I am going to answer in a way that is relevant to many of these different questions.

Often there is a perception of powerlessness in relation to civil society organizations and individual citizens with respect to the MDGs, and may be a perception of powerlessness with regard to well intentioned people, like Mark Sundberg and myself, that describe all nice things but all we are doing is talking about them, and that is not where the action is.

This is one of the themes that emerge in this report, the role of civil society participation is central to the development process, and indeed citizens are anything but powerless. Just to give you an example of this, there was an award given today at the World Bank to a man named Samuel Poore and an organization called the Public Affairs Center in India. The center, its core work is to develop citizen report cards, essentially monitoring the quality of public services in the city of Bangalore, across a dozen services, how citizens perceive the services. This was done in 1994, 1999 and 2003, and what was quite remarkable was the degree to which the process of generating these data and putting these results into the public domain produced over the course of a decade a strong political response aimed at improving the quality of these organizations that were being monitored to the extent that one say major, major shifts, up to 80 percent satisfaction over the course of a decade, and signaling that indeed a great deal of power and authority lies in civil society organizations, indeed this particular organization is spread across several cities and several countries around the world.

I think there is an important lesson, it is citizens that hold governments accountable, but the other lesson which I think is very useful in terms of asking what should an individual's civil society organizations do, if an organization is trying to achieve results that are evident, a particularly effective way to do this is to focus in a very precise way around a single objective. For example, there are civil society organizations, there is an organization in the UK which works globally with many organizations around the world on what would be priority ways of improving access to water on the part of local citizens and this organization, it goes all the way from the kind of grassroots involvement to highly sophisticated and monitoring and prioritization of how resources should be allocated in terms of water reduction.

So I think these two features, the power of civil society as a source of accountability because accountability is the key to good governance as the GMR highlights and the potential civil society organizations have to focus to make a difference and to enjoy the gratification and to monitor and see that over the course of a significant period of time they can see that the areas they have targeted are indeed improving.

The question relating to trade is an important one because of the role trade has had in generating growth opportunities for many developing countries, and indeed, it is right that there has been very little progress made in 2005. The results from the Hong Kong ministerial were very disappointing. The report highlights the importance of trade for growth, and points to two developments in recent years that are both encouraging and discouraging. In the absence of multilateral trade liberalization, there has been a proliferation of regional trade agreements. In the last 15 years, there have been roughly 250 regional trade agreements that have been signed covering a third or more of global trade. The net impact of these is difficult to determine, but they tend to under-perform. Whereas they may help generate trade for some countries that are members of these regional agreements, they also divert trade from trading relations outside the regional group. The net impact is often quite disappointing.

But another issue that is raised in the Global Monitoring Report is about the pledges that have been made and the attention being given to aid-for-trade. Multilateral trade liberalization will not affect all countries equally and for some countries that have weak capacity to respond, it is likely to have a negative impact in the short run until they are able to strengthen trade performance. There has been progress in the last year to pledge more resources, from the U.S., Japan, and the European community, for example, for aid-for-trade. Resources are to be made available to countries with a weaker trade capacity, in order to strengthen trade infrastructure, to improve the investment climate, or marketing capacity, to help them improve their trade response over time. But the bottom line remains that progress with multilateral liberalization has been very week and much more needs to be done if the Doha Development Agenda is going to deliver on its promises. Prospects still appear weak.

The criticism or concern that the MDG goals are overly optimistic has been raised by others as well. It is true that if you look historically, few developed countries have been able to achieve the rates of progress that are implicit in the goals for developing countries to reach the MDGs. However, I don’t think this devalues the importance of the MDGs for a few reasons. First, I think that it is important to have goals that are ambitious in order to push to reach higher standards. These goals represent important and pressing objectives for the international community related to human development and poverty alleviation.

Secondly we have new technologies—new ways of achieving progress and new ways of internationally delivering assistance that make historical rates of progress less relevant. For many countries I don't think these goals are too optimistic and in fact many countries are on target for reaching them. In some regions, and I would point to Sub-Saharan Africa and South Asia it is unlikely that many of the countries will achieve all the goals, but they will make progress toward them, and focusing on the goals is important for monitoring progress and for identifying where additional effort is required.

Indeed, it is the case, as we know, that many African politicians have stolen and kept huge amounts of money, and indeed this money tends to be kept in bank accounts in Switzerland or elsewhere, and this mix of facts signals that corruption is not only an issue for developing countries, but it is a global issue. Corruption of course has--in the process of bribe taking, there is a bribe taker and a bribe payer and, often the bribe payers are from the private sector in developed countries. For example, until not very long ago, in a significant number of industrialized countries, bribe payments were viewed as a cost of doing business that was perceived to be tax deductible. I believe it is still the case that paying bribes is tax deductible in New Jersey, for example. I think that is a signal of the global side of this. Similarly, it turns out that it was not only the Swiss banks that were the receptors, but Sani Abacha, the president of Nigeria, moved his resources to Switzerland, was through British banks as a means of being laundered. So all of this is a clear signal of the way that corruption is a global issue.

That is the bad news. One piece of better news is that some hundreds of millions of dollars of Abacha's resources have in fact been repatriated to Nigeria. This was with the cooperation of the officials in Switzerland and the World Bank staff. Another part of the better news is that the United Nations has a recent convention on corruption, it was promulgated in 2005, it has a significant number of countries that are signatories. This is not a convention that you sign on the dotted line and then you are done with it. It involves ongoing implementation and monitoring of that process. Indeed, what the Global Monitoring Report highlights this year is it highlights a whole series of emerging new initiatives to begin to address corruption as a global problem, not simply the United Nations Convention on corruption as one example, but there are a whole series of anti-money laundering -- and the extracted industries transparencies which brings together mining countries and donor countries who agree on codes of conduct for the reporting of revenues from extractive industries, the flow of those revenues made publicly available, and how those revenues are used. Now, none of this is to say that there now exists a robust global system for controlling corruption, but it does signal that there is increasing movement in the global arena, to control global checks and balances, and the Global Monitoring Report highlights this, and it will continue to do so as it proceeds further.

This is a question of the quality of services delivered. Indeed if poor quality services are delivered, not only do outcomes suffer in terms of what is actually being achieved, but also there is not adequate demand for services uptake. Parents don't want to send their children to schools that offer poor quality education services or clinics that lack properly trained nurses or adequate drugs.

This is an issue that has come up in discussing how to meet the MDGs in Ethiopia. It is appropriate to think of the trade-offs between quantity and quality, and to consider how to use scarce resources to both expand service delivery, but at the same time bear in mind quality concerns. Ethiopia requires on the order of 100,000 teachers to meet the primary education completion goal by 2015. Should they be aiming for the standards that are set by the Education For All fast track initiative of, I believe, 40 students per teacher, or is a more relaxed target appropriate? Should they be aiming for teachers that have high school plus two years of vocational training as a minimum, or could they take teachers on board that have perhaps secondary training alone or even less than secondary completion?

There have been arguments made that having very high quality standards may also be inappropriate, and it relates to an earlier question about retaining labor. If teachers are going to go into rural areas and remain there, recruiting people from that locality and perhaps relaxing standards may be part of the answer to retaining labor and ensuring adequate supply over the short term, but, of course, quality is critical to maintain over the longer run.

One last point to make is on the question of measurement and assessing quality over time. Our ability to do that is very limited now. We may be able to count the number of students enrolled and those who complete school, but the quality of that education is very difficult to measure. This argues for some effort to build quality indicators, to introduce standardized tests that enable one to consider basic literacy or mathematical capabilities, and which could allow tracking the quality educational services over time. That I think needs to be brought into the MDG agenda moving forward.

Poor coordination of international assistance is a problem not just for the international financial institutions, the UN, the IMF, the World Bank, and other IFIs, but also for bilateral donors. It is something which undermines aid effectiveness and which constrains the usefulness of support to developing countries.

Having said that, there are some encouraging initiatives to help strengthen collaboration between the Bretton-Woods institutions and the UN, and let me mention three of those. The first, which I think is something of a breakthrough in 2005 was the Paris Declaration on Aid Harmonization and Alignment, signed in March of last year. This is an effort to recognize that much better coordination is needed across donors as well as cross the multilateral development banks, in conjunction with developing countries themselves, to ensure aid is better aligned to country strategies, that there is not duplication and fragmentation of assistance. There have been specific benchmarks to measure performance, 12 different indicators that are to be monitored over time against benchmarks, and reported on annually. I think this is an important effort to strengthen aid harmonization and to avoid the problems of weak coordination of assistance programs.

Secondly, there is an initiative to bring together the World Bank and the OECD-DAC, which is the Development Assistance Committee of richer donor countries, together with the United Nations development program, to work on issues of scaling up aid, and to consider what that will require in terms of delivering aid and working with governments in this effort.

The third initiative I would mention are results and resource round tables, alongside consultative group meetings, that provide a critical convening authority across organizations, donors, and the government, to coordinate development assistance. This process needs to be country-led, and the programs country-owned, for successful coordination and implementation of Round Table results.