Click here for search results

WDR 2004: Video Conference Consultations

World Development Report Archives

World Development Report 2004

(February 17 – 19, 2003)

The World Bank held three video-conferences on February 17, 18, and 19, 2003 as part of the consultation efforts for the World Development Report (WDR) 2004 on “Making Services Work for Poor People”. Representatives from civil society, government, private-sector, and donor agencies from fourteen countries participated in the 3-hour sessions. (see participants list attached). These consultations, which will help shape the World Bank’s flagship publication this year, were hosted by the WDR 04 Team and Civil Society Team (CST) at World Bank headquarters in Washington and co-facilitated with the Global Development Learning Network (GDLN) of the World Bank Institute (WBI). The purpose was to share an early draft overview of the report and get feedback from colleagues around the world.

The first consultation on February 17 included stakeholders in Cambodia, Indonesia, Philippines, and Vietnam. WDR Team Director Shantayanan Devarajan presented, and Carolyn Reynolds of the Civil Society team moderated. The second video conference was held on February 18 and included Jordan, Nigeria, Tanzania, Zambia, and Sweden. WDR Team Co-Director Ritva Reinikka presented, and John Garrison of the CST moderated. The third conference held on February 19, was with Peru, Russia, Ukraine and Venezuela. Shanta presented for the second time and John Garrison moderated. Veit Burger, of the GDLN was the master of ceremonies for all three discussions as well as the technology coordinator. The success of these conferences highlighted the value of distance-learning technology for consultative purposes. Not only were the WDR Team co-directors able to dialogue with a diverse group of stakeholders from more than a dozen countries in different parts of the world in a short period of time, but these meetings took place when Washington was virtually shut down due to one of the worst snow storms in recent history.


Each videoconference opened with an introduction by the moderator, followed by a 30-minute presentation by a WDR director. After the presentation, there were 2-hour discussions with the participants.

Introduction Digest - John Garrison and Carolyn Reynolds

Every year the World Bank produces a World Development Report and each year the report addresses a different development problem. The WDR is not an operational document, it is a study put together by a team of World Bank specialists and it reflects the team’s best research and thinking. While not all of the conclusions or proposals in the study will necessarily be made operational, the reports do influence and help guide the World Bank’s work. This year, a high level of outside consultation has been part of the WDR process. The team has been consulting with a variety of stakeholders from the onset of the study, when it first developed the hypothesis and outline for the study. There have been meetings in Washington, in Japan, Egypt, Uganda, and Bangladesh as well as other countries. While these discussions have predominantly been with civil society representatives, they have also included government officials, academics, and private sector consultants. This effort reflects a World Bank-wide commitment to consulting more widely in its policies and studies, as well as encouraging greater civil society participation in Bank-financed development activities throughout the world.

At present, the WDR team is finalizing a draft document and the February consultations will feed into that draft. The timeline for the report is as follows:

Mid-March: WDR draft posted on website
Mid-April - Mid-May: Web-based discussion of draft report.
Early June: Final report finalized
September: Final report published
December: Development Marketplace competition linked to the

Presentation Digest - Shantayanan Devarajan and Ritva Reinikka
(See attached power point presentation for more details)

Main Messages
The main messages of the report so far are these: Services are failing poor people; they’re failing them in access, in quantity and in quality. But we can make services work for poor people and the way to make them work is first to put poor people at the center of service delivery and then 1) improve their ability to monitor and discipline service providers 2) amplify their voice in policy making and the political process, and 3) increase the incentives that service providers have to serve them better

How do we know services are failing poor people
We can look at the data and at the progress toward the Millennium Development Goals. The world is largely on track in terms of achieving the income-related MDG, but is off track on the universal primary education goal, on the gender equality goal, and on child mortality. Health and education will thus require greater attention and an accelerated effort to achieve the goals by 2015.

How are services failing poor people
We have several observations:
The first is that services are failing poor people in terms of public spending. Most of these services--health, education, water, sanitation, electricity and transport—are either financed, regulated or provided by the public sector and yet the amount the governments spend on these services accrues mostly to rich people not to poor people. For example, in a very poor country like Guinea, over 48 percent of spending on health goes to the richest quintile and less than 8 percent goes to poorest quintile. In fact, in very few low and middle income countries does the amount going to the poorest quintile exceed the amount going to the richest quintile. As a result, the health and education outcomes of the rich and the poor in these countries are vastly different.

Furthermore, services are failing poor people in terms of access. For example, looking at the relative distances to schools or clinics for poor and rich children in rural areas, the ratio of 10 to 1 is not atypical. Similarly, access to drinking water varies substantially between rich and poor people in a whole variety of countries. In a country like Morocco for example, almost 100 percent of people in the richest quintile have access to a drinking water source, whereas in the poorest quintile, less than 10 percent do.

When thinking about how to address these problems it could reasonably be assumed that they could be dealt with by spending more or reallocating spending. Both of these may help. But then there is another problem, which is that money allocated for these services often doesn’t reach the front line service provider.

We did a study in Uganda about seven years ago where we asked the question: for every dollar of public, non-wage, recurrent spending that is meant for primary education, how much of it reaches the primary school, and the answer was 13 cents on the dollar, and the wealthier schools got the greater amount of these funds. Now the Ugandans have been able to solve that problem, they have been able to raise that number from 13 cents to 80 cents on the dollar.

But even if you could solve that problem, the problem of money reaching the front line service provider, you also have the problem that the quality of services is often very low for poor people. One classic example of this is the high levels of teacher and health worker absenteeism in poor areas. In Bangladesh, we just did a survey that showed the absenteeism rate in primary health care centers is about 79 percent, and this is not atypical.

Not all the news, however, is bad, because there have been innovations and developments that have allowed services to work for poor people, even in the countries where we have been failing in other areas. And one of the things that we are trying to do in this WDR is to present a framework of service delivery that will enable us to understand both the failures and the successes and also how we can move forward, how we can actually make services work for poor people.

The framework is very simple. It starts with the relationship between clients and service providers. And it starts with a very simple observation: In a competitive market transaction, for example, buying shoes or vegetables, you don’t hear about a service delivery problem, because the consumer or the client is able to hold the provider accountable for those services. The client provides money and uses that money as a way of holding the provider accountable; also the transaction is not complex or difficult to monitor. If the client gets the shoes, he knows he’s gotten shoes and not something else. If he doesn’t get shoes or gets bad shoes there are lots of ways he can sanction the provider: he can refuse to pay, he can refuse to give him any repeat business, or there may be legal recourse.

For several important reasons, very good reasons, societies have decided not to provide services like health, education, water, sanitation and so forth in competitive market transactions, so these services don’t have the automatic accountability mechanisms between clients and providers. What they have instead is a long route of accountability, which is that the client, as a citizen, influences the policy maker, and then the policy maker influences the service provider. But that’s the problem.That long route of accountability has two links and both links need to be working for the system to work. The citizen-policy maker link needs to function properly and the policy maker-provider link likewise needs to work, and both of those links are prone to have problems associated with them.

Citizen-policy maker link: Politics, or the political economy, is the main problem with the citizen-policy maker link. Free or subsidized public services can often become the currency of political patronage. Politicians use these services as a way of currying favor with their constituents rather than actually promoting the welfare of poor people. Furthermore the problems and the advantages with the political economy of service delivery appear to be somewhat independent of whether or not the country is a democracy. For example, Cuba has the best health and education indicators in Latin America. At the same time India—a well-functioning democracy—has a very poor record of health and education outcomes for poor people.

Policy maker-provider link: Even if the policy-maker cares (or can be made to care) about health and education for poor people, there is still the problem that the politician cannot provide the service. He has to make sure that there is a service provider. But this relationship--between the policy maker and the provider--can break down.

The workability of this policy maker-provider relationship often has to do with whether or not the service can be monitored and whether the provider faces incentives to actually serve poor people. One way policy makers have been able to improve the responsiveness of providers is to take services where the output can be monitored, where it is easy tell whether or not the service was delivered, and actually write an explicit contract with the provider. In these situations, the results can be quite impressive. (Power Point Reference Cambodia example)

But in many cases, services cannot be easily monitored or observed. You can monitor whether or not a child is enrolled in a school, but you can’t really monitor especially from a distance what exactly is going on in the classroom, you can’t even monitor whether the teacher actually shows up, which is one of the reasons why teachers tend not to show up.

The Client Role. What may have been ignored in the past is the idea that the client can actually play an active role in service provision, that the client might actually be able to monitor and discipline the provider (and in some cases be the provider). And there are actually two important ways that the client can play that role. One is if the client has a choice, if the client can actually exercise choice in the kinds of providers goes to. That creates an incentive for the provider to deliver the service and it’s important that when the client exercises choice the provider is affected by that choice. (Power Point Reference, Female Secondary School Assistance Program in Bangladesh)

But you can’t have choice everywhere and there are limits to how much choice can do. When you can’t have choice the other role that clients can play is to be active participants in the service provision process. (Power Point Reference, EDUCO program in El Salvador)

What We Can Do
With the objective of trying to improve services, there are two main things that we as a global community can do to improve services for poor people.

· We can generate information that is credible, information about what works and what doesn’t work.
· We can also disseminate information.

For instance, in the Uganda example, where only 13 cents on the dollar actually reached the primary school, when the government of Uganda found that out, they publicized it, they put it in the newspaper and this had a profound effect on the empowerment of communities.For one thing, parents started asking the principal of the school to publish the entire school budget and of course also the government had to make sure an increased portion of the spending actually reached the school.

So information has got to be one of the most important things we can do, both the creation of information and the dissemination of information to strengthen service delivery for poor people, but can we do anything more? Can we actually say something about the types of service delivery arrangements and the advantages/disadvantages of each. The answer is yes, but we should be very careful that we tailor those service delivery arrangements to the characteristics of the service and the characteristics of the country, or the service beneficiary area. In addition services arrangements have to take into account the attitudes and politics of the government. That’s why we say in the report, it’s not one size that fits all, maybe eight sizes do, or maybe eight sizes with adjustable waist bands.

Characteristics of service: The main distinctions of the service is whether it is easy to monitor or difficult to monitor. For example immunization is easy to monitor but what the doctor does in the privacy of a clinic is difficult to monitor.

Characteristics of the service beneficiary group—The main characteristics are whether the groups are heterogeneous or homogeneous

Political economy. Finally when we come back to the political economy, there are democracies and non-democracies that care about poor people and there are democracies and non-democracies that seem not to care about poor people and so we make the distinction between pro-rich politics and pro-poor politics. We don’t know exactly how to predict this, but we think the track record of countries demonstrates this. So there are different possible service delivery arrangements.

The most problematic of all the different situations regarding delivery arrangements is when you have the following combination: Difficult to monitor services (such as the doctor in the clinic), a heterogeneous population (different genders, needs etc), and a political system that doesn’t seem to care about poor people. In that situation you can’t rely on central government provision or contracting, what you want to or need to rely on is a system that strengthens as much as possible the clients hand in the service provision process.

In this situation you might want to consider demand-side subsidies, giving power to the consumer. You also might want to consider co-payment by households, and this is controversial, but what we find is that in pro-rich politics regimes, especially when services are difficult to monitor, when free service is legislated, poor people are usually left out of the service. Paradoxically, by allowing people to pay for a portion of those services, the diversion of services to the non-poor is reduced while the ability of the poor to monitor and discipline the service provider is increased.

Video Conference Discussion Notes from separate sessions:

February 17, 2003

February 18, 2003

February 19, 2003

Permanent URL for this page: