Where are the Poor?

Experiences with the Development and Use of Poverty Maps

Norbert Henninger, Mathilde Snel
Download complete report
[PDF file 7.5 MB]

Executive Summary

Finding ways to reduce poverty and inequity is a daunting challenge for local, national, and international decision-makers. One important aspect of this challenge is the spatial heterogeneity of poverty: poor people tend to be clustered in specific places. Aggregated, national-level poverty data mask this subnational variation. For example, the figure below presents maps of poverty data from Ecuador at increasing levels of resolution, from national to regional, provincial, and municipal. These maps show that even the lowest-poverty region of the country (with a poverty rate of 30-45%) contains provinces with considerably greater incidence of poverty (45-60%) as well as municipalities with extremely high poverty rates (60-75%).

Poverty mapping -- the spatial representation and analysis of indicators of human wellbeing and poverty -- is becoming an increasingly important instrument for investigating and discussing social, economic, and environmental problems. Decision-makers need information tools such as poverty maps to help them identify areas where development lags and where investments in infrastructure and services could have the greatest impact. Once largely the domain of economists and social scientists, poverty maps are now being used by policymakers and many non-governmental entities, including civil society groups, academic institutions, and private businesses. However, the new and diverse applications of poverty mapping emerging over the past five years have not been well documented.The World Resources Institute (WRI) in collaboration with UNEP/GRID-Arendal has conducted a study examining the uses and impacts of poverty maps. Our interest in this topic grows out of extensive experience in mapping biophysical indicators, including those related to coral reefs, water resources, frontier forests, and drylands. We have found indicator maps to be a powerful tool for stimulating and advancing policy dialogue. In particular, poverty mapping provides a means for integrating biophysical information with socioeconomic indicators to provide a more systematic and analytical picture of human wellbeing and equity. Environmental factors represent one dimension of the complex physical, biological, and socioeconomic system that influences human welfare and poverty.Poverty mapping -- including methods for producing maps as well as ways of using them to influence policy and expenditures -- is an evolving discipline. Specific objectives of this study are to:

• document the uses and impacts of poverty mapping;
• increase understanding of how national and subnational decision-making can benefit from incorporation of poverty maps;
• encourage innovative applications of poverty mapping in new areas such as environmental management;
• foster expansion of the community of poverty mapping users and stakeholders, including non-governmental and civil society organizations, universities, the media, and the private sector;
• support more widespread use of mapping in developing countries, including development of human and technical capacity to create and interpret poverty maps; and
• encourage greater coordination and collaboration between map producers and users.

Sources: Boundary files from Centro Internacional de Agricultura (CIAT). Poverty estimates from Hentschel et al. 2000.

High-resolution maps can help uncover poor areas that might otherwise go undetected. Shown here are poverty maps for Ecuador at increasing levels of resolution, from national to regional, provincial, and municipios (districts). Higher-resolution maps reveal that the low-poverty region (the Andean region, shown in pale yellow, with a poverty rate of 30-45%) contains several provinces with a considerably higher incidence of poverty (45-60%, shown in dark yellow). One such province, shown at highest resolution, encompasses several municipios (districts) with extremely high poverty levels (60-75%, shown in orange) alongside areas of moderate and low poverty.

The intended audience for this report is decision-makers in international and national development agencies with a poverty reduction mandate, along with the growing network of practitioners using various techniques to produce poverty maps. In addition, we hope it will be useful for a broader audience -- including universities, non-governmental organizations, and the private sector -- interested in learning more about experiences with and potential new applications of poverty mapping.To conduct a systematic examination of the application and use of poverty maps, we used a country case study approach, featuring a series of telephone and in-person interviews with map producers and users in selected countries. The cases considered here encompass 14 countries from Africa (Burkina Faso, Madagascar, Malawi, Mozambique, and South Africa), Asia (Cambodia and Vietnam), and Latin America (Brazil, Ecuador, Guatemala, Jamaica, Nicaragua, Panama, and Peru). These examples reflect a variety of poverty mapping methodologies, in most cases focusing on more statistically rigorous methods, known as small area estimation. However, we also examined other approaches with a longer history of use, such as the Human Development Index pioneered by the United Nations Development Programme and various “unsatisfied basic needs” indexes used primarily in Latin America.Highlights of the uses and impacts of poverty mapping include:

• Poverty maps have become important tools in implementing poverty reduction programs, including international efforts (such as the World Bank-initiated poverty reduction strategies for Highly Indebted Poor Countries) as well as purely national initiatives. One such example comes from Nicaragua, whose poverty reduction strategy relies heavily on poverty maps to allocate US$1.1 billion in capital spending over five years (2001-5).
• Poverty maps help improve targeting of public expenditures by identifying where the neediest populations are located. For instance, in Guatemala, poverty mapping is being used to restructure the National Public Investment System to improve geographic targeting of hundreds of millions of dollars (US$576 million in FY 2002) of annual expenditure.
• Emergency response and food aid programs are beginning to make use of newer, more data-intensive mapping methods. In South Africa, information from a poverty mapping initiative was combined with data on sanitation and safe water supplies to create a geo-referenced strategy for containing a cholera outbreak in KwaZulu Natal province in early 2001. Implementation of this strategy effectively contained the disease in three months, with one of the lowest fatality rates (0.22%) ever recorded. Cambodian poverty maps are being used to identify the poorest communities for distribution of US$50 million (2001-03) in World Food Program food aid, especially “food for work” interventions.
• In several countries, high-resolution poverty maps are contributing to state- and local-level decision-making. Brazil's largest state, Minas Gerais, is using poverty maps to redistribute statewide tax revenues totaling US$1 billion annually toward poorer municipalities that are making an effort to invest in health, education, sanitation, and environmental conservation.
• In the cases studied, the production and distribution of poverty maps resulted in increased transparency of public decision-making, by raising awareness of poverty, igniting policy debates at local and national levels, and encouraging broader civil society participation in decision-making. One such instance was reported from Panama, where officials of the Social Investment Fund indicated that the use of poverty maps in decision-making helped them resist pressure from politicians to alter funding allocations once they had been made.

Four central themes emerge from the case studies: issues related to user demand; methodological issues; the importance of dialogue, collaboration, and dissemination; and capacity development and long-term sustainability.

• Our case studies indicate that, close linkage of map production with user demand promotes use and impact. While most cases address financial distribution issues, map producers and users point to untapped potential for environmental and other applications. Potential for private-sector use of poverty maps appears to be great, though relatively few such applications were documented.
• With respect to methodology, the use of a credible, transparent method was identified as an important factor in the eventual acceptance and impact of a poverty map. We found that map producers often faced challenges regarding access to data, which some successfully addressed through collaborative arrangements with national statistics departments and other relevant agencies. Producing poverty maps at high levels of resolution using up-todate information also promotes extensive use and impacts.
• The cases we examined indicate that collaborative interagency approaches supported by senior officials greatly enhanced map production and use. We also found that active, strategic dissemination can have a powerful impact on awareness and use of poverty maps.
• Among the most significant impacts of poverty mapping have been its effects on institutional strengthening in developing countries. In addition to the expected rise in the technical skills of participating researchers, case studies contained numerous reports of increased institutional credibility and standing in policymaking circles in the wake of a poverty mapping exercise.

Based on these findings, we recommend that seven major steps be taken to ensure that poverty mapping is sustained in the countries studied and expanded to all developing and developed countries over the next 10 years.

1. Recognizing the need for universal application. Every country in the world should map the distribution of its poor. Policies and programs to reduce poverty require that countries have solid, detailed information about where poor people live and the resources the poor can use to improve their lives.
2. Involving stakeholders from the very start. National governments, through a multi-institutional, multi-stakeholder approach (and with the support of the international community, when needed), should conduct a preliminary poverty mapping needs assessment that identifies detailed steps to the development of a useful and accurate map of the poor. These steps should address: purposes and intended uses of poverty maps; agencies and universities to be involved in the mapping effort; methodological approaches and datasets to be used; technical support required; data quality and accuracy of maps required; budget and funding sources; target audiences and outreach strategies; and evaluation and monitoring of uses and user feedback.
3. Sustaining the effort. It is essential that governments take steps to cultivate political support to sustain the technical effort over decades. These actions typically involve: committing technical and financial resources; providing training to key staff; developing a cadre of broadly trained analysts; creating incentives to retain skilled analysts in the public sector; securing appropriate equipment; conducting a high quality census and periodic national household survey to generate essential data; developing data and map standards; providing support to universities and research centers that can assist with map development and rigorous evaluation of map uses and impacts; and committing to more transparent decision-making that takes into account both technical (i.e., poverty maps) and political considerations.
4. Ensuring access to data. Most countries will need to ensure that national legislation is in place to support the distribution of poverty information to all those who want it. This means all potential users in: various levels of government, research institutes and universities, the private sector, nongovernmental organizations, media, and the poor themselves. Widespread access to poverty-relevant data will facilitate broad use of poverty maps and ensure strong demand for poverty information and more accountable decisionmaking. It will open the mapping process to independent organizations, helping to counterbalance any mistrust of official government information.
5. Encouraging innovative uses. This report identifies and documents numerous ways in which poverty maps can and are being used to make decisions; support social, economic, and environmental goals; and strengthen governance. Much more remains to be done, however, to ensure that poverty mapping is applied as broadly as possible. In the area of environmental quality and natural resource management, there are numerous opportunities to integrate poverty mapping into natural resource management and pollution control. For example, poverty mapping can be used to analyze the relationships between poverty and transportation, industrial hazards, exposure to air and water pollution, access to natural resources (wildlife, forests, grasslands, coastal and mineral resources), and natural hazards (flooding, storms, drought, and climate change). The interconnections between poverty and land and resource tenure, siting of protected areas, subsidies for food and agriculture, and a host of municipal services should also be assessed.
6. Supporting research on methods. The World Bank has taken the lead in providing technical and statistical support to a number of countries; other donors should step forward to support poverty mapping research and applications as well. The international community should take up the challenge of creating a new cadre of in-country experts who both understand the methodologies and can work with national and international colleagues to carry out studies and conduct research on new methods.
7. Developing a poverty mapping strategy. Enabling countries to develop and maintain poverty maps and make them widely available will take time, strong technical and institutional support, and leveraged funding. However, it is essential that countries that have been successful in this endeavor -- along with international donors and interested nongovernmental organizations -- develop a long-term strategy to provide less developed countries with needed support. WRI and UNEP/GRID-Arendal are ready to join with the World Bank, developing countries, and interested organizations to develop a strategy that brings the promise of poverty mapping to many countries through many different avenues. We welcome ideas and participation.