|The examples provided on this page, and the documents which support them, are only meant to be illustrative and do not indicate World Bank endorsement of any data, methodologies, or geographic boundaries.|
Poverty mapping, the spatial representation and analysis of indicators of human wellbeing and poverty within a region, is useful in a variety of ways.
Understanding poverty and its determinants
Highlighting geographic variations - Aggregate, national level indicators often hide important differences between different regions or areas. In almost all countries, these differences exist and can often be substantial.The map below provides an example of how disaggregated poverty indicators can reveal additional information to aggregate, national level, indicators. It shows how poverty in Ecuador is concentrated in the rural cantons of the Sierra, and in the north area of the Costa.
Source: Ecuador Poverty Assessment (2004)
Simultaneously displaying different dimensions of poverty and/or its determinants - In addition to revealing disaggregated information, another advantage of poverty maps is their legibility - maps are powerful tools for presenting complex information in a visual format that is easy to understand. They can summarize multiple dimensions in a simple display, something difficult to do otherwise. The disaggregation by geographic area may for example be used to simultaneously display two or more indicators - for instance by presenting poverty headcounts and location of schools or medical centers. Maps encourage visual comparison and make it easy to look for spatial trends, clusters, or other patterns.
Understanding poverty determinats - A poverty map can be used to display simultaneously the outcome of interest (income poverty, incidence of disease, school enrollment, etc.) and its determinants (school location, infrastructure, health center location, natural reseources endowment, access to input and output markets, etc.).This allows to deepen our understanding of the determinats of poverty. The spatial representation can therefore complement regression analysis to help us understand the influence of these determinants and their interaction. For example, Demombynes and Özler 2002 use data on crime and welfare in police jurisdictions in South Africa to analyze the effect of local inequality on crime.
Selecting and designing interventions
Poverty maps can be used to inform policy makers on what intervention to select (when chosing from a number of policy options) and on the details of its possible design.
Selecting interventions - Poverty maps can be used to identify areas in which development has been lagging behind, and which may therefore benefit most from additional resources - e.g. from additional infrastructure, or from transfer programs such as subsidized credit and funds for public works.
In addition, maps with information on poverty and many of its determinants can be used to identify which intervention to implement in a specific area. When deciding on a public investment program for example, information on the specific needs of various areas can help increase the effectiveness of investments – since some areas would benefit more from certain types of investments than others. It might be the case, for instance that investments in roads and transportation would be more effective in some regions, while investments in public service infrastructure would improve economic opportunities in others.
Poverty mapping is also used for emergency response and food aid programs. An example is given by the early warning and mitigation of natural disasters system of USAID’s “Famine and early warning systems network". This system helps select interventions in time for effective implementation during droughts or other natural disasters.
Designing interventions - Once an intervention has been decided, poverty mapping can further be useful for its actual design. Here, we list a few examples of use of poverty maps in the design of interventions.
Geographic targeting of resources - Targeted interventions aim at specific groups of the population, and the identification of the target population can be based on a variety of criteria. One possibility is to decide to target specific geographic areas. Sometimes, interventions decide to target specific individuals within specific areas (combining geographical targeting with some kind of test of individual compliance with the criteria selected - for instance, linked to income levels, enrolment in schools, participation in a specific health care program, etc.). Geographic targeting has limitations (e.g. when used in isolation, it targets both poor and non-poor in poor regions, and ignores the poor who live in non-poor regions), but its low design and administration costs make it often more effective than other options (see Bigman and Fofack 2000).
The smaller the geographic regions for which indicators are available, the greater the effectiveness of interventions - indeed, this allows to reduce leakage (i.e. transfers to the non-poor) and increasing coverage (i.e. minimizes the risk that a poor person will be missed by the program). Studies in India and Indonesia, for instance, show that states or provinces are too heterogeneous for targeting to be effective (Datt and Ravallion 1993, Ravallion 1994). This underlines the need for the collection of indicators for small areas, which are relatively homogenous (see Small area estimation maps).
Designing interventions with regional variants – Detailed information and analysis can help define where interventions are needed and which interventions are needed within the areas selected. It can also help chose the type of program that is most relevant. For instance, a poverty map could reveal lack of coverage of health care services in a few regions, but reveal that some of these regions have low levels of poverty while others experience much lower welfare. Low coverage might be mostly explained by the absence of health care services in the first group, and by the cost of health care services in the second group. Hence, a health care program could use some form of cost-recovery in the first group of regions, and be subsidized in the poorer ones.
Informing decentralization – Poverty maps can also help inform decentralization. For instance, they can help inform the level at which a certain type of intervention or service is best managed and controlled by showing the area which benefits from them. They can also be used to inform formula for fiscal transfers that accompany decentralization of responsibilities. (see Henninger and Snel 2000).
Fostering participation at the local level – At the local level, poverty maps can also play an important role in communicating information. Because all stakeholders, including non-specialists, can easily understand them, poverty maps are an important tool for the participatory definition of priorities and interventions.
For more information on applications of poverty maps see Documents and Links.
Back to Poverty Mapping