Multidimensional Poverty and the AF method

Conceptually, poverty has traditionally been understood as a lack of money with statistics on consumption or expenditure acting as a proxy for a household's quality of life. However, in recent decades a consensus has gradually built that focusing on monetary poverty alone, or on any single indicator for that matter, is not enough to capture the lived reality of poverty and that measurement can do more. 

Multidimensional poverty conceives of poverty as an experience of overlapping deprivations. A person who is poor can suffer from multiple disadvantages at the same time – for example they may have poor health or be malnourished, they may also lack clean water or electricity, they may have had little schooling, or be unemployed or precariously employed. Multidimensional poverty measures identify whether or not people are living in poverty based on whether they experience a critical mass of possible deprivations.

The study of multidimensional poverty has its roots in participatory studies in which people in poverty were asked about their lives. In these studies, people described their experience as one of multiple deprivations, and part of its burden was that so many disadvantages struck them simultaneously. 

There are multiple reasons why it is important to measure poverty using a multidimensional approach:

• To enhance our understanding of poverty to inform action.
  "The heart of public action lies in a clear-headed understanding of what the problems are" – Professor Amartya Sen

  Multidimensional poverty measures complement monetary measures to create a more comprehensive picture of poverty. Measures such as the Multidimensional Poverty Index (MPI) reveal both who is poor and how they are poor, illuminating the set of different disadvantages they experience at the same time. As well as providing a headline measure of poverty for the population being analysed, MPIs can reveal the poverty level and indicator composition among different sub-groups of people or in different areas of a country. MPIs can help prioritise the poorest communities, guiding more impactful multisectoral interventions.
  
  MPI does not simply report the percentage of people who could be considered poor. An MPI also shows the intensity of poverty – the average share of weighted deprivations faced by poor people. 
  
  MPIs go beyond dashboards. A dashboard provides information on poverty which is presented in silos, it cannot shed light on which people experience two or more of those deprivations at the same time. In contrast, an MPI shows where the greatest deprivations cluster and persist in a community.

• To reveal poverty which would otherwise be hidden and address it 
  Studies since the early 1980s have shown that people who are poor according to income or consumption poverty measures are not necessarily the same people as those who are multidimensionally poor. Sometimes, the mismatches can even be high. 

  A lack of services relating to electricity, schools, healthcare, or waste disposal for example, can impede the lives of people even if their monetary resources are slightly above the poverty line. 

• Internationally recognised approach
  The internationally recognised conceptual definition of poverty has broadened to include a multidimensional perspective since the turn of the 21st century. Goal One of the UN’s Sustainable Development Goals (SDGs) aims to end poverty in all its forms and dimensions. The SDGs also recognise multidimensional poverty explicitly in the second of the 169 SDG Targets - namely, Target 1.2 which is by 2030: "[to] reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to national definitions."

A Multidimensional Poverty Index, or MPI, is associated with an ‘information platform’ showing the overlapping deprivations that affect multidimensionally poor people. Basic analyses highlight which deprivations cluster together among specific groups so that integrated policies can address these deprivation bundles effectively. Official national MPIs are one of the leading measures of multidimensional poverty that are in use around the world and are used by over forty governments.

The MPI is customisable and can be tailored to the needs of a country or dataset being analysed. The data underlying a national MPI is usually a household survey. 

The MPI is built with a selection of indicators chosen to reflect important characteristics of the lived experience of poor people across a range of aspects in their lives. These aspects are termed 'dimensions', and often cover health, education, living standards, and work among others. The MPI can provide information on the nature of poverty across the whole group surveyed, and can also be disaggregated by subgroups to provide information on how poverty, for example, is different among children and adults, or people living in rural areas rather than urban areas, or citizens in the north of a country rather than the south.

The global MPI is an example of an MPI that has been customised to enable comparisons across countries. OPHI publishes the global MPI annually with the United Nations Development Programme to highlight multidimensional poverty across 100 countries in developing regions.
  
For more information on the MPI visit our pages on the National MPI and Global MPI.

The Alkire-Foster (AF) method is designed to be an intuitive counting method to measure complex topics, which was created by Professors Sabina Alkire and James Foster.

In a measure based on the AF method indicators form the fundamental components. To take the MPI as an example, indicators are selected that capture important and measurable aspects of the lived experience of poverty. Dimensions are the conceptual groupings of those indicators. 

The AF method first considers if a person (or household, depending on the ‘unit of analysis’) is deprived or non-deprived in each indicator. Each indicator has a weight, which shows its relative importance. A person’s deprivation score is computed by adding up the weighted indicators in which they are deprived. To identify who is poor, a poverty threshold, or cutoff, is set. Each person is categorised as poor or not poor depending on whether their deprivation score is equal to or greater than the poverty cutoff. Robustness Tests are performed to ensure the structure of the MPI is not unduly sensitive to its parameters.

In the case of the MPI, the AF method is a useful framework for measuring multidimensional poverty because of the amount of consistent and policy-relevant information it generates. The three key statistics that are used to describe multidimensional poverty are: 

• Incidence (H): the proportion (expressed as a percentage) of the population who are multidimensionally poor. It is sometimes called the ‘poverty rate’ or ‘headcount ratio’.
• Intensity (A): the average percentage of weighted indicators in which poor people are deprived – that is, the average deprivation score among poor people.
• MPI: the headline figure of poverty representing the share of possible deprivations that poor people in the society experience. The MPI is computed by multiplying the incidence by the intensity (MPI=H x A) and the MPI ranges from 0 – 1, where 0 represents no poverty and 1 represents universal poverty. Given that ‘MPI’ is the name of the overall tool, sometimes the term ‘MPI value’ is used to describe this headline figure.

The MPI (unlike the incidence) can also be broken down by indicator to show which deprivations are highest and how these vary across population subgroups. The MPI value and all related statistics can also be potentially disaggregated by characteristics such as age, urban/rural areas, and states or districts to show who in the population is poorest and how they are poor.

The advantages of the AF method for measuring multidimensional poverty using an MPI are as follows:

• The AF method provides information on the incidence, or percentage of people in poverty. Analysing the percentage of people in poverty alongside the numbers of poor people, is commonplace. As the MPI is updated, the trends in incidence intensity and indicator deprivations are of interest. These are reported with their standard errors.
• The AF method provides information on the intensity of poverty experienced by poor people. This is the average proportion of weighted deprivations that people face. This is important for a ‘pro-poor’ approach to poverty reduction.  Rather than focusing on a single threshold of poverty (the cutoff), policymakers have information about the entire distribution of deprivation scores of poor people. This information is important so that the focus is not placed solely on poor people just under the poverty cutoff for multidimensional poverty.
• The AF method can be disaggregated, it can be broken down or filtered, for example, by region and by other groups for which data are representative to provide information to policymakers about the priorities and needs of especially vulnerable communities.
• The AF method is customisable and flexible. Different dimensions, indicators, and cut-offs can be used to create measures tailored to specific uses, situations, and contexts. These can be chosen through participatory processes, such as in El Salvador.
• The AF method importantly shows how weighted deprivations cluster or overlap. Analysis can show if there are common deprivation profiles, or deprivation bundles, that could be effectively solved by specific multisectoral interventions.
• The AF method can show changes over time or trends to monitor and track progress in poverty reduction.
• The AF method can be used alongside monetary measures to enhance the understanding of poverty in a society.
• The AF method can reveal progress that might otherwise have been hidden. A measure based on the AF method reflects improvements even if the overall goal of moving more people out of poverty is not achieved. If progress is registered in either incidence of intensity, the MPI value will reflect this and decrease. If a population is increasing, but poverty is going down, the MPI will show this nuance.

The AF method underpins the MPI and other measures advanced by OPHI, such as Bhutan's Gross National Happiness Index (GNH) and the Women’s Empowerment in Agriculture Index (WEAI). 

What are we trying to measure when we talk about acute multidimensional poverty? Beyond the graphs and charts what does this relate to? See this gallery.