Reflecting on the "Putting Children First" Conference
A Blog by Christian Oldiges, OPHI, Oxford University
The overarching aim of the Sustainable Development Goals (SDG’s) is to half child poverty in all its dimensions by 2030. In her opening remarks at "Putting Children First", Dr. Akousa Aidoo (Vice-Chairperson and Rapporteur of the UN Committee on the Rights of the Child, 2007-2013) highlighted the need to act now as “child poverty is everyone’s problem and it should be a national government’s priority.” It is now widely established that child poverty is multidimensional in nature, and during the opening day of the “Putting Children First” Conference (Addis Ababa, October 23 to 25 2017), several stakeholders emphasized this basic yet important insight; including Jane Kabubo-Mariara from the PEP Network, who reiterated that “targeting (only) monetary poor children could miss many key dimensions.” The need for public policies directly linked to tackle multidimensional child poverty was shared widely among the audience. In the following, I explain - based on real world examples - how governments can easily apply the Alkire-Foster (AF) method to identify multidimensionally poor children and compute Child Multidimensional Poverty Indices (C-MPIs) that can be used to guide public policies.
The AF method is a flexible technique for measuring poverty. Once indicators are agreed upon and indicator-wise cut-offs are set for the measure of interest, it is a matter of counting each person’s deprivations. Summing up these deprivations, one applies a minimum threshold of deprivations which determines whether a person is multidimensionally poor or not. These simple steps allow us to calculate the incidence of the multidimensionally poor (H) and the average share of deprivations that the poor individuals experience. The latter gives us the intensity of multidimensional poverty (A). The product of the H and A yields the MPI, also called the adjusted headcount ratio, which is the number of deprivations the poor face as a share of the deprivations the entire population could face.
The Global Multidimensional Poverty Index (MPI) is probably the most famous application of the AF method. It is an international comparable measure of acute poverty, which takes into account deprivations in health, education, and living standards, and it is annually published in UNDP’s Human Development Report. Inspired by the Global MPI, many national governments have implemented National MPIs tailored to their specific circumstances. A case in point is the Colombian MPI launched by Nobel Laureate President Santos. Based on five dimensions spanning health, education, childhood conditions, employment, and living standards, the Colombian government is tracking and fighting multidimensional poverty annually. Yearly evidence and breakdowns by indicators and regions, enables the Colombian government to design public policies to improve, for example, housing and education in specific areas. The Colombian MPI has become a great tool for targeting the poorest regions and identifying the needs of the poorest Colombian sub-groups.
Following the basic steps of the AF method, it is very intuitive and straight forward to focus on children. This can be done in two ways. One, we can disaggregate the Global MPI, or any national MPI, by age-groups, as done recently by OPHI. The findings are startling. While children make up for less than a third of the world’s population, half of the multidimensionally poor people are children – hence children are disproportionally more likely to be poor than adults. At the same time, the intensity of poverty is much greater for children than for adults, while sub-Saharan Africa and South Asia together host 87 percent of the 689 million multidimensionally poor children.
Another way to focus on child multidimensional poverty is to create an index solely tailored to child specific needs. We can think of child development indicators which vary by age-group. Doing so, we consider both child-specific requirements and important household characteristics. One prominent example is Bhutan’s C-MPI: there are four dimensions, health, education, living standards, and childhood conditions. Both education and childhood conditions include age-specific indicators. Childhood conditions, for instance, includes malnutrition for children aged 0 to 4 years, child labour for children aged 5 to 14, and early marriage and child-bearing for children 15-17. Ideally, if the data allowed the latter would be an indicator for “Not in Education, Employment or Training” – NEET. In this way, age-specific indicators and cut-offs are applied and hence age-specific deprivations are counted at the individual child level.
At this point of writing, there are several countries in process of designing a C-MPI, including Panama which just launched a national MPI in 2017, and South Asian countries such as Afghanistan and Bangladesh. In any context, the first step is to identify child specific and country relevant indicators. The choice is crucial and depends both on data availability as well as on normative decisions and the national policy agenda. Ideally, national plans, constitutions, and the SDGs inform and justify the choice of indicators and indicator cut-offs. In contrast to other multidimensional indices, a C-MPI, based solely on the AF method, can be disaggregated by indicators and various characteristics. For instance, we can disaggregate by age and gender, and see if poor children live in MPI poor households. Even the household-level Global MPI (when age-disaggregated) shines a powerful light on child poverty as the Global MPI statistics showed. Research is ongoing to establish whether, empirically, these two methods identify different children. Important for policy makers and the public, a C-MPI is very transparent, as any change in the C-MPI can be traced to each and every indicator. This is a unique feature and makes the C-MPI an ideal tool for child focused public policies.
Christian Oldiges is a Research Officer at the Oxford Poverty and Human Development Initiative (OPHI), University of Oxford, and is involved in designing Child Multidimensional Poverty Indices in several South Asian and sub-Saharan African countries.
 OPHI Briefing 46, Alkire, Jindra, Robles, Vaz: Children’s Multidimensional Poverty: disaggregating the global MPI
 This number may be even bigger, as it only refers to the children who are covered by the global MPI sample of countries in 2017.
 For Bhutan’s Child MPI visit: http://www.nsb.gov.bt/mweb/docs/about/main.php?id=187&task=view