Lenhardt, Amanda; Rodriguez Takeuchi, Laura; Samman, Emma (2014). 'Decomposing inequalities to track progress in human development and wellbeing' Paper presented at the annual conference of the HDCA, 2-5 September 2014, Athens, Greece.

Paper Title: Intrahousehold inequalities in child wellbeing. A barrier to progress?

Author: Laura Rodriguez Takeuchi, ODI

Empirically, the extent of intra-household inequality is difficult to assess. For analytical convenience, most policy analysis assumes that within households resources are assigned according to need, treating individual wellbeing as the average adult-equivalent of the household to which they belong. There is a considerable shortage of data analysis on the wellbeing of children specifically, with most information derived from their households or carers. The neglect of intra-household inequalities affects the assessment of the levels of poverty and inequality, and could lead to a skewed view of the patterns of progress in improving child wellbeing. We now have much better data to look at this progress. Using two rounds of Multiple Indicators Cluster Surveys (MICS) for approximately 20 countries, this paper aims to measure the extent of gender inequalities in children within the same household. It shows how much intra-household inequality contributes to the overall levels of inequality and how it has evolved over time. Achievements in four dimensions are examined: nutrition, education and birth registration. To measure inequalities the paper uses a Theil Index decomposition to capture the extent of within-household and between-household inequality for each of the indicators. The analysis shows different depths of within-household inequality across countries and indicators which relate to uneven distributions of gains for boys and girls within households. Finally, household ratios of girls to boys' achievements are used to identify some household characteristics associated with a larger extent of intra-household inequalities.

Keywords: intra-household, child wellbeing, inequality decomposition

Paper Title: Measuring change in intersecting inequalities in health and education. What progress has been achieved?

Author: Amanda Lenhardt, ODI

In order to eliminate extreme poverty, policies will need to be enacted which not only seek to address the root causes of inequality, but also development aims that account for the marginalisation of certain groups. When tracking progress towards a new set of goals, this progress should be disaggregated to look at the gains made by groups that might be excluded from the aggregate gains that are observed. Failure to do so could mean that while non-excluded groups progress out of poverty, marginalized groups will continue to be 'left behind' and gaps between them and the rest will grow.  Intersecting inequalities refer to overlapping disadvantages that an individual or group faces which lead to the deepest marginalization in a society. A person's ethnic identity, religion, gender and spatial location can interact in a way that leaves them excluded from the economy, political systems and social life altogether. The particular overlaps that characterise marginalization in any given country or region vary by context, but poverty is strongly associated with identities that are ascribed at birth – race, caste, and ethnicity – and most pronounced when intersected with disadvantaged locations, economic groups  and gender. This paper aims to contribute to the understanding of how intersecting inequalities manifest across a sample of contexts using DHS surveys for 16 countires with 2 survey points between 1990 and 2012. It also draws out indications of reduced intersecting inequalities that indicate effective targeting of the causes of inequality, or that progress is being made on these human development challenges in an inclusive way. This paper applies a general entropy measure to decompose inequality at a group level and compares outcomes for intersections of group identity across place of residence, economic groups, ethnic identity and gender. This paper also expands the uses of these measures to look beyond income by instead looking at health and education outcomes. 

Key words: intersecting inequalities, inequality decomposition, inclusive human development


Data collection for the measurement of group-based difference

Author: Emma Samman (ODI)

The importance of understanding better group-based as well as individual inequalities has gained prominence in academic circles in recent years, and it has more recently started to gain traction in the policy world as well. This past year, the report of the High Level Panel on post-2015 advocated a specific focus on group-based inequalities under the banner of leaving no-one behind: 'We should ensure that no person – regardless of ethnicity, gender, geography, disability, race or other status – is denied universal human rights and basic economic opportunities. We should design goals that focus on reaching excluded groups, for example by making sure we track progress at all levels of income, and by providing social protection to help people build resilience to life's uncertainties… Metrics should be put in place to track progress on equal access and opportunity across age, gender, ethnicity, disability, geography, and income'. Alongside this call, it urged a 'data revolution': 'Data must also enable us to reach the neediest, and find out whether they are receiving essential services. This means that data gathered will need to be disaggregated by gender, geography, income, disability, and other categories, to make sure that no group is being left behind'.This paper takes as its starting point these recent calls for a focus on group-based inequalities within countries, alongside renewed impetus for data collection efforts that are able to monitor the needs of marginalized groups – and considers how they might be realized. It is structured in three parts. First, it assesses what types of measures would be desirable, with a view to monitoring a post-2015 agreement that is explicitly concerned with reducing group-based difference. It then considers what types of measures are possible, considering current data collection efforts, and the resources and capacity of National Statistical Offices, as well as how politics and public attitudes shape data collection in many countries. Finally, it considers how other forms of data collection could potentially complement current measurement efforts to shed greater light on the circumstances of marginalized groups. To this end, it will review community-based monitoring systems, community score-cards and SMS-based data collection efforts. It considers both the potential for these types of data to inform officially collected statistics, and also their limitations.

Key Words: group-based inequalities, inequality decomposition, data revolution