Skip to main content

In quest of inclusive progress: Exploring intersecting inequalities in human development

For 16 countries with appropriate data, this paper seeks to ascertain to what extent wealth status, urban/rural place of residence and ethnicity – and overlaps between them – explain inequalities in education and health; and how these inequalities have changed over time. Our focus is on women’s years of education and on the proportion of children in a household who have died.

We show that people who belong to one or more disadvantaged groups experience outcomes that are significantly below the average and that group-based identifiers explain a significant share of inequality in those outcomes in most countries. The gaps between the most disadvantaged group and the population average are pronounced in both education and health: the number of years of education ranges from one to nearly six, and the share of children who have died differs between three and 18 percentage points. Overall, women in the lowest wealth quintile from minority ethnic groups had the lowest average outcomes in education in 11 of 16 countries and the highest share of child deaths in 14 of 16 countries.

We then explore the relationships between place of residence, ethnicity and wealth – and find that people in the bottom wealth quintile live largely in rural areas and disproportionately belong to particular ethnic groups. We do not find a strong relationship between ethnicity and rural–urban status – within our sample, there is no clear pattern either within or across countries.

Turning to group inequalities, on the basis of a decomposition exercise, we find that the three characteristics we focus upon explain a significant share of inequality in education and in health in many countries. The contribution is especially high for education. Wealth explains between around 20% (Zimbabwe, Mali) and just over 40% (Bolivia) of total inequality. Place of residence and ethnicity each explain between less than 5% (Philippines) and 25% (Bolivia) of total inequality. For health, group based inequalities appear to be less significant – the three characteristics each account for less than 6% of inequality, but with variation between countries.

We next explore to what extent ‘intersecting inequalities’ – our focus is on belonging to two deprived groups – condition inequality in education and health. Two main findings are highlighted here. First, the effect of belonging to a disadvantaged ethnic group and living in a rural area is significantly larger than either of the component parts in many countries, for education and for health. For example, when looking at years of education  in Bolivia, ethnicity and place of residence each explain around 25% of total inequality. When examined jointly, they together explain close to 40%. The fact that the total contribution is often significantly higher than either component (though less than the sum of both) hints that this combination can be particularly pernicious. Where wealth is joined with place of residence and then with ethnicity, the effect of the intersection is only slightly higher than that of wealth. Second, inequality based on wealth and ethnicity jointly has changed less than the other two forms of overlapping disadvantage (wealth and residence; ethnicity and residence) over the past 20 years in education, while there have been fewer observable changes in inequalities in health.

Finally, we aim to give some insights into the effect of being at the intersection of two disadvantaged groups – in other words, whether there is a spillover effect associated with belonging to two disadvantaged groups such that the disadvantage is compounded. We explore this in a very preliminary way, so the findings are illustrative rather than indicative – but they hint at an additional burden of experiencing overlapping disadvantage.

These findings speak to the need for policies to address not only individuals from disadvantaged groups but also those who are doubly disadvantaged. Policy actions should prioritise people experiencing overlapping disadvantages. In some cases, technical solutions such as better targeting will be crucial, though the importance of ethnic marginalisation also draws into relief the need to tackle politically sensitive drivers such as social discrimination. Many of the policies are national – intersecting inequalities are experienced in different ways from country to country. At the same time, the Sustainable Development Goals (SDGs) and growing focus on ‘leaving no one behind’ add an important international angle. It will be crucial to monitor the success of the SDGs at a disaggregated level and to identify how to collect the data needed to do this well.