What we (don’t) know about gender gaps in multidimensional poverty …
Gender gaps are pervasive in many dimensions of well-being. Globally, almost two-thirds of the world’s illiterate adults are women, because of past (and sometimes present) gender inequalities in access to schooling. Women are also often more “time poor” than men due to the double burden of labor market activities and domestic chores and more “asset poor” due to gender biased laws, traditions and institutions.
While these patterns are well-known, most multidimensional indices of poverty use the household as the unit of analysis to determine the poverty status of individuals. Meaning that if a household is identified as being poor, everyone in that household is considered poor. This includes UNDP’s Global Multidimensional Poverty Index (developed together with the Oxford Poverty and Human Development Initiative (OPHI) at the University of Oxford), as well as the multidimensional poverty measure introduced in the World Bank’s 2018 Poverty and Shared Prosperity Report (PSPR). Why? In part, it is because some dimensions of well-being – such as material living standards, monetary poverty, access to services, etc. – are hard to measure for individuals and so most household surveys collect this information for the household as a single entity.
Moreover, the household-centered approach does not use individual-level data even when they are available. Consider the dimension of education. Household-level measures of multidimensional poverty collate the information about the educational attainment of household members into an indicator for the household – for example, the household is considered as deprived if no adult member has completed primary education. This assures that all members of the household are assigned the same poverty status. Hence, information on differences between individuals living in the same household, such as by gender or age, is lost.
Towards a multidimensional poverty measure for individuals
Chapter 5 of the Poverty and Shared Prosperity Report, which builds on analysis by Stephan Klasen and Rahul Lahoti, illustrates how we might get multidimensional poverty measures for individuals and not simply households. Following the extended measure of multidimensional poverty introduced in chapter 4, deprivations across five dimensions – monetary poverty, education, health and nutrition, access to services, and security – are aggregated into an individual multidimensional poverty measure for adults in five countries (Ecuador, Indonesia, Iraq, Mexico, and Tanzania). Children are excluded from the analysis, because educational attainment resides in the future for infants and young children, and because children require different indicators of nutritional status than adults.
However, due to data limitations, only the education and health & nutrition dimensions give us insights at the individual level. Adults are considered deprived in the education dimension if they have not completed primary schooling, and they are considered deprived in the nutrition indicator if they have a body mass index below 18.5, a common measure of underweight. The other three dimensions — monetary poverty, access to services, and security — may be analyzed only at the level of households with the existing data. The multidimensional poverty measure is hence only partially individualized. But despite these drawbacks, the analysis reveals interesting gender differences in multidimensional poverty.
Figure 1 shows the share of men and women who are deprived in the two indicators on which data about individuals are available: education and nutrition. For each country and indicator, deprivation rates among men and women are compared through two approaches: one relying on the household, whereby all household members are assigned the same deprivation status, and the other relying on the individual, measuring individual deprivations directly. In education (panel a), the household approach reveals some gender differences in education deprivation that disadvantage women. Not surprisingly, these gender gaps, which are muted under the household approach, are amplified if the data on individuals are used. In all five countries, women are much more likely to be deprived in education than men when deprivations are measured across individuals. In terms of nutrition (panel b), gender gaps are small, even if measured with reference to individuals, and they do not show a consistent pattern.
Figure 2 summarizes this information to show the share of men and women who are multidimensionally poor using the individual approach. Multidimensional poverty is more prevalent among women than among men in all countries studied. These gaps may even be wider among the most vulnerable groups. For example, the individual multidimensional poverty measure highlights widowhood as an important vulnerability factor among women, which is not revealed in the household multidimensional poverty measure.
Next steps …
This analysis is only a first step toward a more comprehensive measure of individual multidimensional poverty. Existing research points to gender and other intrahousehold inequalities in monetary poverty, but a more refined measure of individual multidimensional poverty that includes deprivation from lack of access to services (due to social norms that assign domestic work to women), exposure to different forms of violence, etc. should be considered in the future. Still, even a partially individualized measure as the one we discuss here for five countries reveals that women are more likely than men to be multidimensionally poor, driven by largely women’s disadvantaged position in educational attainment.
To learn more about global poverty in general, and specifically about differences in well-being within the household, read the recently released Poverty and Shared Prosperity report 2018, “Piecing Together the Poverty Puzzle.”