Beyond income: A multidimensional approach to tackling inequality

Rising income inequality in many developed and developing countries captures the attention of social activists and policy makers alike. Yet income is just one dimension of inequality. It exists in health, education, and social services, whose dimension-specific inequalities may reinforce or dampen the impact of income inequality. Focusing solely on income inequality offers only a partial view of what Amartya Sen has termed “economic inequality” limiting the scope and accuracy of a country’s policy responses.

Imagine a country or a region where the rich live in areas with the best public schools and the best publicly funded hospitals while the poor live where the schools and hospitals are of poor quality. Now, consider a counterfactual where the poor have access to top-quality public education and healthcare. The government implements a program to improve the quality of education in the lagging regions.

The impact of such an intervention can be assessed in two ways. In the first scenario, the poor will have access to better quality schooling, resulting in higher test scores for their children, while in the second scenario, the schools will be improved for the rich. Under some innocuous assumptions, we conclude that economic inequality would decline in the first case and rise in the second. But income and health inequalities are unchanged in both scenarios, while education inequality will be lower compared to the pre-policy situation. The so-called “dashboard” approach of measuring each dimension-specific inequality level separately (or taking a weighted average) ignores potential interactions across dimensions. However, such effects could be of critical importance as they convey valuable information on how people and societies experience inequality. Improving the quality of education for the poor will increase social mobility and allow many families to escape poverty, thus reducing income inequality. Investing in education in affluent areas might result in heightened political tensions and could stunt economic growth.

To account for these interactions and better gauge the extent of economic inequality, measures of multidimensional inequality have been developed. Despite significant advances in the range of tools available for measuring inequality across multiple dimensions, the policy impact of these measures has been muted.   

In our recent paper, we propose new multidimensional inequality measures that are easily implementable and transparent and overcome many deficiencies of existing measures. The paper aims to identify axiomatically sound multidimensional inequality measures with attributes well-suited for policy. The measures follow a traditional two-stage format, suggested by Maasoumi (1986), which aggregates dimensions first and then applies a unidimensional measure like the Gini coefficient to the distribution of aggregates.

We show that only a linear form can be used for aggregation of the individual-level components. Previous studies have considered linear aggregation, but this is the first paper to select this structure based on the axiomatic properties of its associated measures. We demonstrate that multidimensional inequality can be expressed as a weighted average of specific inequalities and a non-negative term reflecting the relevant aspects of the joint distribution across dimensions.

We also develop a calibration approach based on data in an initial period and normative policy weights. Once the multidimensional inequality measure has been calibrated, it can be used to gauge the multidimensional inequality through time and, with additional assumptions, through space.
 

Application and Impact in Developing Countries

We illustrate the application of our methodology by analyzing changes in multidimensional inequality in Azerbaijan from 2016 to 2023. We use data from the second (2016) and the fourth (2023) rounds of the Life in Transition Survey (LITS). We construct the multidimensional inequality index based on three dimensions captured by the monthly per capita income, years of education, and respondent’s subjective health assessment. The measure is calibrated for 2016 using normative weights of ½ for income and ¼ for the education and health dimensions.

Table 1 presents the specific and multidimensional inequality levels for Azerbaijan. The mean monthly per capita income increased by almost 59 percent from about 852 PPP dollars in 2016 to 1350 PPP dollars in 2023. Income growth was accompanied by an increase in income inequality, from a Gini of 0.253 in 2016 to 0.339 in 2023. The average years of education and health self-assessment remained relatively stable. The inequality in years of education grew while the inequality in health assessment slightly declined.

Table 1: Specific and multidimensional inequalities in Azerbaijan, 2016–2023.

Multidimensional inequality M(x)  as measured by Gini, increased between 2016 and 2023, from 0.144 to 0.230. This increase is due to (i) changes in the specific inequalities, (ii) changes in the effective weights as dimensional means change, and (iii) changes in the rearrangement term R(x). The rearrangement term fell slightly from 0.037 to 0.029, reflecting greater alignment of dimensions in the second year, meaning that people with higher incomes also got better education and health. To reduce economic inequality in the country, the government of Azerbaijan might invest in improving the quality of education in the country’s poorest regions.

Expanding Understanding of Inequality

By adopting such a multidimensional approach, developing countries, with their limited resources and high inequality rates, can better address their policy challenges. Tackling income inequality is notoriously difficult, and such policies often generate undesirable second-order effects. Our framework allows countries to reduce economic inequality by providing public goods to people experiencing poverty, which is straightforward from the implementation perspective and much more palatable politically. By recognizing the interconnectedness of various life dimensions, policy makers can devise more effective strategies to promote equitable growth and ensure that no one is left behind.