Published on Data Blog

The Prosperity Gap incorporates a penalty for high inequality

This page in:

This is the eleventh blog in a series of blogs about how countries can make progress on the interlinked objectives of poverty, shared prosperity and the livable planet. For more information on the topic, read the 2024 Poverty, Prosperity, Planet Report

Growth in average incomes alone is not a good indicator of development progress if this growth only benefits a small group of people. How the benefits of growth are distributed among the population matters for countries’ long-run development path. To this end, the World Bank has introduced a new measure that aggregates into one index countries performance on growth and the equitable distribution of this growth among the population: The Global Prosperity Gap. This new index tracks the average income shortfall from a $25-a-day threshold (in 2017 PPP$). As individuals move closer to this threshold, the Prosperity Gap narrows, indicating higher welfare. The index rewards income growth but gives a higher weight to income growth of the poor and hence penalizes increases in inequality. Essentially, the measure improves as income grows or inequality falls and worsens as incomes fall or inequality rises. (for further detail on the index, see this and this blog or this paper). 

How does the Prosperity Gap incorporate penalty for inequality?

To see the impact inequality has on this new measure, let’s look at some examples.  Table 1reports the Prosperity Gap, average per capita daily income or consumption, and inequality for two low-income countries (Benin and Cameroon) and two middle-income-countries (Peru and Colombia). Benin and Cameroon have similar levels of mean household consumption ($5.02 and $5.37 respectively), but Benin has a much lower Prosperity Gap (lower gap implies better welfare) than Cameroon because of Benin’s lower level of inequality. Said differently, despite Cameroon’s 7 percent higher mean consumption, average consumption would need to increase 15 percent more in Cameroon (or by 8.3-fold) compared to Benin (7.2-fold) to reach the same prosperity threshold of $25-a-day.

Likewise, despite Colombia being significantly richer than Peru in terms of mean income, Peru has a lower Prosperity Gap (or higher welfare) due to Peru’s much lower level of inequality. Put differently, higher inequality implies that Colombia will need an average income of $22.60 per day (or 30 percent higher than its current level of $17.30 per day) to have the same Prosperity Gap as Peru that has a mean income of $12.40 per day. This illustrates how the Prosperity Gap index factors in an “inequality penalty.”


Inequality delays progress towards prosperity

High inequality can also delay how quickly countries reach the $25-a-day threshold. To see why, let’s have a look at Figure 1, which plots the number of years it will take select upper-middle-income countries to reach the $25 threshold, assuming each country grows at 2 percent per capita annually (dark blue bars), or 4 percent per capita annually (yellow bars). In both scenarios we keep inequality within the country fixed. Countries are ranked according to their mean per capita income, with the poorest at the top and the richest at the bottom. While we would generally expect richer countries to face a shorter time horizon than the poorer countries to reach the prosperity threshold, this is not always the case. For instance, Serbia with a lower mean income than Brazil, will reach the threshold much faster due to its lower inequality. This thought experiment highlights that, by actively working toward reducing inequalities today, countries can pave the way for faster shared prosperity gains in the future. (Select the country-group at the top left of in Figure 1 to explore time horizons faced by other countries.)

Figure 1: Inequality delays progress towards prosperity


There is nothing inevitable about inequality

Inequality is determined by historical conditions and market dynamics, but also by policy choices.  Policies can target structural sources on inequality at three interconnected levels.  First, they can address inequality in acquiring human capital and other assets, before individuals join the labor market, such as policies aimed at reducing differences in educational attainment. These differences oftentimes reflect an inequality of opportunity, encompassing factors that are linked to circumstances that are out of an individual’s control, such as birthplace, parental income, gender, race, and others. Second, policies can address inequality in using skills and assets, which arise from market and institutional distortions in the labor, product, capital, and input markets. These distortions include anticompetitive or discriminatory practices, or policies and regulations that provide preferential treatment or restrict market access for some, while limiting access for others, thus curtailing their productive potential and limiting earning opportunities. Third, fair fiscal policies can be leveraged to improve the redistributive impact of taxes and transfers.

Structural inequalities at each stage reinforce each other and are dynamic. For example, disparities in parental income, indicative of past inequality, directly affect the current generation's productive capacities and their ability to leverage these capacities, contributing to intergenerational mobility. A comprehensive approach that addresses barriers at all three levels is necessary to effectively reduce inequality and enhance socioeconomic mobility.

The authors gratefully acknowledge financial support from the UK Government through the Data and Evidence for Tackling Extreme Poverty (DEEP) Research Program.


Maria Eugenia Genoni

Senior Economist, Global Lead on Data Systems and Statistics Operations, Poverty and Equity Global Practice, World Bank

Christoph Lakner

Program Manager, Development Data Group, World Bank

Nishant Yonzan

Economist, Development Data Group, World Bank

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000