Media Briefings

GLOBAL INEQUALITY: New household evidence from the fall of the Berlin Wall to the Great Recession

  • Published Date: April 2014

Income inequality at the global level remains high – and the most important source is income differences between countries. These are among the findings of new research by Christoph Lakner and Branko Milanovic, to be presented at the Royal Economic Society’s 2014 annual conference.

 

The researchers analyse new data based on national household surveys from 1988 to 2008, assigning every person in the world the average income of his or her income bracket. Incomes are measured per person and adjusted for differences in living costs across countries. According to the report:

 

·      Between 1988 and 2008, there was little reduction in total global inequality. But the global distribution and its regional composition have undergone profound changes.

 

·      The proportion of the world’s poorest people living in each of five broad areas – China, India, sub-Saharan Africa, the mature economies and the rest of the world – has changed a lot.

 

·      Average incomes in China have increased significantly. And a larger proportion of the world’s poorest people now live in sub-Saharan Africa.

 

·      The upward movement of China has changed the overall shape of the global distribution from a twin-peaked to an almost single-peaked distribution.

 

The authors comment:

 

The most important source of inequality is income differences between countries. Thanks to the strong growth in some parts of Asia, this source of inequality has declined, but it still explains more than 75% of global inequality.’

 

More…

 

In recent years, there has been a renewed interest in the global distribution of income, where every individual in the world is assigned her/his income. This presents formidable measurement challenges, for example, the availability of household surveys, or the comparison of incomes over time and across countries.

 

Christoph Lakner and Branko Milanovic present new evidence on global inequality using a newly compiled and improved database, based on national household surveys from 1988 to 2008.

 

The data are drawn mostly from the World Bank’s PovcalNet database and the Luxembourg Income Study. Every person in the world is assigned the average income of his or her income decile. Incomes are measured in per capita and adjusted for purchasing power parity differences across countries.

 

With a Gini coefficient of around 70% in 2008, income inequality at the global level remains high, compared with inequality within countries. Over these 20 years, global inequality declined from around 72% in 1988. The most important source of inequality is income differences between countries. Thanks to the strong growth in some parts of Asia, this source of inequality has declined, but it still explains more than 75% of global inequality, as measured by the Theil-L index.

 

A recent literature based on tax unit records documented that household surveys might fail to measure top incomes accurately. As a result, surveys might underestimate inequality and could fail to notice changes in inequality at the very top of the distribution.

 

This new study offers a first (preliminary) analysis of how this might affect estimates of global inequality. Top incomes are approximated by the gap between national accounts consumption and survey means, in combination with a Pareto-type imputation of the upper tail. The estimated global Gini is much higher, at nearly 76% in 2008, and the downward trend almost disappears.

 

The main part of the study looks below the surface at the movement of individual country-decile groups. The green line in Figure 1 shows how the 20-year income growth rate between 1988 and 2008 (on the left axis) varies with the position in the 1988 global distribution. The graph shows that those country-deciles that in 1988 were around the global median grew fastest over the next 20 years, their incomes almost doubling.

 

The blue bars (read from the axis on the right) show that almost 90% of these people came from Asia. On the other hand, those individuals who were around the 90th global percentile in 1988 experienced the slowest growth in the next 20 years. Almost all of them came from rich countries (the red bars in Figure 1). The mature economies include the 27 member states of the European Union and the other high-income countries in the world.

 

As a result of these differences in income growth rates across the world, the global distribution and its regional composition have undergone profound changes over this 20-year period. Figure 2 decomposes the global income distribution into China, India, sub-Saharan Africa, the mature economies and the rest of the world.

 

There has been a clear rightward shift of the Chinese distribution, that is, towards higher incomes. By contrast, an increasing proportion of the poorest people in the world come from sub-Saharan Africa, which is also confirmed by the evidence on global poverty (Chen and Ravallion, 2008).

 

The upward movement of China changed the overall shape of the global distribution from a twin-peaked to an almost single-peaked distribution. The share of the world population living on between $2 (PPP) and $16 (PPP) per day (that is, between $730 and $5,840 per year) increased from 23% to 40%.

 

ENDS

 

 

Notes for editors:

‘Global Income Distribution: From the Fall of the Berlin Wall to the Great Recession’ by Christoph Lakner and Branko Milanovic

 

For further information, contact:

Romesh Vaitilingam: romesh@vaitilingam.com, +44 7768 661095