Research Brief
Trends in global inequality – a comprehensive approach

Recent analysis of inequality trends — which relies on the World Income Inequality Database (WIID) Companion, a pair of rich new datasets — assesses the extent to which inequality trends depend on certain views about inequality, including whether absolute or relative income changes matter more and which changes in the performance of income growth in different parts of the distribution we should care more about. It also examines whether the inequality trend is driven more by differences between countries or within countries and quantifies the contributions of different countries and regions to the global trends.

Absolute inequality

The results of the analysis (Figure 1) indicate that between 1950 and 2019 the dollar income distances between people (absolute inequality) unambiguously increased almost continuously, both between countries and within countries, and therefore globally. Only deep recessions seem to have temporarily reduced absolute inequality.

Relative inequality between countries

In terms of the relative income distances between people (relative inequality) the story is more nuanced (Figure 2).

By most indices, the first decades included in the analysis are characterized by some overall stability in inequality, with a slight upward trend.

Figure 1: Global absolute Gini index (1950-2019)

This upward trend was driven by the fact that large low-income regions, particularly China and India, were left behind during the sustained economic growth that followed the Second World War. This upward between-country inequality trend was further aggravated by population growing in the developing world faster than in western countries. These factors were, however, largely compensated by declining inequality within countries, particularly in China and India.

In most recent decades, we observe a sharp decline in global inequality once these drivers were totally reversed. A large decline in inequality between countries was driven by stronger economic growth in emerging countries, especially in China and, to a lesser extent, India during the globalization era that followed (Figure 3).

Figure 2: Relative measures of global income inequality (1950-2019), Gini and Generalized Entropy 

The decline of relative inequality between countries accelerated rapidly as China’s mean income converged on the global mean. In recent years, this trend is clearly decelerating, however, as China gets richer. Once China’s mean income surpasses the global mean, Chinese income growth can be expected to eventually increase, rather than decrease, relative inequality between countries. The Chinese contribution to recent changes in inequality is already close to zero for various indices.

The recent decline in inequality between countries has only partially been offset by the disequalizing effects of faster population growth in sub-Saharan Africa, now the poorest region globally, and growing within-country inequality — including within China and India.

Figure 3: Country contributions to between- and within-country inequality, Gini: selected countries

Significant variations based on the measures used

The turning point in the global trend in relative inequality varies, by measure, between the mid-1970s and the late 1990s. The turning point is in the mid-1970s if we pay more attention to the relatively good performance of the incomes of the world’s bottom 40 per cent, but in the late 1990s if we place greater emphasis on the increasing concentration of global income in the top 10 per cent during the 1980s and 1990s, driven in part by the collapse of socialist regimes in Eastern Europe but also similarly rapid increases at the top in several other countries. With less sensitivity to either end of the distribution (e.g., the Gini index), the decline in global inequalities starts in the early 1990s.

One important point of discrepancy when assessing the global trends emerges if we pay much higher attention to the very bottom of the income distribution. In that case, inequality sharply declines until around 2005, when it begins to increase again with the stagnation of incomes among the poorest 5%.

Relative inequality within countries

It is important to note that these discrepancies among different inequality sensitivities arise mainly from how the different indices evaluate the recent trend in inequality between countries, rather than within countries. Between countries, the inequality trend will depend on what distributional changes are given more relevance, but within countries there is agreement about the direction of the inequality trend across different measures, even if the magnitude varies. Inequality within countries has shown heterogenous trends by region, with important recent declines in sub-Saharan Africa, Latin America, or MENA countries, and recent increases in the most populated areas, including East Asia, South Asia, and North America.

In sum, to assess global inequality trends over time, researchers should be more explicit about their value judgements. Different trends are obtained depending on whether emphasis is placed on absolute or on relative income changes, and on how much relevance is given to income changes at different parts of the distribution. The lack of a common narrative on global inequality is likely aggravated by the decoupling of the trends observed between and within countries, and the heterogeneity of within-country inequality trends across regions.

What is clear is that while absolute income distances have unanimously increased, relative inequality has mainly declined — at least in the last two decades — reflecting a large improvement in the living conditions of the world’s low- and lower-middle-income groups. But inequality remains high, incomes for the poorest 5% are stagnant, and there is no guarantee that the convergence in country mean incomes that drove this process will continue for much longer.