Notes on Global Income Inequality: A Non-Technical Summary
Branko Milanovic with a useful primer on the persistent problem.
1. What is global inequality?
Global inequality is inequality between all citizens of the world. It treats the world as a single unit (as we normally treat individual countries). The data used to calculate global inequality come from the nationally-representative household income surveys that are increasingly (when available) corrected for the underestimation of top incomes using fiscal data. It also adjusts for the differences in price levels between countries by expressing all incomes in international (or PPP) dollars that in principle have the same purchasing power anywhere in the world. Income is defined as annual after-tax and after-transfers income per capita (where total household income is divided equally among household members).
2. How accurate are such estimates?
The estimates of global income inequality are probably biased downward for two reasons. Some of the poorest countries (many of them in Africa) do not field regular household surveys, or are engaged in civil or international wars, and thus are not included in the calculations. However, the available data cover more than 90% of world population and more than 95% of world income.
The richest people tend often not to participate in surveys or underestimate their fiscal income in order to minimize taxes they pay. Thus both the top and the bottom of global income distribution are underestimated. The underestimation of the top is thought to be slightly increasing, but not to the extent to affect the long-term change in the level of global inequality (discussed next).
3. Long-term evolution of global inequality.
The long-term evolution in global inequality (to the extent that we can estimate things well in the 19th century) can usefully be divided into three periods.
The first period was the one of steady increase in inequality from the 1820s (when the first estimates are made) until 1914, and then of a somewhat slower, and irregular, increase up to 1950. The increase was driven by the “take-off” in economic growth and thus incomes of Western European countries, followed by North America and Japan. Meanwhile, Indian and African incomes stagnated, and China’s income went down. This created massive divergence and drove global inequality up. In addition, within-national inequalities in many countries (e.g. UK, US, Germany, Japan) increased during the Industrial Revolution.
Therefore, between the Napoleonic wars and World War I, we can with some confidence say that global inequality was pushed up by both divergence in mean country incomes and increasing within-national inequalities. The latter mostly reflected changes in functional income distribution, that is in class distribution between landowners, capitalists and workers. The between-country developments however dominated then, and continue to play greater role in the evolution of global inequality than within-country developments.
The second period was between 1945 and 1980. Global inequality was at its all- time peak, as the world became divided in three very distinct (by their income levels) worlds. The rich countries were indeed the “cities” of the world and large Third World areas were “the countryside”. Both India and China, just maintained their relative income positions worldwide (that is, their mean income compared to the world mean was constant).
The third period commenced with fast growth of China that was followed by Vietnam, Thailand etc., and then by India. This, for the first time since the early 19th century, reversed the direction of change, and began to drive global inequality down. China was the primary engine, but around 2000, India began to play an important role. Currently, global inequality is about 63 Gini points, which is some 7 Gini points (or 10%) less than in 1980s. The level however is still extremely high: the world is about as unequal as South Africa, which is the most unequal country in the world. For comparison, US Gini (of after-tax) income is just over 40, and Brazil’s over 50 Gini points.
4. Several implications of global inequality
China. As China’s income (GDP per capita) is now slightly above the world mean, it no longer contributes to the reduction in global inequality. Moreover, China’s faster growth (compared to the rest of the world) will begin to contribute positively to global inequality, at first modestly, and afterwards more strongly. Thus, we should not reflexively consider China any longer as an engine of global inequality reduction.
China itself is of course very unequal despite the fact that inequality is not on the rise since approximately 2010. China’s inequality level exceeds that of the United States, and it has one of the highest urban-rural gaps in the world: the average income of China’s urban population is equal to that of Hungary, while the average income level in rural areas is equal to Vietnam’s.
India and Africa. This makes the roles of India and Africa more important. The recent disastrous developments in India (with two successive years of likely high negative growth) as well as the long-standing problem of lack of convergence of African countries, opens up a real possibility that global inequality may cease its decrease, and might increase again.
This is even more likely as Africa is the only region of the world with projected high population growth. A back-of-the-envelope calculation that would require Africa to grow at about 5% per capita annually, implies a growth rate of 7% or even 8% for the economy as a whole. For comparison, in very “good” years before the Financial Crisis, African (population-weighted) growth was around 3-3.5% per capita, and more recently, before the covid crisis it was 1.5% per capita. Absence of sufficient African convergence will likely increase migrant flows, especially towards Europe. Thus, the European migration crisis should be seen as a secular, not at all as a temporary, issue.
Global inequality in historical perspective. The income changes described above bring the distribution of relative incomes within Eurasia at the same point where it was around approximately 1500. At that time, incomes in the richer parts of China were on par with incomes in richer parts of Europe (Italian city states, the Netherlands). Prior to that, it is likely that the Yangtze valley and coastal areas of China had even slightly higher incomes than Europe. The level of incomes at that time was, at best, 2 to 3 times the subsistence, so the differences in absolute incomes were small. Nevertheless, this fact is important in order to understand better that the period from approximately 1800 to 2000, with large income gaps between European (and North American) areas compared to China and India, was a historical anomaly.
Russia. Likely future equalization of incomes between Europe and East Asia (China) highlights the potential problem of lower mean incomes in the vast and sparsely-populated land mass of Russia and Central Asia.
Global positional reshuffling. As Asian countries improve their relative positions, an increasing number of Asian countries’ citizens (not only Chinese and Indians, but also citizens of Thailand, Indonesia, Vietnam etc.) will populate the top quintile of global income distribution. This is a development of historical importance as the top parts of the global income distribution were, in the past two centuries, populated mostly by Western European, North American and Japanese citizens. The current development is the most dramatic reshuffle in (notional) individuals’ relative income positions since the Industrial Revolution.
The importance of such development cannot be overemphasized. Even if gaps between the rich, the middle class, and the poor in advanced economies do not increase, these three national groups will belong to different parts of the global income distribution. Western distributions, reflected in global income distribution, may increasingly resemble Latin American distributions. The income gaps may not be as large, but relative global positions of different domestic classes may be substantially different.
Rich countries’ middle class. The big losers in this reshuffle will be again the middle (and lower) classes of the rich countries. This is not shown only in the so-called “elephant graph” that summarized the lack of growth among the rich countries’ middle classes between 1988 and 2008 (or 2014), but also in what was just mentioned regarding middle classes’ relative global positioning. A person in (say) Italy whose relative global position drops from the 85th to the 70th global percentile may not at first feel very much of a change, if his domestic position vis-à-vis the top stays the same. Still he or she will gradually realize that their access to some global and often positional goods (travel, type of housing, electric cars and hi-tech gadgets) becomes more difficult. As the world gets more globalized, such loss of status will be more keenly felt. Even the most attractive locations may be increasingly purchased by richer foreigners. What seems today to be a fringe phenomenon of “Venecization” is just a reflection of changing relative economic power between the countries and globalization of the world.
Europe. These developments, both through greater migration from Africa to Europe, and through the loss of relative income position of Europe compared to Asia, will influence European populations at several levels. That effect, due to the geographically different position of North America, may not be as dramatic there.
5. Meaning of global inequality
It is not immediately obvious what is the meaning of global inequality nor why lower global inequality would be advantageous. Two reasons come to mind though: first, lower between-country inequality is supposed to moderate labor flows, and second, it makes global inequality of opportunity between individuals less. A very high level of global inequality (as currently) means that life chances are heavily skewed in favor of people born in rich countries (after adjusting for education level and effort). This is not different from having high inequality of opportunity within a nation—except that the latter is politically considered problematic and there are instruments, notably through government policy, that are supposed to correct it. But on the global level, short of global government, there is no political institution that can handle inequality of opportunity.
Nostalgia. The fact that many people in America and to lesser extent in Western Europe often seem to look longingly toward the 1950s and 1960s makes sense from the point of view of their having then had much higher incomes than people in Asia and Africa. Recently, a (left-wing) author rued the times when “even a working-class Englishman could stand strong and tall” in the rest of the world. But what is not explicitly recognized is that that period of relatively high incomes of the West was, by definition, the period of relatively low incomes in the Third World, and thus the period with historically highest level of global inequality. These relative positions are unlikely to be reproduced in near future—nor would it be desirable, from the global perspective, if they were.
This first appeared on Branko's blog and was reposted with permission.
Photo by Nataliya Vaitkevich from Pexels
