Lambert here: The rent is too damn high.
“Capitalisation of the world: global distribution of income from property, 2000-2020” [Branko Milanovic and Marco Ranaldi, London School of Economics (CC BY 4.0. )].
As global GDP has grown, the share of capital income – “unearned income” that does not require working – has increased in most countries. Branko Milanovic and Marco Ranaldi assess the scale and distribution of capital income globally, showing that capital income has expanded rapidly but remains concentrated among a small fraction of individuals at the very top. This raises important questions about the role of capital ownership in shaping global economic inequality.
Between 2000 and 2020, global GDP increased by more than 60 percent in real terms. At the same time, the share of capital income (income from property, i.e. previously called in British statistics “unearned income” or income that does not require working) in GDP increased in most countries of the world. The increasing share of capital income in GDP, as seen in national accounts data, has its counterpart in the increasing share of income from capital received by individuals and reflected in household survey data. This aspect, however, has not been extensively studied in the recent literature, especially at the global scale. How large is capital income in household incomes around the world, and how unequally is it distributed across individuals?
Our key objective is to assess the level and change in the size of capital income around the world and to examine the inequality with which it is distributed. To do so, in our working paper we use standardised nationally representative household surveys from the Luxembourg Income Study (LIS) for 43 countries, covering about 3.9 billion people in the benchmark year 2000 (64 percent of the world population), and 51 countries and 4.56 billion people (58 percent of the world population) in the benchmark year 2020. The representation of advanced economies (Europe, North America, Australia, South Korea, and Japan) and Latin America is almost complete. China and India are also included. Other parts of the world are much less represented in LIS data and thus in our sample.
Defining capital income
This exercise, however, faces several challenges, starting with the definition of capital income. The most unambiguous and narrow definition is to use cash income received from capital or property. This implies that two important sources of capital income are disregarded because of data limitations: self-employment income and imputed income from property. Income from self-employment represents a significant share of total income in less developed countries, where the informal sector is quite large. However, breaking self-employment income into the shares due to capital and labour is arbitrary without knowledge of occupation, type of employment, amount of capital employed, etc. Imputed income from capital is most obvious in the case of owner-occupied housing: people own homes for which they would have otherwise had to pay rent; that rent which, in a way, they pay to themselves is also income from property. But survey data on imputed income from owner-occupied housing are notoriously poor, incomplete, or methodologically inconsistent across surveys and countries, and in many cases such estimates are not available at all.
We therefore use three unambiguous definitions of cash income from capital; all obtained from micro (individual-level) data from the Luxembourg Income Study. The first includes interest, dividends, and rents, as grouped under the variable capital income in LIS. For a broader definition, we add income from individual funded pensions and then also income from private funded pensions. The former is, in principle, cash income from voluntary savings over the lifetime that is subject to some age and other withdrawal conditions, and the latter is practically the same except that savings out of wage income for such funded pensions are legally mandated. In both cases, income is received from investments made in various financial instruments.
Income is always expressed in household per capita amounts and in constant 2020 dollars. We use US dollars (obtained by converting national currencies into dollars at the market exchange rate) rather than dollars of equal purchasing power because capital incomes are intrinsically linked to asset prices, financial markets, and returns that are largely determined, valued, and often transacted worldwide at market exchange rates.
How is capital income globally distributed?
Table 1 shows the main results across the three definitions of capital income. Several patterns stand out. First, the inclusion of pensions significantly increases the average amount of cash income from property. At the worldwide level (population-weighted), the increase in 2020 is almost 50 percent, with capital income going up from $523 per person to $774. The increase is particularly pronounced at the top of the distribution. At the 99th percentile, the two types of pensions account in 2020 for almost as much as “classical” property-related sources such as interest, dividends, and rents. These results highlight the rising importance of private pensions as a source of income from property, particularly in rich countries.
Second, median worldwide income from property is zero, and even with the broadest cash definition of income from property, income at the 75th percentile is very small in 2020 and was zero in 2000. This indicates that most of the world population has no income from property or that such income is very small.
Table 1: Global distribution of incomes from capital according to three definitions of cash capital income
Note: all amounts are expressed in constant 2020 dollars, and income values are for the world distribution, population weighted. Ranking is by capital income per person.
Third, the implication of the last point is that income from property is extraordinarily concentrated, with Gini coefficients (regardless of the capital income definition) that are seldom seen elsewhere: about 94 to 97 Gini points which is close to the theoretical maximum of 100, obtaining when the entire income is received by one individual or one household.
Fourth, there was a substantial worldwide increase in income from property between 2000 and 2020. We focus here on the third (broadest) definition of capital income. The mean real capital income per person increased by almost 90 percent over the twenty-year period (from $413 to $774), implying an average annual growth rate of 3.2 percent, compared with an average real per capita GDP growth rate of 1.2 percent over the same period. Despite this growth, the median world capital income remained, as already noted, zero. Moreover, only the top 12 percent of the population in 2000 and the top 27 percent in 2020 had positive non-trivial cash income from capital (non-trivial is defined as at least $100 per person annually). In other words, 73 percent of the world population have zero or near-zero income from property. That percentage increases to 80 when we estimate the prevalence of positive incomes from property in the countries not included in LIS data.
Fifth, the top 1 percent (not shown in the table) received on average $26,526 in 2000 and $41,430 in 2020 from the ownership of capital: an increase of 56 percent, or an average annual rate of 2.2 percent (vs. 3.2 for the world as a whole). The share of the richest (by capital income) top 1 percent thus declined from 64 to 53 percent.
Sixth, as the last points imply, capital income has grown more slowly at the very top of the capital income distribution than at the mean, leading to a decline in global capital income inequality. The decline was driven primarily by the extraordinary growth of capital income in China, at an average per capita annual rate of 18.6 percent (to be contrasted with an equivalently defined average growth of 2.4 percent in the United States). China entered this period with very little capital income, and as its population accumulated capital it moved up the global capital income distribution, thereby reducing overall inequality. This process mirrors what has been observed since the 1990s at the level of global disposable income: rapid growth in populous Asian countries, and particularly in China, has reduced global income inequality.
Will these results remain if the household survey data discussed so far are adjusted for the well-known underestimation of capital incomes at the top? We proceed to do this next by using three types of adjustments. In the first, we take the mean value of capital income from National Accounts (NA), contrast it with the mean value of capital income obtained from household surveys (HS), and then allocate the gap proportionally to all individuals; in the second we allocate the gap only to the top 5 percent of national capital income distributions, and in the third, we add the estimated returns to the wealth of billionaires worldwide, as listed by Forbes magazine in the years 2001 and 2020. In all three adjustments, levels of inequality are higher than in no-adjustment case. with Ginis coming close to the theoretically maximum value, but, alike in the no-adjustment case, there is a decrease of inequality between 2000 and 2020.
The role of capital ownership in global inequality
The global distribution of capital income is characterised by two striking features. First, capital income is absent or negligible for most of the world population: even under the broadest definition, the median remains zero and a large share of individuals receive little or no income from property. Second, capital income is extraordinarily concentrated among a small fraction of individuals at the top of the global distribution. At the same time, the rapid growth of capital incomes in large emerging economies—most notably China—has contributed to a modest decline in global capital income inequality over the past two decades. Taken together, these results highlight a world in which capital income has expanded rapidly but remains highly unevenly distributed, raising important questions about the role of capital ownership in shaping the evolution of global economic inequality.