Where do the poorest Americans fall in the distribution of income among all people ever born? : The Independent Review : The Independent Institute

Data from historical and contemporary sources suggests that the poorest Americans — those in the bottom five percent — have higher incomes than about 95 percent of all humans who have ever lived.


Founding Editor of The Independent Review Robert Higgs reminds us that “[i]income inequality is a statistical artifact. . . . The aggregate of the measurement is arbitrary: why, for example, should inequality be measured for the entire US population rather than for the population of the city or state in which one lives, the the entire North American population (including Mexico), the entire Western population The population of the hemisphere, or the entire world population? ” (2014).

He could have added “or even the entire world population throughout history.” In fact, measuring inequality in this way can be very illuminating. Therefore, in the calculations below, I try to put the incomes of today’s poor Americans into the the largest historical perspective.

Branko Milanovic (2010) provides data showing where people in different income brackets (twentieths of the distribution) in several countries fall within the global income distribution. He finds that the income of the bottom 5% of Americans puts them at the 68th percentile of the global income distribution in 2010. Poor Americans are quite wealthy by today’s global standards. However, they are much richer compared to everything people who have ever lived.

Virtually everyone in the premodern era lived at a small multiple of the subsistence level. Although subsistence cannot be measured precisely, Milanovic, Lindert and Williamson (2007) estimate (following Angus Maddison 2001) that subsistence is equivalent to $400 per year in 1990 dollars, which equates to approximately $855 a year today. Jeffrey Williamson (2009, 29-30) provides estimates of the relationship between average income and subsistence income for a wide range of countries over a period of almost two thousand years. In the years before the onset of modern economic growth, these ratios averaged around 3.1 – falling below 2 in places like medieval Byzantium (1000 AD), Mughal India (1750), Qing China (1880) and colonial Kenya (1927), while slightly exceeding 2 in ancient Rome and exceeding 5 in a handful of places, including England (1759) and Holland (1732). Although a few places still have very low average income to subsistence income ratios – Tanzania and the Democratic Republic of Congo fell below 2 at the turn of the millennium – the ratio is now over 60 in places like Great Britain. Britain, France, Germany, Italy, Japan and the United States (where it is above 80).

The annual US Census Social and Economic Supplement reports incomes of $0 and $800 for the bottom two percentiles in 2020 (DQYDJ 2021), which is obviously incorrect.[1] Income estimates are well below the consumption levels of the American poor, as shown by data from the Bureau of Labor Statistics (2020). The BLS Consumer Expenditure Survey calculates that households with incomes in the bottom 10.4% (those earning less than $15,000 in 2020) earned 0.93% of total income in the United States. , while their consumption accounted for 4.7% of the total. The gap is due to transfers from government and non-government sources (such as family members), undeclared income in the formal and informal economy, and other factors. The consumption of poor Americans is about five times (5.065) their income level, although this ratio is likely an even higher multiple for the poorest of the poor. Accordingly, in comparing the lower breakdown (5%) of modern Americans to everyone who has ever lived, I use the consumption estimate, although other adjustments can be made. In 2020, the average household income in the bottom breakdown was $4,300 per year, so annual consumption was closer to $21,800. There are on average 1.7 people in these households (referred to as “consumption units” in the Consumer Expenditure Survey report), which means that consumption per person averaged about 12,820 $.

Max Roser (nd, using data from Van Zanden et al. 2014) estimates that of the 1.09 billion people in the world living in 1820, about 30.5 million had incomes above $12,800 per year ( in today’s dollars), which places today’s bottom-dwelling Americans somewhat above the 97th percentile (97.2) of the 1820 population. Angus Maddison (2001, 28) estimates that the average income levels of AD 1 and AD 1000 were about one-third lower than in 1800. If all income estimates given by Roser are reduced by one-third, this implies that Americans today the lower twentieth would be a little above the 98th percentile (98.1) among people living in the premodern period.

Kaneda and Haub (2020) estimate that approximately 94.4 billion of the 116.8 billion humans who have ever lived were born in 1650, and 101.6 billion were born in 1850.

Using these figures (see Table 1), approximately 1.8 billion of people born in 1650 were wealthier than those of the poorest five percent of Americans living today, and about 200 million of those born between 1650 and 1850 were wealthier than the lower ventiles of today.

There are currently about 7.9 billion people living, and Milanovic estimates that about 32% of today’s population have incomes above the bottom ventile of the United States, so about 2.53 billion people around the world have higher incomes than the poorest 5% of Americans. Similar calculations using Roser’s estimates place modern Americans in the bottom 80th percentile of the world in the mid-20th century. There are approximately 7.3 billion people born after 1850 who are no longer living. If modern bottom-breaking Americans have incomes at the 80th percentile among them, that’s 1.46 billion of the latter group who had incomes higher than those of the poorest 5% of Americans today.

Surely there are measurement errors here, but I’ve tried to be on the safe side, such as ignoring the bias of the consumer price index, which underestimates the value of today’s dollars by compared to the 1990 dollars used by Roser.[2] I assumed that consumption was equal to income before the advent of the welfare state in recent times.

Table 1 summarizes these results. Adding the four groups (2.53 billion alive + 1.46 billion born after 1850 but no longer living + 200 million between 1650 and 1850 + 1.8 billion born before 1650), approximately 5.99 billion people had incomes higher than those of the poorest 5% of Americans. today.

In other words, the poorest Americans have higher incomes than about 95% of all humans who have ever lived.

This is a profound and unappreciated point: the poorest among us today are very well off economically by the standards of all of world history. They live lives that surpass even the aspirations of most people who have ever lived. Virtually everyone living in the United States today is an economic winner. Will any of us speak out against this inequality?


[1] Leeson, Hardy, and Suarez (2022) report that beggars in Washington, D.C. subway stations earn about $6.10 per hour.

[2] Roser’s numbers are in 1990 dollars, which I converted to 2020 dollars using the Consumer Price Index, which estimates prices will be 2.14 times higher in 2020. This ignores the upward bias of the CPI due to substitution bias, quality change bias, new goods biases and other biases. See, for example, Hausman 2003. Removing these biases would reduce the ratio from 2.14 to about 1.67 by my rough calculations.


Bureau of Labor Statistics. 2020. Table 1203: Pre-tax income: Share of overall annual expenses and sources of income, Consumer Expenditure Survey, 2019.

DQYDJ. 2021. Average, median, top 1%, and all household income percentiles in the United States in 2020.

Hausmann, Jerry. 2003. Sources of Bias and Solutions to Bias in the Consumer Price Index.
Economic Outlook Journal 17, no. 1:23–44.

Higgs, Robert. 2014. Income inequality is a statistical artifact. Lighthouse.

Kaneda, Toshiko and Carl Haub. 2020. How many people have ever lived on Earth? Bureau of Demographic Research.

Leeson, Peter T., R. August Hardy, and Paola A. Suarez. 2022. Hobo Economicus. economic review, coming.

Maddison, Angus. 2001. The World Economy: A Millennial Perspective. Paris: OECD.

Milanovic, Branko. 2010. The haves and the have-nots: a brief and idiosyncratic history of global inequality. New York: Basic Books.

Milanovic, Branko, Peter H. Lindert, and Jeffrey G. Williamson. 2007. Measuring Ancient Inequalities. National Bureau of Economic Research Working Paper no. 13550.

Rosar, Max. nd Our world in data. World distribution of income, 1820 to 2000.

Van Zanden, Jan Luiten, Joerg Baten, Marco Mira d’Ercole, Auke Rijpma, Conal Smith and Marcel Timmer. 2014. How was life? Global well-being since 1820. Paris: OECD.

Williamson, Jeffrey G. 2009. History without proof: Latin American inequalities since 1491. National Bureau of Economic Research Working Paper 14766.

Previous Development Applications - City of Burlington
Next North Oxford residents want post office in old bank building