Suppose that all people in the world are allocated only two characteristics over which they have (almost) no control: country of residence and income distribution within that country. Assume further that there is no migration. We show that more than one-half of variability in income of world population classified according to their household per capita in 1% income groups (by country) is accounted for by these two characteristics. The role of effort or luck cannot play a large role in explaining the global distribution of individual income.
He backs up his assumption of zero migration by pointing out:
Assignment to country is fate, decided at birth, for approximately 97% of the people in the world: less than 3% of the world's population lives in countries where they were not born.
And this graph from his paper is quite striking:
Here are the axes. Tom Davies, in the comments below, has it right:
"The graph (for year 2008) shows on the horizontal axis a person's position in their own country's income distribution, and on the vertical axis, a person's position in global income distribution. Thus, the poorest Americans (points 1 or 2 on the horizontal axis have incomes that put them above the 50th percentile worldwide). Note that 12% of the richest Americans belong to the global top 1%."
Here's why I got it wrong: I was looking for an uploadable version and I went to the article I linked to. It turns out that it's a different graph: it's actually more informative than the one in the Review of Economics and Statistics. But the graphs tell the same story.
I found an uploadable version of the above graph here.
Do you notice a huge contradiction between the last sentence of the abstract and the whole message of his article?