If you've ever seen a chart or map or animated gif map or something about economic inequality, chances are it uses something called the Gini coefficient. It's the standard measurement, but it's also way too complicated to explain, as the Center for Global Development's Alex Cobham illustrates in this video:
Cobham and other development experts have come to prefer something called the Palma ratio. The ratio is named after Gabriel Palma, an economist at the University of Cambridge who noticed that the 40th through 90th percentiles of a country's income distribution tend to always get the same share of income. The amount going to the bottom 40 and top 10 percentiles varies quite a bit, but the middle doesn't change much if you look across different years, countries and levels of economic development. For example, here's the 2010 breakdown. Note that the green columns, showing the middle 40th-90th percentiles, don't change much at all, even when the tails are changing dramatically:
That lead to a great idea: How about, instead of doing the complicated gymnastics needed to calculate the Gini, we just divide the top 10 percent's share of income by the bottom 40 percent's? That's the Palma ratio. And it's much easier to grok than the Gini. A Palma ratio of 5 means the country in question has an top 10 percent that makes 5 times as much as its bottom 40 percent; a Gini ratio of 0.5 means, well, it's hard to explain.
Despite being easier to explain, the Palma aligns nearly perfectly with the Gini. Cobham and Andy Sumner of King's College London ran a regression and found that the two components of the Palma (the top 10 percent and bottom 40 percent's shares of income) explain 100 percent of variation in countries' Gini ratios. But that's at least partly because the World Bank uses limited data to compute Gini coefficients. In situations where there are better data, the two can diverge, and in a way that makes the Palma look much more attractive as a measure. The Gini overweights changes in inequality that happen in the middle of the income distribution, which is exactly not what we should be looking at, given how stable the middle is.
This is particularly troublesome when you're using the Gini to measure how much a policy increases or decreases inequality. "The difficulty with using Gini for that is that you'll get unclear answers," Cobham says. "If the policy is good for the top 10 percent and bad for the bottom 40 percent but progressive in the middle, then the Gini could show a regressive measure as progressive."
How do various countries stack up on the Palma? The table to the right, from a Danish Institute of International Studies (DIIS) report, provides a good sampling of scores. Developed countries like the United States and United Kingdom tend to have ratios below 2. Especially equal countries, like Denmark or Japan, have ratios below one. But a number of poor countries are in that zone, too. India's ratio is below the United States, for instance, as is Tanzania's. And Burkina Faso and China aren't too far above us.
It's middle-income countries where inequality has really gotten out of hand. Brazil's Palma is above 3, and South Africa's is above 7. They've had plenty of economic growth, but it's been unequally distributed, and their welfare states are not yet at a point where that can be counteracted through things like public education, transfer payments, progressive taxation and so forth.
But the most shocking number from that report is the global Palma: the ratio of the top 10 percent's share of world income to the bottom 40 percent's share, taking every country into account. The ratio, DIIS estimates, is about 32. Those of us in the richest 10 percent globally make 32 times more than the bottom 40 percent.
It's another reminder that, while extreme poverty in the United States is very real, the biggest inequalities, by far, are at the global level. "The political instruments for reducing income inequality between the richest 10 per cent and the poorest 40 per cent of the world’s population do not exist," author Lars Engberg-Pedersen notes. "Progressive taxation, provision of social security, etc. are country-level instruments, and official development assistance comes no way near addressing global inequality."
It's nowhere close to enough, but I'd be remiss without noting that giving money directly to poor people in Kenya is one way individuals can chip away ever-so-slightly at the problem.