22 August 2006

America's Third World Economy

Daily Kos diarist 7November amidst a larger argument about the estate tax throws in an interesting tidbit about where the United States ranks of the Gini index, a widely accepted measure of income inequality.

The U.S. has high income inequality as measured by the Gini index.

Out of 124 countries, the U.S. ranks 92nd most equal (meaning it has more income inequality than most countries) with a 46.6 Gini index. This has increased significantly since 1970:

Gini coefficients for the United States at various times, according to the US Census Bureau:

1970: 0.394
1980: 0.403
1990: 0.428
2000: 0.462 [1]

[1] Note that the calculation of the index for the United States was changed in 1992, resulting in an upwards shift of about 0.02 in the coefficient.


Thus, on a comparable basis with prior years, the Gini coefficient for the U.S. in 2000 was about .442, rather than 0.462. In contrast, France, for example, has become a much more equal society in the same time period. The Gini index the U.S. had in 1970 would rank as #67 today, instead of the actually current ranking of #92.

Japan is second most equal. All of the other countries in the top 15 for equality are European (including the former Soviet Union), although there are a few exceptions in the top quarter (i.e. the top 31 countries). South Korea is #26, because, like Japan, it has a modern economy. Mongolia is #22, partially because of its former status as a Soviet satellite and partially due to likely oddities in the data that don't fully reflect the inequality in its society.

Rwanda (#16) (the tenth income percentile is around $7,500 per year), Ethiopia (#20) (the tenth income percentile is around $5,400 per year) and Bangladesh (#27) (the tenth income percentile is around $13,600 per year) make the list presumably because even the well to do are struggling in those countries.

Not all Western economies are in the top quarter, but almost all are in the top half (i.e. through #61). France is #34, Canada is #36, Switzerland in #37, Australia is #46, Greece is #48, Israel is #49, Ireland is #50, the U.K. is #51, Italy is #52, and New Zealand is #53.

The only reasonably developed countries outside the top half are Portugal at #65, Singapore is #77 (a particularly notable ranking in light of the fact that there is almost no privately owned residential real estate in Singapore -- everyone rents), and Hong Kong (ranked as a country despite being part of China for this purpose) at #83.

Who is in the American peer group, at the low end of the third quartile and in the bottom quartile?

82 Thailand 43.2
83 Hong Kong 43.4
84 Ecuador 43.7
85 Uruguay 44.6
86 Cameroon 44.6
87 Cote d'Ivoire 44.6
88 People's Republic of China (mainland only) 44.7
89 Bolivia 44.7
90 Philippines 46.1
91 Costa Rica 46.5
92 United States 46.6
93 Guinea-Bissau 47
94 Dominican Republic 47.4
95 Madagascar 47.5
96 The Gambia 47.5
97 Burkina Faso 48.2
98 Venezula 49.1
99 Malaysia 49.2
100 Peru 49.8
101 Malawi 50.3
102 Mali 50.5
. . .
109 Mexico 54.6
. . .
116 South Africa 57.8
117 Brazil 59.3
. . .
124 Namibia 70.7

Are the rankings hiding a lot of bunched together Gini index scores? Judge for yourself. The most equal, Denmark, has a Gini index of 24.7, Spain which is the last country in the top quartile is at 32.5, Armenia which rounds out the second quarter is at 37.9, Guinea-Bisseau which rounds out the third quarter is at 47. Namibia is worst at 70.7.

Limitations of Gini Index Data

The Gini index probably overrates income inequality in the United States relative to other countries as shown by two different measures discussed below where the U.S. ranks #79 and #77 respectively, as compared to #92 for the Gini index. Some likely factors (from the Wikipedia article for the Gini index) behind this disparity include:

The Gini coefficient measured for a large geographically diverse country will generally result in a much higher coefficient than each of its regions has individually. . . .

The meaning of the Gini coefficient decreases as the data become less accurate. . . .

Economies with similar incomes and Gini coefficients can still have very different income distributions. This is because the Lorenz curves can have different shapes and yet still yield the same Gini coefficient. As an extreme example, an economy where half the households have no income, and the other half share income equally has a Gini coefficient of ½; but an economy with complete income equality, except for one wealthy household that has half the total income, also has a Gini coefficient of ½.

It is claimed that the Gini coefficient is more sensitive to the income of the middle classes than to that of the extremes. . . .

[T]he Gini coefficient is influenced by the granularity of the measurements. For example, five 20% quantiles (low granularity) will yield a lower Gini coefficient than twenty 5% quantiles (high granularity) taken from the same distribution.


The U.S. is more geographically large and diverse than many of the countries to which it is compared. For example, even if the South, the Northeast, the Midwest, and West were individually had high levels of income equality, the regional variation between incomes in the South and those in the Northeast, for example, would contribute to a higher Gini index for the United States.

The very top of the U.S. income distribution has a particularly large share of total U.S. income, even though the bottom 96% or so, have income distributions more equal than other countries with Gini indexes similar to the United States.

The U.S. also has very fine grained and accurate data available, something probably not true in the outlier countries identified below.

Also, most concern about income inequality is related to concerns about the poor (who may suffer needlessly for want of affordable social programs) and the rich (who may be exploiting the system for their own unjustified benefit), not to the extent of modest differences in the middle.

90-10 ratios

There are other ways than the Gini index to judge income inequality. One crude measure compares the ratio of the 90th percentile of income to the 10th percentile of income. In the United States this ratio is 15.9 to 1 which ranks #79 in the world out of 124 countries ranked by this measure.

A few countries with higher Gini indexes than the U.S. have higher income equality than the U.S. by this measure (with Gini index ranking in parenthesis). Mongolia (#22) has a ratio of 17.8 to 1. Burundi (#39) has a ratio of 19.3 to 1. Singapore (#77) has a 17.7 to 1 ratio. Iran (#79) has a 17.2 ratio.

All the countries ranked #83 to #124 by Gini index above, have a higher ratio than the United States, except Cameroon which has a 15.7 ratio. Still a ranking by this ratio leaves the United States in the middle of the third quartile in any case.

80-20 ratio

A similar measure compares the ratio of incomes for those at the 80th percentile of income to those at the 20th percentile of income. In the United States this ratio is 8.4 which ranks #77 by this measure.

This also shuffles results compared to the Gini index. Mongolia (#22) with a 9.1 ratio, Burundi (#39) with a 9.5 ratio, Singapore (#77) with 9.7 ratio, and Iran (#79) with a 9.7 ratio, are again outliers that fall below the United States by this measure. So does every country ranked #81 or lower by Gini index, except Thailand (#82)which has an 8.3 ratio.

Peer Comparison By Ratio Measures

By income ratio measures (90-10 and 80-20) on which the United States ranks 15.9 and 8.4 respectively, Japan has the least income inequality. Its 90-10 ratio is 4.5, and its 80-20 ratio is 3.4. Namibia remains worst in the world by these measures as well with a 90-10 ratio of 128.8 and and 80-20 ratio of 56.1.

Some of the other notable ratios are set forth below (after their Gini index ranks).

1 Denmark 8.1/4.3
2 Japan 4.5/3.4
3 Sweden 6.2/4
4 Belgium 7.8/4.5
5 Czech Republic 5.2/3.5
6 Norway 6.1/3.5
10 Finland 5.6/3.8
14 Germany 6.9/4.3
18 Ukraine 6.4/4.3
19 Austria 7.6/4.7
24 Netherlands 9.2/5.1
25 Russia 7.1/4.8
26 South Korea 7.8/4.7
31 Spain 9/5.4
32 India 7.3/4.9
34 France 9.1/5.6
36 Canada 10.1/5.8
37 Switerland 9.9/5.8
42 Poland 8.6/5.5
46 Australia 12.5/7
48 Greece 10/6.2
49 Israel 11.7/6.4
50 Ireland 9.7/6.1
51 U.K. 13.8/7.2
52 Italy 11.6/6.5
53 New Zealand 12.5/6.8
65 Portugal 15/8
70 Turkey 13.3/7.7
73 Cambodia 11.6/6.9
74 Turkmenistan
75 Ghana 14.1/8.4
76 Senegal 12.8/7.5
77 Singapore 17.7/9.7
78 Kenya 13.6/8.2
79 Iran 17.2/9.7
81 Nicaragua 15.5/8.8
82 Thailand 13.4/8.3
83 Hong Kong 17.8/9.7
88 China 18.4/10.7
90 Phillipines 16.5/9.7
92 United States 15.9/8.4
99 Malaysia 22.1/12.4
104 Nigeria 24.9/12.8
109 Mexico 45/19.3
112 Zimababwe 22/12
116 South Africa 33.1/17.9
117 Brazil 68/26.4

Conclusion

The bottom line is that the United States, by any measure, has level of income inequality associated with a developing country, not that of a mature industrialized economy.

Lassiez-faire economists who argue that income inequality is a necessary component of economic success for a nation as a whole are simply wrong in the face of the empirical facts. Highly productive societies tend to have low levels of income inequality, while most societies that struggle to provide for themselves either have high levels of income inequality, or near universal gross impoverishment.

3 comments:

Robert said...

The bottom line is that the United States, by any measure, has level of income inequality associated with a developing country, not that of a mature industrialized economy.

Which would seem to tell us more about the "associations" rather than the data, since we are a mature industrialized economy; the strongest in the world, in fact.

Highly productive societies tend to have low levels of income inequality, while most societies that struggle to provide for themselves either have high levels of income inequality, or near universal gross impoverishment.

...except for the MOST productive society on earth, ours, which has a high level of income inequality. Data > Theory.

Andrew Oh-Willeke said...

The United States is one of a whole lot of highly productive societies on the planet, and it isn't at all obvious that we are the best off.

The United States is certainly not anywhere near the fastest growing economies on the planet. When the U.S. economy was growing at a rapid pace, the inequalities in income weren't nearly so great.

The United States is seeing its manufacturing base crumble. Income inequality has skyrocketed since about 1970, a time period which has coincided with the collapse of the American manufacturing economy, and a period of rising trade deficits and sustained large government borrowing.

Also, the high GNP of the United States is being sustained by the most hours worked per capita of any country on Earth. It isn't at all obvious that our per hour productivity is the highest on the planet as a result. Other development nations, even Japan, manage to live relatively affluent lives with more time away from work.

And, it also isn't at all clear that we aren't simply living on borrowed time and old capital. Great empires take time to decline. It is entirely possible that our third world income inequalities are sowing the seeds of long term economic anemia.

Anonymous said...

If the Gini coefficient is based on household income, then some inqualities, for example, that a woman may not work and therefore have no income may not show up; furthermore, in a country like the US, in which many women work, albeit in many cases not earning the same as men for various reasons, a Gini coefficient based on individual income would reveal more apparent inequality than there actually is. The point being, it's a pretty crude measure which is probably revealing different things in different countries.