The Gini Index - how does your inequality measure up?

All my life I've thought that Gini was just a refreshing bitter lemon beverage now only available in France. However today I read An Open Letter to George Bush (thank you Chicken or Beef? and learned that Gini is also the name of a coefficient or index used to measure a population's income inequality. According to Wiki its named after Italian statistician Corrado Gini.

In a population with perfect equality everyone has the same income and the Gini coefficient is 0. In a population with perfect inequality of income one person has all the income and everyone else has none. Naturally most countries fall somewhere inbetween but the general rule is that when the rich get richer and the poor get poorer, income inequality measured by the Gini coefficient rises. The Gini index is just the Gini coefficient multiplied by 100, so it goes from 0 to 100.

The open letter to George Bush makes the point that our Gini index "is far higher in the United States than in any other developed country and is continuing to move upward". It also implies that our level of income inequality may be approaching a level that is extreme "which can be socially corrosive and economically dysfunctional." Furthermore it claims "we believe your tax policy has exacerbated the problem of inequality in the United States." Having blogged on the topic of income inequality before this is naturally of great interest to me. So I decided to dig up some figures and find out what the truth is.

Now it took me a long time to dig up Gini figures for the US that include Bush II years, all easily located sources stop at 2001. However I eventually dug up the document "Income, Poverty, and Health Insurance Coverage in the United States: 2003" by the US Census Bureau. This document includes historical figures for the Gini index from 1967 through to 2003 and jolly interesting it is too. Plotted on a graph it looks like this:

The huge increase between 1992 and 1993, just at the end of Bush Snr's term of office coincided with a change in methods used to collect income data (see the footnotes for this chart). Also the figures in the 2003 document are apparently for pre-tax income, a conclusion that I drew by comparison with other tables of Gini data (see this pre and post-tax Gini chart). As I've mentioned before while pre-tax inequalities are bad, post tax ones tend to be even worse due to the steady elimination of progressive income taxation by successive Republican governments. Unfortunately I haven't been able to dig up any post-tax (disposable income) based Gini figures which would probably be the most interesting ones to evaluate what if any role the Bush tax cuts have had on income distribution.

I did managed to locate an analysis for 2002 that shows computations of the Gini coefficient when various adjustments to total income are made including taxes, deductions, benefits paid, etc. Table 2 of this document indicates that in 2002 the Gini coefficient starts out at 0.448 when considering gross income not including capital gains. After adjusting for things like capital gains, taxes, social security payments and taxes, federal and state taxes the Gini ratio rises to 0.487. Then it goes on to adjust for the value of received benefits like medicaid, medicare, free school lunches, and the benefit from an increase in home equity. These factors all serve to decrease the Gini coefficient to a final value of 0.40.

So clearly calculating the Gini coefficient is complicated and its going to take some hard work to produce concrete figures that indicate what exactly government policy has done to the inequality of income distribution. However, variances for calculation method aside, what you should do is take a look at Gini figures world wide. Basically is 0.4 or 0.5 or whatever good, bad or indifferent? Kim Scipes did just this in his article "International Income Inequality: Whither the United States?", so did the World Bank, and even the CIA. You can even get a map of global Gini values.

Issues with absolute measurement aside (most figures are actually based on gross income since that's the only data available) when you look at these tables you see something interesting. Relative to the rest of the world the USA is basically doing about as well as your average eastern European country, but better than your average Latin American country. Compared to the rest of the "developed world" the USA is usually lagging far behind at the back of the pack. Indeed based on distribution of income on a per family basis, Scipes concludes that we're doing about as well as sub-Saharan Africa and only Latin America and the Caribbean are doing worse than us at divvying up the income. Indeed if you consider the CIA world factbook figures then out of the 112 nations listed the USA comes 69th right inbetween Ghana and Turkmenistan. And that's using the 1997 Gini index value of 40.8. If we're actually at 45 or worse now then we'd be 84th position between Kenya and Costa Rica.

Clearly, if our income distribution inequality is getting worse then either we are entering into some new and undiscovered state of developed country with solid economy and happy populace, or we are actually pushing the envelope of stability with inequality and increasingly disgruntled populace. I know which I would choose and I think its only a matter of time before the cracks that have been plastered over by the veneer of media whitewash will begin tearing wide open. Clearly given the still virtually deadlocked state of our political system we haven't reached that point yet, but perhaps as many are predicting, four more years of Bushanomics will be precisely that impitous to produce that "socially corrosive and economically dysfunctional" state that will lead to, well, rioting in the streets Latin America style...