## Measuring inequality

The standard measure of inequality is the Gini coefficient. It is equal to one if all income goes to one person; if one woman owned everyone in an economy, fed them what she liked and made them work according to her will, if all income belonged to her and everyone else was her slave, the Gini coefficient in her monarchy would be one. If, on the other hand, everyone got exactly the same income, the Gini coefficient would be zero.

If people are ranged from the poorest to the richest, and if the cumulative proportion of the population is plotted against the cumulative proportion of income it commands, one gets a Lorenz curve. If everyone got the same income, the proportion of income and the proportion of population would be the same, and a graph of one against the other would be a 45-degree straight line. If all the income went to one person, the graph would hug the x-axis till it reached that last person, and then rise vertically. The ratio of the area between the Lorenz curve and the 45-degree line to the total area under the line is the Gini coefficient.

The Organization for Economic Cooperation and Development has recently been looking at trends in income inequality. The countries with least unequal incomes amongst those for which OECD got data (mostly for 2008) are the Czech Republic and Scandinavian countries — Norway, Sweden, Denmark and Finland — all with Gini coefficients of 0.25 or less. Czech Republic was communist; the Scandinavians have extremely progressive taxation, and their people are honest in paying taxes. The countries with most unequal income are Mexico, Turkey, the US and Israel, with Gini coefficients of 0.48, 0.42 and 0.38 and 0.38 respectively. Most OECD countries cluster in the range of 0.25 to 0.33.

Take a look (I find, by Googling, that Gini coefficient of India is about 0.325 at present and it is increasing).

### One Response to “Measuring inequality”

1. swissecon Says:

Thanks for posting this.

I would, however, add that we should be careful with these inequality statistics. There are many details that might lead to severe misunderstandings.

I have written some of them down a few months ago:

http://swissecon.wordpress.com/2011/07/08/714/