Here are some ideas for things to do with the above app:
- Do you know what the US Gini coefficient of income is? If not, play around with different income distributions for a while and try to guess. You can check here.
-
So I don’t really care about income inequality per se, I care about equality of consumption. That is, I mostly care about whether people get to consume similar amounts of societal resources, as opposed to what their before-tax income is. The Gini coefficient isn’t quite a fair measure of overall equality, because it doesn’t include taxes (which are progressive), government transfers (which are mostly targeted towards lower-income people), or the amount that the government spends on public goods which are shared equally by everyone (and therefore increase equality). It also doesn’t include the cost-of-living differences between different places–higher income people live in states and suburbs where prices are higher.
Sometimes people directly try to measure consumption inequality, and they get lower numbers than the income inequality numbers–this report finds that US consumption Gini coefficient is about 0.1 lower than the US income Gini coefficient.
-
- What’s the Gini coefficient of a population where there are two equally sized population groups, where one group has much higher income than the other?
- Suppose I start out with a very equal population, with an income of 2 each. What happens to the Gini coefficient as I add many higher-income people?
- Compare the Lorenz curves for the populations [10, 20, 35, 35] and [15, 15, 30, 40]. Their Gini coefficients are the same, but I think that this is a weakness of the index–I think the latter is a more equitable income distribution.