Income Inequality — Security Framework

Security Framework

Begin by identifying whether a figure concerns income, consumption or wealth and whether the unit is a person, adult or household. Pre-tax income and disposable income also have different scopes. A comparison that changes these definitions may not describe a real distributional trend.

For four incomes of 10, 20, 30 and 40, the mean and median are both 25. Raising the highest to 140 makes the mean 50 while the median remains 25. The average rises even though the other three incomes are unchanged. The example shows why a mean needs distributional context.

The Lorenz curve relates cumulative population shares to income shares in the standard non-negative case. The Gini summarises relative inequality and may be shown as a zero-to-one coefficient or zero-to-100 index. Thus 0.30 and 30 can describe the same value. It is neither a poverty headcount nor a top-income share.

Doubling all incomes preserves relative shares and the Gini while doubling absolute gaps. Under a fixed illustrative poverty line, some people may still cross out of poverty. Empirical figures need equally careful labels: the World Bank's October 2026 brief gives a consumption Gini of 23.0 for 2023, while the WID study's 22.6 per cent figure is an estimated top one per cent income share for 2022–23. They are different statistics.