Everyone Gets Ten Times Poorer and the Gini Does Not Move
Published 10/9/2026 · 3 min read · Business tools
The Gini coefficient compares a distribution with perfect equality on a scale from zero to one: five identical incomes score 0, and the set 10,000 / 20,000 / 30,000 / 50,000 / 90,000 scores 0.38. Because it is built entirely from ratios between the values, multiplying every income by the same factor leaves it untouched — the same five people at 1,000 to 9,000 also score 0.38, and so do they at 100,000 to 900,000. That scale invariance is deliberate and it is also the measure's most important limitation: it describes how unevenly a total is shared and says nothing whatsoever about whether the total is large or small. A country can grow poorer every year with a perfectly flat Gini, and it can grow richer while the index worsens.
Five incomes of 10,000 to 90,000 give a Gini of 0.38. Divide every one of them by ten and the answer is still 0.38 — the measure sees proportions and nothing else.
Two different distributions, one index
The third row of the table scores exactly the same as the second, and the two are not the same society. In the first, the poorest of the five has 10,000; in the second, 5,000 — half as much — while the middle of the distribution is tighter and the top is identical. Compressing one part of a distribution and stretching another can leave the summary unchanged, because a single number cannot encode a shape. That is why serious work on inequality reports the index alongside the shares taken by each fifth of the population, or a full Lorenz curve: those show where the movement happened, and the coefficient only shows that some happened.
Before tax or after, per person or per household
Two published Gini figures for the same country in the same year routinely differ by a large margin, and almost always because they measure different things. Market income before taxes and transfers gives one number; disposable income after them gives a considerably lower one, since that is what redistribution is for. Counting households rather than individuals changes it again, and adjusting household income for the number of people it supports changes it a third time. None of these is wrong and none is comparable with another. Before comparing two figures, find the definition each used — it is always stated, and it is usually the whole explanation of the gap.
| Incomes | Gini |
|---|---|
| 40,000 five times over | 0.00 |
| 10,000 / 20,000 / 30,000 / 50,000 / 90,000 | 0.38 |
| 5,000 / 25,000 / 35,000 / 45,000 / 90,000 | 0.38 |
Worked with our own calculator
Gini coefficient calculator
Given
- Data (values)
- 10000, 21000, 33000, 57500, 108000
Result
- Gini coefficient
- 0.405
- Gini index (%)
- 40.52%
- Count
- 5
These figures are produced by the calculator below, not typed in by hand — they are recomputed whenever the tool changes.
Run it on your own figures →Frequently asked questions
- What counts as a high Gini?
- There is no threshold in the mathematics, only ranges that different bodies describe as low, moderate or high, and those descriptions vary with the definition being used. What is comparable is a country against itself over time, and one country against another when both figures come from the same source using the same method. A single index quoted without the year, the source and whether it is measured before or after redistribution is not enough to conclude anything, and the same country will appear in different positions on two league tables built from different definitions.
- Does it work on things other than income?
- On any non-negative quantity shared among a population. It is used for wealth, which is far more concentrated than income and therefore scores much higher in the same country; for land holdings; for how unevenly customers contribute to revenue; and for how concentrated a market is among its suppliers. The interpretation carries over unchanged — zero is a perfectly even share and one is everything held by a single member — while the intuitions built on income figures do not, since what counts as a lot of concentration differs by the thing being shared.
Articles you may find interesting
All guides →Related tools
Sources
Spotted a mistake in this article?