What Is the Gini Coefficient? How Income Inequality Is Measured

Search Fintrev

Search our guides, reviews and glossary.

The Gini coefficient is a single number between 0 and 1 that summarizes how unevenly income (or wealth) is distributed across a population. A value of 0 means perfect equality: everyone has the same income. A value of 1 means perfect inequality: one person has all the income and everyone else has none. When the same measure is multiplied by 100 and reported on a 0 to 100 scale, it is usually called the Gini index, and some sources write it as a percentage. A Gini coefficient of 0.40, a Gini index of 40 and a reading of 40% describe the same distribution.

What the Gini Coefficient Measures

Italian statistician Corrado Gini introduced the measure in 1912. It answers one question: if you lined up everyone from the lowest income to the highest, how far does the actual distribution depart from a world where everyone earns the same?

Because it is a ratio, the Gini is scale invariant. If every income doubled, each person’s share of total income would stay the same and the Gini would not change. That makes it comparable across countries with different currencies and price levels, but it says nothing about living standards. Measures such as GDP per capita or incomes adjusted for purchasing power parity answer that separate question.

The Gini can be calculated for any distribution: annual income, disposable household income, consumption spending or net worth. An unqualified national Gini almost always refers to income, because income data are easier to collect than wealth data.

How the Gini Coefficient Works: The Lorenz Curve

The standard way to picture the Gini is the Lorenz curve. The horizontal axis shows the cumulative share of the population, ranked from poorest to richest. The vertical axis shows the cumulative share of income those people receive.

• If income were perfectly equal, the bottom 20% would earn 20% of income, the bottom 50% would earn 50%, and so on. That produces a straight 45 degree diagonal called the line of equality.
• In any real economy, the bottom groups earn less than their population share, so the Lorenz curve sags below the diagonal. The more it sags, the greater the inequality.

Label the area between the line of equality and the Lorenz curve A, and the area under the Lorenz curve B. The total area under the line of equality (A + B) is 0.5, because it is half of a 1 by 1 square. The Gini coefficient is:

G = A / (A + B) = 2A = 1 − 2B

An equivalent formula works directly from individual incomes, without drawing a curve. It is based on the average absolute difference between every pair of incomes:

G = (Σᵢ Σⱼ |xᵢ − xⱼ|) / (2n²μ)

where x is each income, n is the number of people and μ is the mean income. This version shows what the Gini captures: how far apart incomes are from one another on average, relative to the average income.

How to Read a Gini Reading

There is no official cutoff for “high” or “low,” but as a rough guide:

• Closer to 0 (or 0 on the index): incomes are bunched together.
• Around 0.30 to 0.40: moderate dispersion, typical of many high-income economies.
• 0.50 and above: income is heavily concentrated. A national Gini index of 50 sits at the midpoint of the scale and indicates a markedly uneven distribution.
• Closer to 1: a small group receives nearly all income.

Small changes matter. Because the Gini compresses an entire distribution into one number, a move of one or two index points over several years can reflect a sizable shift in who receives income.

Worked Example: Calculating a Gini Coefficient

Hypothetical example for illustration. Take an economy of five people with annual incomes of $10,000, $20,000, $30,000, $40,000 and $100,000. Total income is $200,000 and mean income is $40,000.

Step 1: Rank incomes and compute cumulative shares. Each person is 20% of the population.

Step 2: Find the area under the Lorenz curve (B). Treat each 20% slice as a trapezoid: width 0.2, average height equal to the mean of the starting and ending cumulative shares.

B = 0.2 × [(0 + 0.05) + (0.05 + 0.15) + (0.15 + 0.30) + (0.30 + 0.50) + (0.50 + 1.00)] / 2 B = 0.1 × 3.00 = 0.30

Step 3: Apply the formula. G = 1 − 2B = 1 − 0.60 = 0.40, or a Gini index of 40.

The pairwise formula gives the same answer. The absolute differences between all ten unique pairs of incomes add up to $400,000, so the sum over all ordered pairs is $800,000. Dividing by 2 × 5² × $40,000 = $2,000,000 gives 0.40.

Small groups have a lower ceiling. With five people, the maximum Gini under the discrete formula is (n − 1)/n = 0.80, reached when one person has everything. For national populations of millions this ceiling is effectively 1, which is why Gini values for small samples or small firms are not directly comparable to national figures.

Why the Gini Coefficient Is Used

The Gini is the most widely reported inequality statistic because it is compact, bounded and comparable. Governments, the World Bank, the OECD and national statistical agencies such as the U.S. Census Bureau publish it to:

• Compare countries. World Bank estimates have placed South Africa among the most unequal countries, with a Gini index of about 63, and Norway among the most equal, at about 23. The United States, at roughly 40 in World Bank figures, is high for a developed economy.
• Track change over time. The global Gini rose from about 0.50 in 1820 to about 0.657 by 1980 and 1992, and a World Inequality Lab estimate puts it at about 0.67 for 2020. Economists estimate that COVID-19 added about 0.5 points between 2019 and 2020, the largest one-year increase since World War II.
• Measure the effect of policy. Comparing the Gini before and after taxes and government transfers shows how much redistribution narrows the gap.
• Serve as a development indicator. Alongside poverty rates and output measures, the Gini indicates whether growth is widely shared. Many of the lowest readings are in wealthier European countries and some of the highest are in poorer ones, but the link with GDP per capita has shifted over different historical periods rather than following a stable pattern.

Limitations of the Gini Coefficient

Different distributions can share the same Gini. Hypothetical example: in Economy X, four people earn $0, $25,000, $25,000 and $50,000. In Economy Y, three people earn $12,500 each and one earns $62,500. Both have a Gini of exactly 0.375, yet in X the poorest person has nothing, while in Y nobody is without income and the inequality comes from one top earner. Their Lorenz curves cross, and the single number cannot tell you which picture applies.

It is more sensitive to some parts of the distribution than others. The Gini responds most to changes in the middle of the distribution and relatively less to changes at the very top or very bottom. Analysts who care specifically about the top 1% or about extreme poverty often use top income shares or ratios such as the Palma ratio (the top 10% share divided by the bottom 40% share).

Data quality drives the result. Informal and shadow economic activity tends to be a larger share of output in developing countries and among lower earners, and unrecorded income at the bottom makes measured inequality look worse than true inequality. At the top, tax havens make wealth especially hard to observe, which tends to understate wealth concentration.

Demographics shape the number. A Lorenz curve ranks people only by income, not by age, region, gender or race. A country with a large retired population, or many students, will show a higher Gini even if lifetime incomes are fairly similar, because people at different life stages earn very different amounts in a given year.

It ignores the level of income. Turkey and the United States have had similar Gini coefficients in OECD data even though Turkey’s GDP per person is far lower.

Methodology varies between sources. Figures for the same country can differ because one uses household income and another individual income, or because only one adjusts for household size (equivalized income). Comparing a Gini from one database with a Gini from another can mislead.

Distinctions Readers Often Confuse

Gini coefficient vs. Lorenz curve. The Lorenz curve shows the full distribution. The Gini is one number calculated from it.

Income Gini vs. wealth Gini. Income is a flow earned over a period. Wealth is the stock of assets minus debts accumulated over time. Wealth is far more concentrated than income in most countries, so a nation can show moderate income inequality while a small group holds most of its net worth.

Market (pre-tax) Gini vs. disposable (post-tax) Gini. Market income is earnings and investment income before government intervention. Disposable income is what households have after taxes and cash transfers. Databases often label these separately (for example, “Gini_mkt” and “Gini_disp”), and the gap between them is a rough gauge of how much taxes and transfers redistribute. Consumption-based Gini figures, used for many developing countries, are usually lower than income-based ones and should not be compared directly.

Gini in other fields. In credit scoring and machine learning, a Gini coefficient measures how well a model ranks outcomes, and “Gini impurity” is a criterion used to split decision trees. These share the name and the inequality concept but do not measure income distribution.

FAQs

The Bottom Line

Anyone comparing countries, tracking inequality over time or judging the reach of taxes and transfers can use the Gini as a compact, comparable starting point. It still hides the shape of the distribution and the level of income, and it shifts with demographics and data quality. Compare figures only when they come from the same source and method, and pair them with measures such as top income shares or GDP per capita.