Income Inequality — Explained
Detailed Explanation
Define what is being distributed
Income inequality concerns differences in income across people or households over a specified period. It is not the same as inequality in consumption or wealth. Income is a flow; consumption measures resources used for current living; wealth is a stock of assets less liabilities at a point in time. A household can borrow or use savings to maintain consumption when income falls, so the distributions need not move together.
Before comparing two figures, identify the variable, the population unit, the period and the method. Per-person income differs from total household income when household sizes vary. Income before taxes and transfers differs from disposable income after them. A comparison that ignores these distinctions may make two valid estimates appear contradictory, or create an apparent trend from a change in measurement rather than a change in people's circumstances.
An average does not describe the distribution
Consider four hypothetical annual incomes of 10, 20, 30 and 40 units. Their total is 100, their mean is 25 and their median is also 25. Now raise the highest income to 140 while leaving the others unchanged. The mean becomes 50 because the total is 200, but the median remains 25. Average income has doubled without any increase for three of the four people.
The example does not show that averages are useless. It shows what they measure and why distributional information is needed alongside them. A median describes the middle of an ordered distribution, while a mean is influenced by the total. Neither alone reveals the entire pattern. Top shares, percentile ratios and measures covering the whole distribution answer further questions.
The Lorenz curve and Gini index
For a standard non-negative income distribution, a Lorenz curve relates cumulative population shares, ordered from lower to higher income, to cumulative income shares. The equal-distribution line provides a benchmark. The Gini summarises the area between that line and the Lorenz curve relative to the relevant total area. It is commonly expressed as a coefficient on a zero-to-one scale or an index on a zero-to-100 scale. Thus 0.30 and 30 can represent the same value on different scales.
The World Bank's Gini metadata explains the measure and the need for care in comparisons. A Gini value is not the percentage of people who are poor and does not identify which group gained or lost. Different distributions can have the same Gini. A change in the index should therefore be interpreted alongside the underlying distribution and measurement basis.
Relative inequality differs from absolute gaps
Return to the incomes 10, 20, 30 and 40. If each doubles, their shares of total income remain unchanged, so the relative distribution and its standard Gini do not change. The absolute gap between the highest and lowest incomes rises from 30 to 60. A statement that relative inequality is unchanged can therefore coexist with a larger absolute gap.
Poverty is another distinct question. With an illustrative fixed poverty threshold of 15 units, the lowest-income person moves above it when incomes double. Poverty under that threshold falls while relative inequality remains unchanged. The threshold and example are deliberately hypothetical. The lesson is that poverty reduction, average growth and inequality are connected but are not interchangeable outcomes or statistics.
Read Indian estimates with their methods attached
The World Bank's October 2026 India Poverty and Equity Brief displays a consumption-based Gini index of 23.0 for 2023. Its methodological note refers to the household consumption surveys and cautions that measurement changes and incomplete capture of wealthier households' consumption affect interpretation. The number should not be relabelled as an income Gini or presented without its reference year.
The World Inequality Lab's 2024 study provides a different kind of estimate. It combines national accounts, tax information, surveys and rich lists to study income and wealth distributions. It estimates the top one per cent's pre-tax income share at 22.6 per cent in 2022–23. This is a top share from a constructed income series, not a consumption Gini. Comparing 22.6 directly with 23.0 as if they were the same indicator would mix both the variable and the statistic.
Causes and responses need specific mechanisms
Income differences can reflect unequal access to education, health, productive assets, reliable work and markets, as well as differences in returns and bargaining positions. These mechanisms should be examined with evidence rather than used as a universal explanation for every observed gap. A household's current income can also fluctuate, so a single observation need not describe its long-term position.
Policy analysis should identify the objective: raising low incomes, improving opportunities, reducing exposure to shocks or changing the distribution after taxes and transfers. Different measures address different mechanisms and have costs and implementation requirements.
Public services may improve living conditions even when a narrow cash-income statistic does not fully capture the benefit. For UPSC, use a worked distribution, distinguish the measures and attach methods to empirical claims.
A stronger answer explains what the evidence shows before proposing a response.