Income inequality is the uneven distribution of income among individuals or households within a population. Income includes earnings from employment, self-employment, investments, and transfers received over a specified period. The concept concerns how resources are distributed, rather than simply how much income an economy produces. It is distinct from wealth inequality, which concerns accumulated assets and liabilities, and from poverty, which concerns resources falling below a specified threshold. (oecd.org)
Income concepts and units of comparison
The measured extent of inequality depends on the definition of income. Market income generally consists of earnings and investment income before government taxes and transfers. Disposable income includes public cash transfers and subtracts income taxes and household social security contributions. Comparing these distributions helps describe the redistributive effect of public policy, although it does not capture every effect of government activity. (oecd.org)
Household income also requires adjustment for household size. Two households receiving identical amounts may have different living standards if one supports more people. An equivalence scale accounts for differences in needs and the economies of sharing housing and other resources. The OECD commonly divides household income by the square root of household size. Under this convention, a four-person household needs twice the income of a one-person household to have the same equivalised income. (oecd.org)
Annual income inequality differs from inequality over a lifetime. Temporary unemployment, study, retirement, or fluctuations in business earnings can affect a single year's income. Inequality of opportunity is another distinct concept: it concerns the influence of circumstances outside individual control, such as family background, on economic outcomes. Neither lifetime inequality nor unequal opportunity can be fully described by a snapshot of annual income. (imf.org)
Measurement
The Lorenz curve ranks people from lowest to highest income and plots their cumulative population share against their cumulative income share. Perfect equality corresponds to the diagonal: the poorest 20 percent, for example, would receive 20 percent of income. The Gini coefficient expresses the area between this diagonal and the Lorenz curve relative to its maximum possible value. It is commonly reported on a scale from zero to one, or as an index from zero to 100. (databank.worldbank.org)
Other measures examine particular parts of the distribution. Top-income shares report the proportion received by groups such as the richest tenth. The S90/S10 ratio compares the average incomes of the richest and poorest tenths. These measures complement the Gini coefficient by making specific distributional gaps more visible. Different income distributions can nevertheless have the same Gini, so a single index does not uniquely describe a society's income structure. (oecd.org)
Most conventional inequality measures describe relative differences. Consequently, inequality can increase while absolute poverty declines, if incomes rise throughout the distribution but grow faster at the top. Conversely, a comparatively equal distribution does not establish that average incomes are high. Poverty indicators and measures of average income therefore answer questions that inequality indices alone cannot resolve. (datatopics.worldbank.org)
Economic mechanisms
Income differences arise through interacting economic and institutional processes. In the labor market, education, experience, occupational opportunities, and employment conditions influence earnings. Technological change can increase demand for particular skills while replacing or reorganizing other tasks. Automation may therefore raise some workers' earnings while reducing demand for others. Its distributional effects depend partly on access to training and the supply of relevant skills. (elibrary.imf.org)
Globalization and international trade also change demand across industries and occupations. Their effects differ across countries and groups rather than following a universal pattern. Labor institutions, protections, and access to financial services help explain why similar international pressures can produce different national outcomes. A 2015 IMF staff study identified technological skill premiums, globalization, weakening labor protection, and limited financial inclusion as relevant drivers, with their importance varying by country. (imf.org)
Income from assets introduces another mechanism. Savings, property, and financial investments can generate interest, rent, and dividends. Unequal asset ownership can therefore contribute to unequal income, while income differences affect households' capacity to accumulate assets. Income and wealth remain separate distributions: a household may own valuable property yet receive little current income. (oecd.org)
Economic consequences and public policy
Research examines relationships between inequality, economic growth, and investment in human capital. Limited access to education and finance can restrict opportunities for lower-income households. However, observed correlations between inequality and growth do not by themselves establish causation. Findings depend on the mechanisms studied, country conditions, and the period examined; inequality is not a single, uniform economic intervention. (elibrary.imf.org)
Fiscal policy changes disposable-income distributions through taxes and transfers. Progressive taxation and cash benefits can reduce measured inequality, while public education and other services affect access to resources beyond cash income. Their effects depend on eligibility, financing, implementation, and which households actually receive benefits. Income redistribution can also affect incentives to work, save, and invest, making policy design relevant to both distributional outcomes and economic activity. (imf.org)
Data and comparability
Inequality estimates commonly use nationally representative household surveys. Comparisons require attention to whether datasets measure income or consumption, use pre-tax or disposable income, and adjust for household composition in the same way. Consumption is generally distributed less unequally than income, so these measures should not be treated as interchangeable. Survey methods and income definitions can also change between years. (databank.worldbank.org)
Reported statistics must therefore be read alongside their reference year, population coverage, and methodological documentation. Coverage of households at the top of the distribution presents particular difficulties. Cross-country tables may combine observations collected in different years, and national Gini coefficients cannot simply be averaged to obtain a global Gini: the combined distribution must also incorporate income differences between countries. (oecd.org)