Gini Coefficient¶

Collapse a whole distribution of a resource into one number on a 0-to-1 scale by measuring how far its Lorenz curve bows away from perfect equality.

Version: v2 · 2026-09-19 · History
Domain-specific #: 450
Domain group: Social Sciences
Origin domain: Economics & Finance
Subdomain: Inequality and Distributional Measurement → Economics & Finance

Core Idea¶

The Gini coefficient (Corrado Gini, 1912) is a single scalar summary of how unequally a resource is spread across units. Geometrically it is twice the area between the Lorenz curve — cumulative population share, poorest-first, against cumulative resource share — and the 45-degree line of equality, from 0 to 1. It is a measurement instrument chosen for desirable axioms, but does not uniquely characterise a distribution: different shapes can share one Gini.

How would you explain it like I'm…

The Fair-Share Score

Imagine sharing a bag of candy with your friends. If everyone gets the same amount, the score is zero. If one kid gets all the candy, the score is one. The Gini Coefficient is that score for how unevenly something is shared.

Zero-to-One Unfairness Number

The Gini Coefficient is one number from 0 to 1 that shows how unevenly something, like money, is shared among people. Zero means everyone has exactly the same, and 1 means one person has everything. You can find it by drawing a curve that lines people up from poorest to richest and shows how much of the total each group holds, then measuring how far that curve sags from a perfectly fair straight line. Scientists also use it for other things, like how unevenly citations or animal numbers are spread out. But two very different situations can have the same Gini number, so it doesn't tell the whole story.

Lorenz-Area Inequality Index

The Gini Coefficient measures inequality in how a non-negative resource, like income, is spread among people or units. Line everyone up from poorest to richest and plot the Lorenz curve: for each share of the population, the share of the total they hold. Perfect equality is the 45-degree line, where the bottom 30% holds 30%. The Gini is twice the area between that line and the Lorenz curve, ranging from 0 (perfect equality) to 1 (one unit holds everything). It doesn't change if you double everyone's income or copy the whole population, and moving some resource from a richer unit to a poorer one always lowers it. Its weakness is that differently shaped distributions can share a Gini, so analysts pair it with other measures like top-income shares.

The Gini Coefficient (Corrado Gini, 1912) summarizes inequality of a one-dimensional non-negative distribution in one number. Geometrically it is twice the area between the 45-degree line of perfect equality and the Lorenz curve, which plots cumulative population share (poorest first) against cumulative resource share. Algebraically it equals the mean absolute difference over all pairs divided by twice the mean, i.e., half the relative mean absolute difference, showing it is pairwise dispersion scaled by the average. It ranges from 0 (perfect equality) to 1 (one unit holds everything). It is scale-invariant, population-replication invariant, and satisfies the Pigou–Dalton transfer principle: a rich-to-poor transfer strictly lowers it. It does not uniquely characterize a distribution—a long upper tail and mass bunched near the bottom can yield the same value—so it is paired with decile shares, Theil indices, and top-share measures. Beyond income and wealth, the same construction measures concentration in species abundance, firm sizes, citation counts, and health outcomes.

Scope of Application¶

Being a functional, not a mechanism, the Gini applies literally wherever its precondition holds: a one-dimensional non-negative distribution across orderable units.

Income and wealth inequality — the canonical use, read alongside decile and top-1% shares.
Land and resource allocation — its 1912 land-holding origin; now also water, quota, emissions allowances.
Industrial organisation — how production or market share concentrates across firms.
Health and human development — the "health Gini" on mortality or attainment across regions.
Ecology and bibliometrics — concentration of species abundance, or of citation counts.
Network science and fairness — the Gini of a degree distribution or allocation across protected groups.

Clarity¶

The coefficient turns inequality from a vague rhetorical referent into a comparable quantity: a full distributional shape collapses to one number on a fixed [0, 1] scale, so a hundred countries become a scatter and a trend a single line. Because the Gini does not uniquely characterise a distribution, a change in it is not self-interpreting — which is what tells a practitioner when to reach for decile shares rather than over-read one scalar.

Manages Complexity¶

The raw object is a full distribution — every household's holding, incomparable across populations of different size and currency. The Gini collapses it to one engineered scalar: scale-invariance strips out currency, population-invariance strips out sample size, the transfer principle fixes its direction. The analyst tracks a number where they would track a distribution, with the many-to-one slack itself signalling when a top-share decomposition is needed.

Abstract Reasoning¶

The instrument licenses a comparative move that ranks distributions across populations and time, warranted by its built-in invariances. It supports a diagnostic move that treats a bare Gini shift as not-yet-interpreted and reads it against top shares to localise the change, and a boundary-drawing move that situates the Gini among rival summaries and picks the one whose sensitivity matches the question.

Knowledge Transfer¶

Because the Gini is a measurement instrument, not a mechanism, nothing is lost at a boundary: the construct itself travels — as literal application, not analogy — wherever its precondition holds, so the same Lorenz-area scalar and reading discipline apply unchanged from income to species abundance to degree distributions. The insight that one-dimensional concentration admits scalar summarisation sits one level up at the parent prime Measurement. The caution: it is a Measure, not inequality itself, and insensitive to the tail, so it travels alongside decile and top-share measures.

Relationships to Other Abstractions¶

Current abstraction Gini Coefficient Domain-specific

Parents (3) — outgoing typed relations

Gini Coefficient presupposes Lorenz Curve Domain-specific

Gini presupposes the Lorenz concentration object because its defining scalar is twice the area between that curve and the equality diagonal.
Gini Coefficient is a decomposition of Aggregation Prime

Gini strips to a deliberate many-to-one collapse of a complete distribution into one comparable scalar at the cost of shape information.
Gini Coefficient is a decomposition of Measurement Prime

Removing Lorenz and inequality terminology leaves a procedure mapping a distributional concentration attribute onto a calibrated zero-to-one scale.

Neighborhood in Abstraction Space¶

Gini Coefficient sits in a sparse region of the domain-specific corpus (71^st percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2769 abstractions)

Nearest neighbors

Theil Index — 0.87
Lorenz Curve — 0.86
Species–Area Relationship — 0.83
Benford's Law — 0.82
Foster–Greer–Thorbecke Poverty Measures — 0.82

Computed from structural-signature embeddings · 2026-10-25