Wealth Distribution: How Inequality Emerges from Fair Rules

simulator beginner ~8 min
Loading simulation...
Gini ≈ 0.50 — exponential distribution from equal starts

After 1000 rounds of random fair exchanges, 200 agents who started with equal wealth develop a Gini coefficient around 0.50. The wealth distribution converges to an exponential (Boltzmann-Gibbs) distribution — the maximum entropy distribution for a conserved quantity.

Formula

Gini = (2 × Σᵢ i × wᵢ) / (N × Σᵢ wᵢ) − (N+1)/N
Boltzmann-Gibbs distribution: P(w) = (1/⟨w⟩) × e^(−w/⟨w⟩)

Inequality from Fairness

Imagine a room of 200 people, each starting with exactly $100. Every round, two random people are paired and one gives the other a fraction of their wealth — a perfectly fair coin flip determines who pays whom. After a thousand rounds, the initial equality has vanished: some people have $500 while others have $5. No one cheated. No one had an advantage. Inequality emerged purely from the mathematics of random exchange.

The Boltzmann-Gibbs Distribution

Statistical physicists recognized this pattern immediately — it is the same exponential distribution that describes molecular energies in a gas. When a conserved quantity (money or energy) is randomly exchanged between agents (people or molecules), the maximum entropy distribution is exponential: P(w) = (1/W) e^(-w/W), where W is average wealth. This is not a coincidence; it is a fundamental result of statistical mechanics applied to economics.

The Lorenz Curve and Gini Coefficient

The Lorenz curve plots cumulative wealth share against population share, sorted from poorest to richest. Perfect equality is a 45-degree line; real distributions bow below it. The Gini coefficient — twice the area between the Lorenz curve and the equality line — quantifies this bow. Watch the Lorenz curve evolve in the simulation as the initially flat wealth distribution becomes exponential. The Gini converges to approximately 0.5, matching many real-world economies.

Policy Implications

This model delivers a profound insight: some degree of inequality is a mathematical inevitability of exchange economies, not evidence of systemic unfairness. However, real economies feature mechanisms that amplify inequality beyond the random-exchange baseline — compound returns on capital (Piketty's r > g), inheritance, and network effects. Understanding the baseline helps distinguish inevitable statistical inequality from policy-addressable structural inequality. The simulation shows what 'natural' inequality looks like — deviations above it are the target for economic policy.

FAQ

Why does inequality emerge from equal starting conditions?

When agents randomly exchange fixed fractions of wealth, some agents win several exchanges in a row while others lose several in a row. The resulting distribution converges to an exponential (Boltzmann-Gibbs) distribution — the same distribution as molecular energy in a gas. This is a maximum entropy result: it is the most likely macrostate given the constraint of total wealth conservation.

What is the Gini coefficient?

The Gini coefficient measures inequality from 0 (perfect equality) to 1 (one person has everything). Geometrically, it is twice the area between the Lorenz curve and the line of perfect equality. Real-world values range from about 0.25 (Scandinavia) to 0.63 (South Africa). The simulation's random exchange model typically produces Gini values of 0.4–0.6.

Is this model realistic?

The model captures a fundamental truth: even completely fair, random exchanges produce inequality. Real economies add mechanisms that can amplify inequality (returns on capital, inheritance) or reduce it (progressive taxation, redistribution). The model shows that some inequality is a natural statistical consequence of exchange, not necessarily of exploitation.

What is the Pareto principle?

The Pareto principle (80/20 rule) states that roughly 80% of outcomes come from 20% of causes. In wealth distribution, it means about 80% of wealth is held by 20% of the population. Vilfredo Pareto discovered this pattern in 1896 studying Italian land ownership, and it appears across many economic and natural systems.

Sources

Embed

<iframe src="https://homo-deus.com/lab/sociology/wealth-distribution/embed" width="100%" height="400" frameborder="0"></iframe>
View source on GitHub