Gini Coefficient: The Number That Measures Inequality

Sweden and the United States are both rich, hardworking countries with big economies. But the money lands very differently. After taxes and government help, Sweden's Gini coefficient sits near 0.28, while the United States runs around 0.39. You will not find a single dollar amount in those two numbers, yet they tell you something real: income in Sweden is spread far more evenly than it is in America, and the gap has held for decades. That is the whole job of the Gini coefficient. It takes the messy incomes of millions of households and boils them down to one number you can track over time and line up against other countries.

What the Gini coefficient actually is

The Gini coefficient is a single number that measures how evenly, or how unevenly, income is shared across a group of people. It is named after Corrado Gini, the Italian statistician who introduced it in 1912, and it has been the go-to inequality measure ever since. The number always lands between 0 and 1:

• 0 means perfect equality. Everyone earns exactly the same. Nobody has a dollar more or less than anyone else.
• 1 means perfect inequality. One person earns all the income and everyone else earns nothing. Neither extreme actually happens. Every real country sits somewhere in the middle, usually between about 0.25 and 0.65. You will also see the same idea written as the Gini index, which is just the coefficient multiplied by 100 (so 0.39 becomes 39). The World Bank reports it that way. It is the same measure, only a bigger number.

Where the number comes from: the Lorenz curve

To see how the Gini is built, you first need a simple graph called the Lorenz curve. Here is how it works. Line everyone up from poorest to richest, then plot two things against each other: the share of people counting up from the bottom, and the share of total income those people hold. If income were perfectly equal, the poorest 20 percent of people would earn 20 percent of the income, the poorest half would earn half, and so on. That traces a straight diagonal called the line of perfect equality. Real life never looks like that. Because people at the bottom earn less than their proportional share, the real Lorenz curve sags below the diagonal, bowing down toward the bottom corner. The more unequal a country is, the deeper that sag. Now picture two spaces on the graph. Area A is the gap between the straight equality line and the sagging Lorenz curve. Area B is everything underneath the Lorenz curve. The Gini coefficient is simply: Gini = Area A ÷ (Area A + Area B) In plain words, it is the size of the inequality gap measured against the whole space below the equality line. Since that whole triangle always works out to 0.5, the math also shortcuts to Gini = 2 × Area A. A bigger sag means a bigger Area A, which means a higher Gini. The U.S. Census Bureau builds its official index the same way, off the difference between the real income curve and a perfectly equal one.

Work one out yourself

The best way to stop treating the Gini like magic is to compute one by hand. Say we have a small economy split into five equal-sized groups (these are called quintiles), each holding this slice of total income, listed from the poorest fifth to the richest:

That running-total column is your Lorenz curve. Now measure Area B under it by slicing the space into five strips and treating each as a trapezoid, whose area is its width times the average of its two heights. Working in decimals, with each strip 0.2 wide:

• Strip 1: 0.2 × (0 + 0.05) / 2 = 0.005
• Strip 2: 0.2 × (0.05 + 0.15) / 2 = 0.020
• Strip 3: 0.2 × (0.15 + 0.30) / 2 = 0.045
• Strip 4: 0.2 × (0.30 + 0.55) / 2 = 0.085
• Strip 5: 0.2 × (0.55 + 1.00) / 2 = 0.155

Add them up and Area B comes to about 0.31. The full triangle under the equality line is 0.5, so Area A is 0.5 minus 0.31, which is 0.19. That gives: Gini = 0.19 ÷ 0.5 = 0.38 A Gini of 0.38 is moderate inequality, right about where the United States sits. Not bad for five numbers and some grade-school geometry. The lumping does cost you something: stuffing the whole top fifth into one block hides how concentrated income is inside that group, so a real calculation with finer data would push the number a little higher.

What the numbers look like in the real world

Once you can read the scale, the country comparisons get interesting. These are rough Gini figures for income after taxes and government benefits, which is the fairest way to compare countries (more on why in a second):

Before tax or after tax? It changes everything

Here is the detail that trips most people up, and the one your professor will test. A Gini coefficient means almost nothing until you know which income it is measuring. Market income is what people earn before the government touches anything: wages, business profits, investment returns. Disposable income is what is left after taxes come out and benefits are added back in, things like Social Security, unemployment checks, and food assistance. Those two produce very different Ginis for the same country in the same year. In the United States the market-income Gini is high and has climbed for decades, while the Congressional Budget Office shows that taxes and transfers pull the after-government number down by a real amount. The Census Bureau's Gini for household money income runs near 0.48 in recent years, while the after-tax, after-transfer figure lands closer to 0.39. Same country, same year, different number, because they are counting different things. So whenever you see a Gini quoted with no label, treat it as half a fact.

Income is not the only thing the Gini measures

Most Gini numbers you read are about income, but the exact same math works on wealth, meaning everything a household owns minus what it owes. And wealth is far more lopsided than income. While the U.S. income Gini sits somewhere between 0.39 and 0.49 depending on which measure you use, the wealth Gini runs above 0.8, because a huge share of families hold almost no savings while a small group holds most of the assets. When you hear that inequality is worse than the headline suggests, this gap between income and wealth is usually the reason.

What the single number quietly hides

I think the Gini is genuinely useful, but it earns its keep only if you respect what it leaves out. Squeezing a whole country into one number throws away information, and three blind spots matter most. First, two very different countries can post the same Gini. A place where the middle class is hollowed out and a place where the very poorest are destitute can land on the same number through completely different Lorenz curves. The Gini tells you how much inequality exists, not where it sits. To find that, you go back to the group shares or look at the top 1 percent directly. Second, the Gini is only as honest as the data behind it. A Gini built on Census money income leaves out things that would push it up, like capital gains, and things that would push it down, like health coverage and other non-cash help. Change what counts as income and you change the number. Third, it is a photo, not a film. A single year's Gini cannot tell you whether the same families are stuck at the bottom every year or whether people move up and down over time. A country with high yearly inequality but lots of movement is a very different place than one where the bottom never changes, yet both can show the exact same Gini.

So how should you use it?

Used well, the Gini is a great first question. In one number it tells you whether a country shares its income more like Denmark, more like the United States, or more like South Africa, and which direction it is heading over time. That is real information, and it is why governments, the World Bank, and the OECD all lean on it. Used badly, it becomes a final verdict that hides everything that actually matters: who is poor, why they are poor, and whether they can climb out. So treat the Gini as the headline, not the whole story. Read the number first, then ask the next question. And if someone hands you a single Gini and acts like the argument is settled, ask them which income it measures, and what it left out.