How Do We Measure the Economy?

Start Here: Does It Count?

Gross domestic product, or GDP, is the most watched number in economics. Before reading what it is, decide which of these would count in this year’s U.S. GDP:

Define It: What GDP Counts

1 Final goods, market value, inside our borders

GDP is the market value of all final goods and services produced within a country in a given period, usually a quarter or a year. Each part of that definition does work:

Another way to avoid double counting is to add up value added: what each business adds to the materials it buys. A steelmaker, a parts supplier, and an automaker each add their share. Their shares sum to the price of the finished car. BMW’s Spartanburg plant, the company’s largest in the world, assembled 412,799 vehicles in 2025 and exported about half of them. Those exports were worth about $9 billion, which made BMW the largest U.S. auto exporter by value. All of that work at the plant is part of U.S. GDP.

WATCH What Is Gross Domestic Product (GDP)? (Marginal Revolution University)

2 Spending equals income

Statisticians can measure GDP by adding up what is spent on new goods and services, or by adding up the incomes earned producing them. In principle the two give the same answer, because every dollar a buyer spends is a dollar of income to someone: workers, owners, lenders, or landlords. On the spending side, GDP has four parts:

3 Where the number came from

Measure It: Reading the Numbers

4 Real versus nominal

GDP can grow because the economy produces more or because prices rise. Nominal GDP, which the Bureau of Economic Analysis (BEA) calls current-dollar GDP, measures output at today’s prices. Real GDP takes out price changes, so it tracks how much is actually produced. In the third quarter of 2026, current-dollar GDP grew at an annual rate of 5.6 percent. Real GDP grew 3.4 percent. The gap between the two is rising prices.

WATCH Nominal vs. Real GDP (Marginal Revolution University)

5 Reading the headline

A headline such as “GDP grew 3.4 percent” needs three cautions:

• It’s an annual rate. The economy didn’t grow 3.4 percent in three months. That is the pace it would reach if the quarter’s growth continued for a full year.
• It’s an early estimate. BEA revises each quarter’s number as better data arrive. Second-quarter 2026 growth was first reported at 1.5 percent and later revised to 2.2 percent. The next estimate of the third quarter comes out November 25, 2026.
• Look inside it. BEA reported that third-quarter growth came from investment, consumer spending, and government spending. The rise in investment mainly reflected businesses adding to inventories and buying equipment. Goods made and stored this quarter count now. They still have to be sold.

6 Per person

Total GDP measures the size of an economy. To get closer to how well people live, divide by population. In 2025, U.S. GDP per person was about $90,000, according to the World Bank. GDP per person is an average. It doesn’t say how output or income is shared.

WATCH Real GDP Per Capita and the Standard of Living (Marginal Revolution University)

Comparing Countries

7 Adjusting for prices: purchasing power parity

Comparing countries means turning yuan, rupees, and pesos into one currency. The simplest way is to use market exchange rates. But a haircut, a meal, or an apartment costs far less in a poor country than in a rich one, and exchange rates don’t capture that. The World Bank explains that because prices tend to be higher in high-income countries, “GDP comparisons based on exchange rates—which do not account for price-level differences—overstate the economic size of high-income countries while understating the economic size of low-income countries.”

Purchasing power parity (PPP) fixes this by valuing each country’s output at what money actually buys there. Economists at the St. Louis Fed illustrated the idea with a hamburger. In July 2010, a Big Mac cost $3.73 in the United States and 32 pesos in Mexico. At a “burger exchange rate” of 8.57 pesos per dollar, the two would cost the same. The actual exchange rate was about 13 pesos per dollar, so a dollar went further in Mexico than the exchange rate suggested. PPP rates do the same thing for a broad basket of goods and services.

8 Adjusting for population

The second adjustment is the same one used within a country: divide by population. Here is how the two adjustments change the picture:

Which country has the biggest economy depends on the measure. At exchange rates, the United States is largest. At PPP, China’s total output is larger. Per person, the average American produces about three times as much as the average person in China at PPP, and more than six times as much at exchange rates. The U.S. numbers don’t change because PPP is measured in U.S. prices. The adjustment can also go the other way. In high-price Switzerland, output per person falls from $114,769 at exchange rates to $102,513 at PPP. In Mexico, it rises from $13,889 to $25,868.

WATCH GDP & PPP (Marginal Revolution University)

Know Its Limits

Every measure leaves things out. These economists showed what GDP misses, and in which direction.

9 Frédéric Bastiat: What is not seen

In 1850, the French economist Frédéric Bastiat imagined a boy who breaks a shop window. Onlookers console the owner: the glazier gets work, and money circulates. Bastiat agreed the glazier gains six francs. “That is what is seen.” What is not seen is that the owner “will not be able to spend them for another” thing, such as new shoes or a book. His conclusion: “Society loses the value of objects unnecessarily destroyed.” GDP sees the repair. It doesn’t subtract the broken window.

READ THE ORIGINAL What Is Seen and What Is Not Seen (Frédéric Bastiat, 1850)

10 Work at home

Cooking, cleaning, and caring for children at home are real production, but no one is paid, so GDP leaves them out. Economists at BEA estimated what this work would be worth at market wages. Counting it would raise GDP by 39 percent in 1965 and by 26 percent in 2010. The share fell mainly because more women took paid jobs. Some of the measured GDP growth in those years was work moving from the home into the market. Counting home production lowers nominal GDP growth from 1965 to 2010 from 6.9 percent a year to 6.7 percent.

11 William Nordhaus: The price of light

The Yale economist William Nordhaus traced the cost of light from fires and candles to electric bulbs. He found that ordinary price data miss how much better and cheaper light became. As a result, the growth in light since the start of the industrial age “has been underestimated by a factor of between nine hundred and sixteen hundred.” Light is one example. Whenever new or better products appear, from vaccines to smartphones, GDP is likely to understate how much better off people are.

12 Erik Brynjolfsson: Free goods

Much of what people use online has a price of zero, so it adds little to GDP. Erik Brynjolfsson and his coauthors asked thousands of people how much they would have to be paid to give up a service for a year. In 2017, the median answer was $17,530 for search engines, $8,414 for email, and $3,648 for digital maps. They concluded that “digital goods have created large gains in well-being that are missed by conventional measures of GDP and productivity.”

13 Charles Jones and Peter Klenow: Closer than you might think

The Stanford economists Charles Jones and Peter Klenow built a broader measure of welfare that also counts leisure, life expectancy, and inequality. Across countries, they found that “GDP per person is an informative indicator of welfare”: the two measures have a correlation of 0.95. But individual countries can differ a lot. France has 70 percent of U.S. GDP per person, but its welfare is about 97 percent of the U.S. level, because the French live longer, work fewer hours, and share consumption more evenly. Each of those adds more than 10 percentage points. Many poor countries do worse than their GDP suggests. Shorter lives put sub-Saharan Africa at about 1 percent of U.S. welfare, compared with 5 percent of U.S. income.

14 Stiglitz, Sen, and Fitoussi: A dashboard

In 2009, a commission led by the economists Joseph Stiglitz, Amartya Sen, and Jean-Paul Fitoussi reviewed what GDP misses. Its recommendations included “look at income and consumption rather than production,” “give more prominence to the distribution of income, consumption and wealth,” and “broaden income measures to non-market activities.” For judging whether today’s living standards can last, it called for “a well-identified dashboard of indicators” rather than a single number. Since then BEA has begun publishing measures that show how income is shared among households.

What Economics Adds

LISTEN How to Avoid Lying With Statistics (with Jeremy Weber) on EconTalk, hosted by Russ Roberts, with a transcript

Key Terms in Plain Words

Gross domestic product (GDP)

The market value of all final goods and services produced in a country in a period.

Final good

A good or service sold to its last user, not used to make something else.

Intermediate good

A good used up in making another good, such as steel in a car.

Value added

What a business adds to the materials it buys.

Nominal GDP

GDP at current prices. BEA calls it current-dollar GDP.

Real GDP

GDP with price changes taken out, so it tracks actual output.

Annual rate

The pace growth would reach if one quarter’s growth continued for a year.

Inventories

Goods produced but not yet sold. Adding to them counts as investment.

GDP per person

GDP divided by population. Also called GDP per capita.

Household production

Unpaid work at home, such as cooking and child care, that GDP leaves out.

Exchange rate

The price of one currency in terms of another.

Purchasing power parity (PPP)

A way to compare countries by valuing output at what money actually buys in each.