Headline growth figures for the United States are estimates that change as better data arrives, and the latest revision to second-quarter 2026 output was large. This article describes the United States only: how the Bureau of Economic Analysis (BEA) produces gross domestic product (GDP) figures, what changed in the latest release, and how to read the next one.
What changed in the second-quarter 2026 figures?
According to the BEA’s third estimate, released September 30, 2026, real GDP grew at a 2.2% annual rate in the second quarter of 2026 (April through June). The second estimate had put growth at 1.5%, so the figure was revised up by 0.7 percentage point (2.2 minus 1.5).
The BEA attributed the upward revision to higher investment, consumer spending and government spending. Investment changes covered both inventories and fixed investment, consumer spending was revised up for both services and goods, and federal defense drove the government spending change.
The same release included the BEA’s 2026 annual update, which covers the first quarter of 2021 through the first quarter of 2026 and keeps 2017 as the reference year. The first-quarter 2026 growth rate now stands at 2.5%, according to both the BEA and the Federal Reserve Bank of St. Louis’s FRED database.
What does GDP actually measure?
The BEA defines gross domestic product as the value of the final goods and services produced in the United States. “Final” matters: the flour a bakery buys is not counted separately from the bread it sells, because that would count the same output twice.
Two versions appear in the news:
- Nominal GDP (current-dollar GDP) values output at the market prices of the period being measured.
- Real GDP (chained-dollar GDP) removes the effect of inflation so that different periods can be compared. The growth rates quoted in headlines are almost always real.
Real GDP is a measure of production, not of how any individual household is doing. A household’s experience depends on its own income, debts and spending, which is why the link between output and living standards is looser than the headlines suggest. A related point is covered in the guide to how real wages compare with prices.
Why does the BEA publish three estimates for every quarter?
For each quarter, the BEA publishes three estimates. The advance estimate arrives about a month after the quarter ends and relies on the best information available at that time. The second and third estimates incorporate source data that were not available earlier, which the BEA says improves accuracy.
The trade-off is speed versus completeness. Waiting for every survey and tax record to arrive would mean reporting the economy’s performance many months late. Publishing early gives policymakers, businesses and markets a timely read, at the price of an estimate that is expected to move.
The second-quarter 2026 sequence shows how much it can move:
| Estimate | Real GDP growth, Q2 2026 (annual rate) |
|---|---|
| Second estimate | 1.5% |
| Third estimate (September 30, 2026) | 2.2% |
| Revision | +0.7 percentage point |
Revisions can go in either direction. A figure being revised is a normal feature of the process, not evidence of a mistake.
What does “annual rate” mean?
The BEA reports quarterly data at annual rates “for ease of comparison.” A quarter covers only three months, so the raw change over the quarter is much smaller than the number in the headline. The annual rate asks: if this quarter’s pace continued for four quarters in a row, how much would output grow over a full year?
A worked example with illustrative numbers: suppose output rises 0.5% in a quarter. Compounded over four quarters, 1.005 multiplied by itself four times is about 1.0202, so the annualized rate is roughly 2.0%. The quarterly change of 0.5% and the annualized 2.0% describe the same quarter; they only differ in the time frame used.
One consequence is that annualized figures exaggerate how much a single quarter’s swing might mean for a whole year. A strong or weak quarter is a pace, not a prediction.
How noisy is quarterly growth?
Looking at several quarters side by side shows why a single reading should not be over-interpreted. FRED, which republishes the BEA’s series, shows these real GDP growth rates at seasonally adjusted annual rates:
| Quarter | Real GDP growth |
|---|---|
| Q2 2025 | 4.0% |
| Q3 2025 | 3.9% |
| Q4 2025 | 0.2% |
| Q1 2026 | 2.5% |
| Q2 2026 | 2.2% |
The fourth quarter of 2025 stands out at 0.2%, between two much faster readings. Several quarters taken together usually say more about the underlying trend than any single release does. Because these figures are as of the BEA’s September 30, 2026 update, they may be revised again in future annual updates.
Does one measure of output tell the whole story?
The BEA also publishes gross domestic income (GDI), which measures the same economy from the income side: wages, profits and other earnings. In theory, output and income should match; in practice they are built from different source data, so they differ. According to the BEA, real GDI rose 2.6% in the second quarter, revised up 0.4 percentage point.
Another measure is real final sales to private domestic purchasers, which strips out volatile items such as inventories, government spending and trade to focus on private demand. The BEA reported that it rose 4.6% in the second quarter, up 0.4 percentage point from the prior estimate.
The BEA also reported the PCE price index, its inflation gauge based on consumer spending, rising 5.0% at an annual rate in the second quarter, revised down 0.3 point, with the measure excluding food and energy at 3.3%, also revised down 0.3 point. These are annualized quarterly rates, which are not the same as the year-over-year inflation figures that often appear in headlines. For how inflation indexes are built and why different measures diverge, see what inflation actually measures.
Using several indicators together is a more reliable habit than relying on one. When output, income and private demand tell a similar story, confidence in the direction of the economy is higher; when they diverge, caution is warranted.
Why do GDP numbers matter beyond the headline?
GDP releases feed into the assessments of many institutions, but the link to any particular decision is indirect. Central banks, including the Federal Reserve, weigh growth alongside inflation and the labor market when setting policy; the reasoning is laid out in why interest rates matter more than most people think. A revision that raises past growth does not by itself change what the Federal Reserve will do, since policymakers look at a broad range of data.
Markets also react to the gap between a release and expectations rather than to the level of growth itself, a theme explored in why markets can rise even when the economy looks weak. A strong GDP number can be met with a shrug if it was widely expected.
For households, the more practical reading is cautious. A stronger growth figure does not mean paychecks rise by the same amount, and a weaker one does not mean jobs vanish. GDP is best read as background context for the economy, not a guide to personal finances.
How to read the next release on October 29
The BEA has scheduled the advance estimate of third-quarter 2026 GDP for October 29, 2026, at 8:30 a.m. Eastern time. Because it is an advance estimate, it will be based on the least complete data of the three estimates. A few habits help when reading it:
- Check which estimate it is. Advance, second and third estimates carry different levels of information.
- Note the annual rate. A quarterly pace is not a full-year forecast.
- Compare against revised prior quarters. The 2026 annual update changed data back to 2021, so older figures quoted in earlier articles may not match the current series.
- Look at the components. Consumer spending, investment, government spending and trade each tell a different part of the story, and a headline can hide offsetting moves.
- Wait for revisions before drawing firm conclusions. The second-quarter experience, where a 1.5% reading became 2.2%, is a reminder that early numbers are provisional.
Reading growth figures with the right amount of caution
The second-quarter revision from 1.5% to 2.2% shows why GDP is best treated as a living estimate. The BEA’s three-stage process trades some precision for timeliness, annualized rates amplify the quarter-to-quarter swings, and no single measure captures the whole economy. Readers who follow output, income and private demand together, and who treat an advance estimate as the first draft it is, will be better placed to understand what the headlines are saying and what they are not.
This article is general information, not personalized financial advice. Rules, data and rates change, and figures may be revised.
Sources: U.S. Bureau of Economic Analysis; Federal Reserve Bank of St. Louis
