Reading the latest movement
In September 2026, the monthly change in nonfarm payroll employment was 29 thousand jobs, down 104 from August 2026 and down 47 from September 2025. Across 956 readings since February 1947, the average is 121 thousand jobs, and the latest reading is higher than 25% of them. The average since January 2017 is 117 thousand jobs.
A difference between periods describes economic change in this edition of the data; a later revision of either observation is a separate event. Monthly changes are noisy; the range of recent months is usually more informative than a single reading.
What the nonfarm payroll employment tells you
The number of jobs on U.S. business and government payrolls, excluding farm work, usually reported as the monthly change in jobs.
The monthly change in nonfarm payrolls is the most market-moving U.S. labor statistic. Bond and stock traders react within seconds of the release because job growth informs expectations for consumer income, inflation and Federal Reserve policy. The National Bureau of Economic Research lists payroll employment among the key indicators it considers when dating recession peaks and troughs. Its large sample and industry detail make it the standard measure of how many jobs the economy is creating.
Businesses use the industry breakdown, which covers sectors from construction to health care to government, to track demand in their own markets and to benchmark hiring plans. State and local economic-development teams compare national growth with state payroll data from the same program. Journalists and campaigns cite the monthly figure heavily, which makes it important to know how large the sampling error is and how much early estimates can be revised.
How to read it
- Focus on the monthly change, and smooth it with a three-month average. BLS puts the 90 percent confidence interval for a single month's change at about plus or minus 136,000 jobs.
- Always read the revisions to the previous two months, which are reported in the same release and can shift the trend as much as the new month.
- Compare job growth with the pace needed to keep up with labor-force growth; that breakeven depends on population and immigration trends and shifts over time.
- Check private versus government jobs and the industry mix. Gains concentrated in one or two sectors are a narrower signal than broad-based growth.
- Strikes, hurricanes and severe weather can depress one month's count when workers are not paid during the reference pay period, typically followed by a rebound.
What moves it
Payrolls grow when demand for goods and services rises enough that employers add staff, and shrink when demand falls, financing costs rise or a shock forces closures. Interest-rate-sensitive sectors such as construction, manufacturing and finance tend to turn first, while health care and government often grow steadily through cycles. Labor supply also matters: population growth, immigration and participation set how many jobs can be filled. Business formation and closures, which the survey estimates through a birth-death model, become an important source of error near turning points.
Records and long-run averages
Computed from every published monthly reading of the monthly change (thousands of jobs) in this edition, February 1947 to September 2026. Records reflect the acquired history only; earlier data published elsewhere is not included.
| Measure | Value | When / basis |
|---|---|---|
| Latest | 29 | September 2026 |
| Record high | 4,631 | June 2020 |
| Record low | -20,469 | April 2020 |
| Long-run average | 121 | Median 156 · February 1947 onward |
| Average since 2017 | 117 | Last ten calendar years |
| Latest percentile | 25th | Share of readings below the latest |
Year-by-year summary
Calendar-year average, range and final reading of the monthly change (thousands of jobs). The total adds the monthly changes, which equals the December-to-December change in the level. A year marked partial includes only the months published so far.
| Year | Average | Total | Low | High | Last reading |
|---|---|---|---|---|---|
| 2026 partial | 68 | 612 | -156 | 214 | 29 September 2026 |
| 2025 | 10 | 116 | -140 | 108 | -17 December 2025 |
| 2024 | 122 | 1,459 | 9 | 237 | 237 December 2024 |
| 2023 | 210 | 2,515 | 68 | 434 | 154 December 2023 |
| 2022 | 377 | 4,526 | 100 | 819 | 100 December 2022 |
| 2021 | 606 | 7,268 | 315 | 942 | 586 December 2021 |
| 2020 | -771 | -9,246 | -20,469 | 4,631 | -185 December 2020 |
| 2019 | 165 | 1,985 | 7 | 296 | 127 December 2019 |
| 2018 | 191 | 2,286 | 60 | 394 | 192 December 2018 |
| 2017 | 176 | 2,115 | 89 | 223 | 151 December 2017 |
| 2016 | 194 | 2,331 | 46 | 373 | 225 December 2016 |
| 2015 | 226 | 2,713 | 96 | 342 | 269 December 2015 |
Readings around recent recessions
Values of the monthly change at each NBER business-cycle peak and trough inside this history, with the range from the peak through twelve months after the trough. Shaded bands on the chart mark the same NBER recession dates.
| Recession (peak to trough) | At peak | At trough | Low in window | High in window |
|---|---|---|---|---|
| February 2020 to April 2020 | 262 | -20,469 | -20,469 April 2020 | 4,631 June 2020 |
| December 2007 to June 2009 | 105 | -458 | -824 March 2009 | 522 May 2010 |
| March 2001 to November 2001 | -37 | -308 | -311 October 2001 | 115 October 2002 |
| July 1990 to March 1991 | -35 | -167 | -323 February 1991 | 77 June 1991 |
Recession dates: NBER Business Cycle Dating Committee. A reading at a peak or trough month is not a forecast signal on its own.
Put the latest level in context
These comparisons use published levels in this snapshot, with matching reference periods. Rate differences use percentage points; growth of a rate is not a percentage-point difference. The chart above may use a different transformation.
| Earlier period | Earlier level | Latest level | Absolute difference | Relative change |
|---|---|---|---|---|
| September 2025 | 158,548 | 159,044 | 496 | 0.31% |
| September 2021 | 147,771 | 159,044 | 11,273 | 7.63% |
| September 2016 | 144,967 | 159,044 | 14,077 | 9.71% |
How have prices, jobs and output changed since 2019? · Research questions and worked comparisons.
Definition and limitations
Counts jobs, not unique people. A person with two payroll jobs can be counted twice; the self-employed generally are excluded. Monthly estimates and benchmarks can revise earlier levels.
Percentage growth is calculated from matching series levels. Both endpoints retain the same adjustment basis. Growth computed from rounded source inputs can differ from the agency’s published headline rounding.
Which measure fits your question? Compare definitions and complementary measures before drawing a conclusion.
Release schedule and revisions
BLS publishes payrolls in the monthly Employment Situation report, usually on the first Friday of the month at 8:30 a.m. ET, from the Current Employment Statistics survey of roughly 120,000 businesses and government agencies. Each month's estimate is revised in the next two releases as more reports arrive. Once a year, levels are benchmarked to near-complete counts from unemployment insurance tax records, with the final benchmark published in the February release.
Next scheduled release on this site's verified calendar: Employment Situation, November 6, 2026.
Read it alongside
- Unemployment rate: The household-survey unemployment rate offers an independent check on what payroll growth implies for slack.
- Average hourly earnings: Earnings from the same survey show whether job growth comes with rising wage pressure.
- Initial unemployment insurance claims: Weekly claims can flag layoffs before they appear in the monthly payroll count.
Frequently asked questions
What is the annual payroll benchmark revision?
Each year BLS realigns the survey-based payroll level with counts from the Quarterly Census of Employment and Wages, which draws on unemployment insurance tax filings covering nearly all employers. BLS publishes a preliminary estimate of the benchmark adjustment in late summer, and the final revision is incorporated with the January data released in February. The revision can change the level of jobs for the prior year or more.
Why does the payroll survey not count the self-employed?
The survey collects data from employers about the workers on their payrolls. Self-employed people, unpaid family workers, farm workers and private household workers are outside its scope. They are captured by the household survey, which is one reason the two surveys report different employment totals.
Does a gain of 200,000 jobs mean 200,000 more people are working?
Not necessarily. Payrolls count jobs, not people, so someone who takes a second job adds to the count without any change in the number of employed people. Conversely, a person moving from a payroll job to self-employment reduces payrolls but not household employment. For counts of people, use the household-survey measures.
What is the birth-death model?
Newly opened businesses cannot be sampled until they appear in administrative records, and businesses that close often simply stop responding. The birth-death model estimates the net employment effect of these openings and closings each month. It works well in steady conditions but can miss turning points, which is one reason benchmark revisions tend to be larger around recessions.
Transformations
The published level keeps the source unit. Year-over-year change compares with the same period a year earlier (52 weeks for weekly data); previous-period change compares with the preceding observation, in percentage points when the series is already a rate. No frequency conversion, interpolation or hidden rescaling is applied. Formulas and methods.
Source and methodology
Producer documentation · Acquisition endpoint / file · Producer release information
Revisions and archived snapshots
Earlier site acquisitions are archived. Matching unchanged series identities can be compared between versions; this is not a complete agency vintage history. Previous-period readings above come from the same acquired snapshot. The revision comparison separates changes to matching historical observations from new coverage and corrected series identities. Earlier agency vintages remain outside this archive. Read the revision policy and download this frozen snapshot.