What Is NFP (Non-Farm Payrolls)?

What is NFP? Non-Farm Payrolls is the monthly change in the number of paid jobs in the United States outside farming. If the US added 150,000 jobs last month, that 150,000 is the number everyone quotes.

The Bureau of Labor Statistics (BLS) publishes it in a report called the Employment Situation. Usually that is the first Friday of the month at 8:30 a.m. New York time, which is 13:30 in London. For a few weeks in March and around the start of November the clocks in the US and Europe change on different dates, and it lands an hour earlier for European traders. Holidays move the date too, so check the BLS release schedule or When is the next NFP? rather than counting Fridays.

Where the number comes from

The BLS asks about 119,000 businesses and government agencies, covering some 622,000 worksites, how many people they had on the payroll (BLS technical note). That survey is called Current Employment Statistics.

“Non-farm” is a little misleading, because more than farms are left out. The count skips farm workers, but also the self-employed, private household staff, unpaid family workers and the armed forces (BLS CES FAQ). Civilian government jobs are in it. So when the government hires a lot of people in one month, the headline jumps without the private economy doing anything.

The same report holds three numbers

Newcomers watch only the headline. The report gives you at least three numbers, and the market reads them together:

• Payrolls: the headline change in jobs.
• Average hourly earnings, from the same survey. The Fed watches pay closely, because pay that rises fast tends to push prices up.
• The unemployment rate, from a separate survey of about 60,000 households. Its releases are on US Unemployment Rate.

Because the two surveys ask different people different questions, they can disagree. Strong payrolls next to a rising unemployment rate is not a typo, and it is not rare either.

Across the reports for 2015 to 2025 as first published, leaving out 2020, every January report (when the household survey moves to new population estimates) and November 2025 (no October household survey to compare with, after the shutdown), payrolls rose while the unemployment rate went up in 30 of 108 reports. In 27, the two surveys’ counts of people in work went in opposite directions, and in all but one of those it was payrolls up, household survey down (BLS releases, kept by ALFRED).

The first number is an estimate

The BLS revises each month’s figure in the next two reports, as late survey answers come in. So every NFP report also changes the two months before it.

Those revisions are not small. Comparing each month’s first estimate with its third, from 2015 to 2025 and leaving out 2020 and October 2025 (first published late, after the shutdown), the median month moved by 34,000 jobs. It moved by more than 50,000 in 39 of the 119 months and by more than 100,000 in 14. In four of them the third estimate had the opposite sign to the first: September 2017 went from jobs lost to jobs gained, and June, August and December 2025 the other way (BLS estimates as published, kept by ALFRED).

Say the headline comes in 40,000 above the forecast, but the two earlier months are revised down by 60,000 together. On balance the labour market is weaker than you thought this morning, whatever the headline says.

Once a year the whole series is checked against unemployment insurance records, which cover nearly every employer. That correction can be large. In September 2025 the BLS said its March 2025 level was probably about 911,000 jobs too high (BLS preliminary benchmark).

Why a jobs report moves currencies

Congress gave the Federal Reserve two goals, maximum employment and stable prices (Federal Reserve). The jobs report speaks to the first directly and, through wages, to the second. So it shifts what traders expect the Fed to do with interest rates, and the dollar follows those expectations within seconds.

• Payrolls above the forecast make higher US rates, or fewer cuts, more likely. That tends to support the dollar.
• Payrolls well below the forecast point to a slowing economy and earlier or deeper rate cuts. That tends to weigh on the dollar.

What moves the market is the gap between the result and the forecast, not the number itself. 200,000 jobs is good news when 150,000 was expected and bad news when 275,000 was. The forecast and every recent result against it are on US Non-Farm Payrolls (NFP).

The first seconds trade on the headline, and wages and revisions can turn the price around within minutes. Reading the release in that order, with the slippage and the scenarios, is in NFP Trading Guide: How Non-Farm Payrolls Move the Dollar.