Skip to main content

What are non-farm payrolls and why do they matter?

Every first Friday of the month, the U.S. Bureau of Labor Statistics releases the non-farm payrolls figure — a single number that can swing stock markets, reshape expectations about interest rates, and set the tone for the next month of economic analysis. This number represents the total change in jobs in the U.S. economy outside of farming, government, and household employment. For investors, policymakers, and anyone trying to understand the economy's health, non-farm payrolls are the single most important economic indicator released each month.

Non-farm payrolls tell you whether businesses are hiring, firing, or staying put. A strong jobs report — say, 250,000 new jobs in a month — signals confidence and growth. A weak report, or worse, job losses, suggests the economy is weakening. The Federal Reserve watches this figure closely because employment is one of its dual mandates alongside price stability. If jobs are growing too fast, inflation might accelerate. If jobs are disappearing, recession may be starting. For this reason alone, non-farm payrolls move markets in ways few other statistics do.

Quick definition: Non-farm payrolls measure the monthly change in the number of paid employees on U.S. business payrolls, excluding farms, government workers, and private household workers. This data point signals labor market health and influences monetary policy decisions.

Key takeaways​

  • Non-farm payrolls are released on the first Friday of each month, covering the prior month's employment data.
  • The figure includes roughly 130 million workers across goods, services, and construction sectors.
  • Strong jobs reports can cool stock markets if they suggest the Fed will raise rates longer; weak reports can boost equities on expectations of rate cuts.
  • The headline number tells you total job creation; sub-components show which industries are hiring or shedding.
  • Revisions to prior months' data are common and often significant, reshaping economists' understanding of labor market strength.

How non-farm payrolls are measured​

The non-farm payrolls figure comes from the Bureau of Labor Statistics' monthly Current Employment Statistics (CES) survey, conducted in conjunction with state employment agencies. The BLS surveys roughly 130,000 businesses each month, asking how many people they have on their payrolls in the pay period containing the 12th day of the month.

The survey covers all private employment — manufacturing, retail, healthcare, finance, technology, hospitality, you name it — plus government jobs. It does exclude farm workers (hence "non-farm"), self-employed individuals, gig workers, and household employees like nannies or housekeepers. This means non-farm payrolls capture traditional, salaried or hourly employment, which is about 80% of the U.S. workforce.

The BLS conducts this survey every month, collecting data from businesses of all sizes. A large corporation with 50,000 employees contributes to the count; so does a small bakery with five staff members. Because the survey covers a fixed sample of about 130,000 establishments, the BLS uses statistical models to estimate the total national payroll figure. This method is reliable but not perfect — and this is why revisions happen.

A typical monthly release shows three numbers:

  • The headline non-farm payroll change (e.g., "jobs rose by 275,000")
  • The unemployment rate (e.g., "fell to 3.8%")
  • The average hourly earnings (e.g., "rose 0.3% month-over-month")

Of these, the headline payroll number is the most watched, though economists and traders also scrutinize wage growth and the unemployment rate reported in the same release.

Why the Fed cares deeply about non-farm payrolls​

The Federal Reserve has two core mandates: maximum employment and stable prices (low inflation). Non-farm payrolls speak directly to the employment mandate. When jobs are growing too fast — say, 400,000+ per month when the economy is already near full employment — wage pressures mount, potentially stoking inflation. When payrolls are shrinking or barely growing, that signals economic slack and potential deflation risk.

Consider the period after the 2008 financial crisis. Non-farm payrolls were deeply negative, shedding jobs for months. The unemployment rate hit nearly 10%. The Fed responded by keeping rates near zero and buying trillions in bonds to stimulate hiring. As payrolls recovered and jobs came back, the Fed gradually normalized policy. Years later, when payrolls strengthened and unemployment fell below 4% around 2018, the Fed raised rates aggressively because labor markets looked so hot.

More recently, after the pandemic shutdowns in 2020, the economy bounced back hard. Non-farm payrolls surged — months of 600,000 to 1 million new jobs. That abundance of hiring pushed up wages and prices alike. By 2022, the Fed was raising rates rapidly, citing strong labor markets as a key reason why inflation was so persistent. The Fed knew that with so many jobs available, workers could demand higher pay, which businesses would grant, which would feed price increases.

In short: strong non-farm payrolls → Fed sees reason to raise rates (or keep them high). Weak payrolls → Fed sees reason to cut rates (or pause hikes). This is why financial markets react sharply on the first Friday of every month. Traders are updating their bets on future Fed action based on the single number released at 8:30 a.m. Eastern Time.

The composition of non-farm payrolls: Which industries matter​

The headline number tells you total job change, but the subcomponents tell the real story. The BLS breaks down the monthly payroll report by major industry: goods-producing (manufacturing, mining, construction), service-providing (retail, hospitality, healthcare, finance, tech), and information.

Over the past two decades, the U.S. economy has shifted decisively toward services. In 2000, goods-producing industries accounted for about 20% of non-farm payrolls; today, it's about 15%. Services now dominate — healthcare, hospitality, retail, finance, and professional services together account for over 70% of jobs. This shift reflects both trade dynamics (manufacturing moved overseas) and structural changes (the U.S. economy became more service-oriented and less dependent on commodity production).

When non-farm payrolls are strong, it's useful to look where the strength is. If all the job growth is in hospitality and leisure — sectors that typically pay less — wage growth might lag despite headline numbers looking healthy. By contrast, job growth in professional services, finance, or tech signals higher average wages and stronger consumer purchasing power. In January 2024, for example, the U.S. added 353,000 jobs, but much of that was in healthcare and leisure services; construction and manufacturing were comparatively weak. This suggested mixed underlying momentum.

The "diffusion index" — the share of industries adding jobs rather than shedding them — is another useful gauge. When diffusion is broad (say, 80% of industry groups gaining jobs), it signals robust growth. When it narrows (60%, or lower), it suggests growth is concentrated in a few sectors, which can be fragile. A recession often begins with a narrowing diffusion index: first a few industries weaken, then the weakness spreads.

Revisions: The non-farm payroll story never ends on day one​

One of the most misunderstood aspects of non-farm payrolls is that the number released on the first Friday is preliminary. The BLS revises the prior two months' figures in the current month's report. These revisions are often substantial.

Here's why: the CES survey is based on a sample of about 130,000 businesses. To estimate the total, the BLS uses statistical models that correct for nonresponse, seasonality, and other survey quirks. Inevitably, the model's estimate is inexact. As more data comes in from late filers and as the BLS receives reports from new businesses or closures, the prior estimates improve. The revisions in the following months correct for these modeling errors.

In practice, revisions can swing the narrative. In December 2023, the BLS initially reported 256,000 jobs added. The following month, that figure was revised down to 165,000. In March 2024, it was revised down again to 105,000. Over three months, the December print lost 150,000 jobs — nearly 60% of the headline. This doesn't mean the initial report was false, but it underscores that the true number is unknown on the day of release.

Investors and Fed watchers who obsess over the initial headline number while ignoring the prior months' revisions are missing a critical part of the labor market story. A savvy analyst always looks at the three-month average of payroll growth, which smooths out the noise. The three-month average is far more stable and predictive of future trends than any single month.

A recent example: in 2024, month-to-month non-farm payroll figures were volatile, ranging from 100,000 to 400,000, but the three-month average hovered around 200,000, suggesting stable underlying job creation despite headline volatility.

Non-farm payrolls and wage growth: The wage-price spiral story​

One reason the Fed has been so focused on non-farm payrolls is their link to wage growth. When non-farm payrolls are strong and unemployment is low, workers have more bargaining power. They can shop around for better offers. Businesses, eager to fill positions and worried about losing staff, bid up wages to compete for labor.

This dynamic played out vividly from 2021 to 2023. As the economy reopened after the pandemic, non-farm payrolls soared — but so did quits, as workers left jobs to find better-paying opportunities. Businesses, desperate to fill vacancies, raised wages aggressively. Average hourly earnings — reported alongside non-farm payrolls — rose at rates not seen since the 1980s. The BLS reported wage growth accelerating to 5%+ annually. This sounds great for workers, but it posed a problem for the Fed: faster wage growth could feed inflation if it outpaced productivity gains.

The link between payrolls and wages is not mechanical. During recessions, payrolls fall but wages often continue to grow (because low-skilled, low-wage workers are laid off first, raising the average). During booms, the relationship can be loose too if immigration surges or discouraged workers re-enter the labor force. Still, the broad relationship holds: tight labor markets (strong payrolls, low unemployment) tend to push wages up, and the Fed worries that wage growth above productivity growth is unsustainable and will drive inflation.

By 2024, as Fed policy tightened and payroll growth slowed, wage growth also moderated — from 5%+ to closer to 4%. This gave the Fed confidence that the wage-price spiral was cooling and that it would eventually be safe to cut rates. The non-farm payroll number, then, is not just an employment indicator — it's a crucial piece of the wage and inflation story.

Market reactions to non-farm payroll surprises​

Non-farm payrolls are one of the rare economic releases that move the market directly. Stock futures swing sharply at 8:30 a.m. on the first Friday. Bond yields move. The dollar strengthens or weakens. Why? Because the market is repricing its expectations for Fed policy and, by extension, for future economic growth and earnings.

The key is the surprise — the difference between what economists expected and what actually came in. If economists forecast 250,000 jobs and the actual number is 250,000, the market yawns. But if it comes in at 400,000 (a large surprise to the upside), traders interpret this as "the Fed will need to stay hawkish longer," which can temporarily depress equity valuations. Conversely, if it comes in at 100,000 (a large miss to the downside), traders might expect rate cuts sooner, which can boost equities on the assumption that lower rates boost growth and earnings.

A detailed example: in March 2024, non-farm payrolls came in at 275,000, above the 200,000 estimate. The market initially sold off, interpreting the strong number as a reason for the Fed to keep rates high. But as the day wore on, traders noticed wage growth had actually slowed month-over-month, and some revised their expectations downward, pushing equities back up. The headline payroll beat masked a more nuanced story about cooling wage pressures — exactly the kind of interpretation that keeps the market interesting.

The relationship between non-farm payrolls and the unemployment rate​

Most people assume that strong non-farm payrolls always mean a falling unemployment rate, and vice versa. In reality, the relationship is looser than it seems. Both numbers come from the same monthly release, but they come from different surveys — non-farm payrolls from the employer survey (the CES), and the unemployment rate from the household survey (the Current Population Survey, or CPS).

The household survey asks a sample of about 60,000 households whether members are employed, unemployed, or out of the labor force. The unemployment rate is the share of the labor force that is actively job-hunting. The labor force itself can grow or shrink depending on whether discouraged workers re-enter or exit.

This means you can have a strong payroll number without the unemployment rate falling, because the labor force might expand at the same time (bringing in formerly discouraged workers). You can also have weak payrolls with a steady or falling unemployment rate if people drop out of the labor force. This disconnect has become more common in recent years as participation dynamics have shifted — aging populations retire, and young people delay workforce entry in response to recessions.

A practical example: In late 2023 and early 2024, non-farm payrolls were slowing (trend around 200,000–250,000 per month), but the unemployment rate held steady around 3.8%. This seemed contradictory until you noted that labor force participation was flat — people were neither entering nor leaving the workforce in large numbers. The labor market was cooling, but not so much that it pushed unemployment up.

Seasonal adjustment: Why the numbers don't match reality​

A final nuance: all published non-farm payroll figures are seasonally adjusted. The raw (unadjusted) data is messier. Retail always hires before Christmas; construction always slows in winter. The BLS removes these predictable seasonal patterns to show the "trend" in hiring.

The seasonal adjustment method is statistical, and it sometimes makes mistakes or needs to be recalibrated. In unusual years — say, after a pandemic when normal patterns are disrupted — seasonal adjustment can produce surprising results. During the 2020 pandemic shutdown, traditional seasonal patterns broke down entirely, making it harder to interpret the adjusted figures.

This is why the BLS publishes both the seasonally adjusted and unadjusted figures. The seasonally adjusted headline is what hits the news, but savvy analysts also look at the unadjusted trend and compare year-over-year changes (which sidestep seasonal issues entirely).

The great financial crisis (2008–2009): Non-farm payrolls shed 8.7 million jobs over roughly 24 months, peaking in January 2008 at 138.4 million and bottoming in February 2010 at 129.7 million. The decline was broad-based: manufacturing, construction, and retail all contracted sharply. The unemployment rate hit 10% in October 2009. The recovery took years. It wasn't until 2014 that non-farm payrolls fully recovered to pre-crisis levels.

The pandemic shock (March–April 2020): Non-farm payrolls fell 22 million in two months — the sharpest drop in modern history. Hospitality and leisure were hit hardest; retail also declined sharply. But the rebound was equally swift and unusual. By the end of 2020, payrolls had recovered most losses. By 2021, they surged past pre-pandemic levels, reaching new records.

The 2021–2022 era of strong payrolls: From mid-2021 through mid-2022, non-farm payrolls grew at an average pace of 450,000+ per month — well above the 200,000 consensus estimate of "healthy" growth. Job openings exceeded job seekers. The Fed kept rates near zero despite this strength, citing supply-chain disruptions. By late 2022, the Fed pivoted to rapid rate hikes, partly in response to realizing that labor markets were too hot. This strong payroll growth set the stage for wage inflation and the inflation boom of 2021–2023.

The slowdown of 2023–2024: As the Fed raised rates through 2023, non-farm payroll growth moderated. Monthly gains fell from 450,000+ to 200,000–300,000. This gradual deceleration signaled that the Fed's policy was working — cooling labor demand without a sharp shock. By early 2024, the trajectory suggested a "soft landing" was possible: growth slowing without recession.

The employment impact of payroll changes​

Common mistakes when interpreting non-farm payrolls​

Mistake 1: Treating the initial headline as gospel. The first number released is preliminary. Revisions over the next two months can be substantial. Always ask: "What did last month's number get revised to?" before drawing conclusions.

Mistake 2: Ignoring composition and focusing only on the headline. A headline showing 300,000 jobs added is meaningless without knowing where the growth is. If it's all in low-wage hospitality, that's different from growth in high-wage professional services. Check the industry breakdowns.

Mistake 3: Expecting payrolls and the unemployment rate to always move together. They come from different surveys and can diverge for months or quarters. When they disconnect, it's a signal that labor force participation or other dynamics are in flux.

Mistake 4: Missing the wage story. Non-farm payrolls are only part of the employment picture. Average hourly earnings, released at the same time, tell you whether the jobs being created pay well. This is crucial for understanding wage-price dynamics.

Mistake 5: Over-weighting a single month's surprise. Financial markets overreact to monthly payroll surprises all the time. A better strategy is to focus on the three-month trend, seasonal adjustments, and revisions. Noise is high; signal is lower.

FAQ​

What time is the non-farm payrolls report released?​

The report is released at 8:30 a.m. Eastern Time on the first Friday of the month (or sometimes the first business day after the first Friday if Friday falls on a holiday). The data covers employment activity in the prior month.

How long does it take for the final payroll number to be locked in?​

The BLS releases an initial estimate, then revises the prior two months in the next release. Additional revisions can occur for up to three years via annual benchmark revisions. For practical purposes, the final payroll figure is known after about three months, though minor tweaks can continue.

Can non-farm payrolls be negative, and what does that mean?​

Yes. When businesses reduce staff or close, non-farm payrolls can be negative (e.g., "–180,000" means 180,000 jobs were lost that month). Negative payrolls are rare in normal times and typically signal a recession. The financial crisis, the pandemic shutdown, and the 1990–1991 recession all saw negative payroll months.

How does non-farm payroll data differ from the unemployment rate?​

Non-farm payrolls measure the total number of jobs (a stock), while the unemployment rate measures the share of the labor force that is jobless (a ratio). Payrolls come from the employer survey; the unemployment rate comes from the household survey. They can diverge for months.

What is considered a "strong" non-farm payroll number?​

This depends on the economic moment. During recessions, any positive number is strong. In normal times, 150,000–250,000 per month is considered healthy (roughly equal to population growth and labor-force expansion). Numbers above 300,000 per month suggest the economy is running hot and wage pressure may build.

Do non-farm payrolls include gig workers like Uber drivers?​

No. Non-farm payrolls only count traditional, on-payroll employees. Gig workers are counted in the household survey (unemployment rate) if they report income and actively job-seek, but not in the CES payroll survey.

Why do economists and the Fed focus so much on non-farm payrolls if the data is preliminary and subject to revisions?​

Non-farm payrolls are the earliest and most reliable signal of labor market health. While revisions happen, the initial estimate is broadly directional and captures the month-to-month trend accurately enough for policy decisions. The Fed and markets use it as one input among many, not the only input.

Summary​

Non-farm payrolls, released monthly by the Bureau of Labor Statistics, measure the total change in jobs outside farming and government. This single number shapes Fed policy, moves financial markets, and reveals whether the economy is accelerating or slowing. While preliminary and subject to revision, the payroll report is the most timely labor market indicator available. Understanding the distinction between headline payrolls and industry composition, between payroll growth and wage growth, and between the payroll survey and the unemployment rate is essential for anyone trying to decode the economy. The first Friday of every month, the non-farm payroll figure arrives as the starting gun for a month of economic interpretation.

Next​

→ The JOLTS report for investors