US Macro

Non-Farm Payrolls — Monthly Change

Non-Farm Payrolls measures the net monthly change in employed workers across all US industries except farming, government domestics, and proprietors. It's the single most market-moving economic release globally — published the first Friday of every month at 8:30 AM ET, watched obsessively by traders worldwide, and used by the Federal Reserve as the highest-frequency read on US labor market health.

May 14, 2026PAYEMSlabor-market · us-macro · fed-policy · bls

NFP

—

There is no single economic data release globally that produces more synchronized market reaction than the first-Friday-of-the-month Non-Farm Payrolls release. From 8:25 AM ET to 8:45 AM ET on those days, every trader at every desk in New York, London, and Tokyo has their eyes on the same screen, the same number, the same Federal Reserve implications. The minute-by-minute volume profile on NFP mornings is among the most concentrated of any trading day. NFP is the canonical macro release.

NFP

—

What it measures

NFP is the monthly net change in non-farm payroll employment in the United States:

The underlying level series — FRED PAYEMS — is the total non-farm payroll employment in thousands of jobs. The "change" version (NFP) is derived from the level. Our nfp indicator displays the period delta — the monthly increase or decrease in jobs.

Release schedule: first Friday of each month, 8:30 AM ET, as part of the Bureau of Labor Statistics' "Employment Situation Summary." The release also includes the unemployment rate (separately published, from the household survey), average hourly earnings, average weekly hours, and various sectoral breakdowns. NFP is the headline number.

Typical print: roughly 150-200K jobs per month is the post-COVID baseline. Job growth of 250K+ is widely interpreted as strong; below 100K as weak. The all-time peak monthly NFP was June 2020 at +4.8 million as COVID layoffs reversed; the trough was April 2020 at -20.7 million.

Why it matters

Two angles.

The Fed-policy-input angle. NFP is the most reliable monthly signal of US labor market strength, and labor market strength is half of the Federal Reserve's dual mandate. Fed decisions are heavily influenced by the rolling 3-month average of NFP. When NFP is consistently above 250K, the Fed sees a labor market that's running hot and worries about inflation pressure. When NFP is below 100K consistently, the Fed sees softening and considers cuts. The September 2024 jumbo 50bp cut was triggered partly by NFP weakness — the August 2024 release had shown +142K jobs (below expectations) plus large downward revisions to prior months, plus the broader benchmark revision showing +818K fewer jobs created over the prior year. Together, these signals shifted the Fed's reaction function decisively.

The recession-signal angle. Recessions are typically declared by the NBER based on a combination of indicators, but NFP is often among the first to weaken. The 3-month rolling average of NFP turning negative is widely watched as a recession signal (though it can lag the actual recession start by 1-3 months). The 2008 recession was visible in NFP by early 2008 (consecutive negative months); the 2001 recession was visible by mid-2001. The 2024 weakness in NFP, especially combined with the Sahm Rule trigger on unemployment, was the most significant recession-warning signal of the post-COVID cycle.

What moves it, and what it moves

Moves NFP:

Business-cycle momentum. Strong consumer spending and business investment lead to hiring; weakness leads to slowing hiring or layoffs. With a 1-3 month lag.
Fed policy with a multi-quarter lag. Aggressive tightening eventually slows hiring; aggressive easing eventually accelerates it. The 2022-2023 hiking cycle produced visible NFP slowdown in 2024.
Sector-specific dynamics. Construction is rate-sensitive (mortgage rates affect homebuilding employment); tech is investor-sentiment-sensitive; healthcare and government grow steadily regardless of cycle.
Immigration. Labor force growth — which feeds into payroll employment over time — has been a major source of variance. Post-2024 immigration restrictions have already produced visible labor-force growth slowdowns.
Survey timing. The CES survey is taken during the pay period including the 12th of the month. A natural disaster (hurricane, wildfire) that affects employment during that week can produce 50K+ swings in the headline.

NFP moves:

Fed policy expectations (most directly — single largest single mover of Fed Funds futures).
UST2Y and UST10Y bond yields.
Equity markets (large moves at 8:30 AM ET on release days).
The US dollar (DXY).
Single-stock equity moves in employment-sensitive sectors (REITs, homebuilders, consumer discretionary).

A worked example: the August 2024 benchmark revision shock

Through 2023 and into early 2024, the US economy was producing what looked like a steady 200-250K NFP per month. The Fed had cited this strength as one reason for keeping rates elevated despite easing inflation.

The first signs of softening came in mid-2024:

July 2024 NFP (released August 2, 2024): +114K — well below the 175K consensus
Plus downward revisions to May (-2K) and June (-27K)
Plus the unemployment rate had risen from 4.1% to 4.3% in the same release (triggering the Sahm Rule)

The market reacted sharply — equity sold off, UST yields fell, Fed Funds futures repriced. But the bigger shock came two weeks later:

August 21, 2024: BLS released its preliminary annual benchmark revision. The revision indicated that for the 12 months ending March 2024, the US economy had created approximately 818,000 fewer jobs than the monthly CES reports had suggested. This was the largest annual benchmark revision in modern history.

The implications:

The labor market had been weaker than thought for the past year.
Fed policy had been calibrated against a labor market that was, in retrospect, less robust than data suggested.
The September 2024 FOMC meeting now faced a different decision than markets had been pricing for months.

The Fed delivered a jumbo 50bp cut on September 18, 2024 — its first cut of the cycle, larger than the 25 bp markets had been expecting before the benchmark revision came out. Powell's press conference explicitly cited labor market data — including the benchmark revision — as a factor.

The episode demonstrated several lasting lessons:

NFP revisions can be substantial enough to reshape Fed policy.
The labor market signals can deteriorate before the unemployment rate fully reflects it (NFP is a leading indicator of UNRATE in some scenarios).
The combination of NFP weakness + Sahm Rule trigger + benchmark revision was a powerful trio of labor market signals.

The current cycle, and the open question

The structural debate:

Trend NFP rate. What's the underlying "right" pace of US job creation in equilibrium? Demographics (slower labor force growth, especially post-immigration restrictions) suggest a slowdown from the 2010s pace of ~180K/month to ~100-130K/month structurally. Whether this is currently underway or whether the recent weakness is cyclical is debated.
Immigration's role. Estimates suggest 2022-2024 immigration added 1-2 million workers per year to the US labor force, enabling stronger NFP growth than demographics alone would imply. Post-2024 immigration restrictions could substantially reduce sustainable NFP.
Sector composition shifts. Hiring has been concentrated in healthcare, government, and a few other services sectors. The cyclical sectors (manufacturing, construction, business services) have been weaker. This composition shift makes interpreting headline NFP harder.
Revisions cadence. The 2024 benchmark revision was unusually large; subsequent quarterly NFP figures may also be revised meaningfully. Markets are pricing in the possibility of further revisions and weighting individual prints accordingly.

What to watch: the first-Friday release each month (NFP plus UNRATE plus average hourly earnings); the 3-month rolling average; the prior-2-months revisions (the 'revision tape'); the annual benchmark revision (typically in August); state-level UI claims data (a higher-frequency cross-check); and the JOLTS (Job Openings and Labor Turnover Survey) for additional labor market detail.

FAQ

Why is it called 'non-farm payrolls' and what's excluded?

The 'non-farm' qualifier exists because agriculture employment is volatile, seasonal, and historically less integrated with the rest of the economy. The Bureau of Labor Statistics produces separate agriculture-employment data; NFP excludes that to focus on the urban/industrial economy. Also excluded: workers employed in private households (housekeepers, nannies, caregivers paid directly by families), self-employed proprietors (unincorporated business owners), and certain government-related categories. The NFP figure covers approximately 158 million workers across private companies and the public sector. Despite the exclusions, it captures roughly 95% of US salaried employment.

How is NFP actually measured?

Via the Current Employment Statistics (CES) survey — sometimes called the 'establishment survey' — which BLS conducts monthly with a sample of approximately 140,000 businesses and government agencies (covering ~700,000 individual worksites). Companies report how many people they had on payroll during the pay period including the 12th of each month. This is different from the household-survey-based unemployment rate (which uses the CPS, a survey of households): the establishment survey counts JOBS (one person can have two jobs and be counted twice); the household survey counts PEOPLE. The two can diverge meaningfully in some months — when they do, economists often watch the household survey for trend signals because it's harder to construct via business-level sampling.

Why is NFP so market-moving?

Three reasons. First, it's the highest-frequency timely macro data: every month, more useful than quarterly GDP. Second, it directly informs the Fed's reaction function — labor market strength is half the Fed's mandate, and NFP is the most-watched signal of that. Third, the release is at 8:30 AM ET on a fixed schedule (first Friday of the month, with rare exceptions for holidays), giving the entire trading day to price the reaction. The combination produces moves: typical reactions to a ±50K surprise in NFP are 10-25 bps in UST2Y, 5-15 bps in UST10Y, 0.5-1.5% in SPX, and 1-3% in single-name growth stocks. The first hour after release is one of the highest-volume periods in any given trading day.

What's the deal with NFP revisions?

Substantial. Each NFP print is initially released as the headline number, but BLS revises it twice in the following two months as more establishment-survey data comes in. Some revisions are huge — the August 2024 release included an annual benchmark revision showing that the US had created approximately 818,000 FEWER jobs over the prior year than initial estimates had suggested. The benchmark revision is the largest single-revision in NFP history. Reasons for revisions: late-reporting establishments, sample expansion as more businesses respond, and BLS's annual benchmark of CES data against state unemployment insurance records (which is more complete but available with a 9-month lag). Traders who watch NFP closely now also watch the revision history — if Q3 revisions are systematically downward, that's a leading signal of labor market softening.

What's the 'birth-death adjustment' and why is it controversial?

Each month, BLS adds an estimate for jobs created at new businesses (births) and subtracts an estimate for jobs lost at businesses that closed (deaths). Because the CES survey can't capture these in real time (a new business isn't yet in the sampling frame), BLS uses a statistical model based on historical patterns. The adjustment can be +/- 100K jobs per month — meaningful relative to the headline number. Critics argue the model over-counts business formations during cyclical turns, leading to systematic over-estimation of NFP during slowdowns. The 2024 benchmark revision was partly attributed to over-estimated birth-death adjustments in 2023-2024. BLS continues to refine the methodology, but it remains a permanent feature of the monthly release that produces some uncertainty around the headline figure.