Guide

Nonfarm payrolls explained

On the first Friday of each month at 8:30 a.m. Eastern, the Bureau of Labor Statistics (BLS) publishes the Employment Situation report. The headline that moves S&P futures within seconds is nonfarm payrolls (NFP) — the net change in jobs on business payrolls, excluding farm workers, private household employees, and a few other categories. NFP is the market's shorthand for hiring momentum, but the number you trade is only the tip of a 40-page release: prior-month revisions, private vs government splits, average hourly earnings, average weekly hours, and detailed sector tables that reveal whether growth is broad or concentrated in healthcare and government while manufacturing sheds workers. This guide explains how payrolls are counted, what the companion statistics mean, how revisions and seasonal adjustment work, how NFP links to Fed policy and GDP, a Harbor Logistics release-day read worked example, an indicator decision table, pitfalls, and a checklist — alongside our unemployment rate guide, monetary policy overview, and economic calendar explainer.

What nonfarm payrolls measure

Nonfarm payroll employment comes from the Current Employment Statistics (CES) program — a monthly survey of roughly 122,000 businesses and government agencies covering about 631,000 individual worksites. Each establishment reports how many people received pay during the pay period that includes the 12th of the month. The BLS publishes the seasonally adjusted change in total nonfarm jobs from the prior month, usually rounded to the nearest thousand.

"Nonfarm" excludes agricultural workers (highly seasonal and poorly captured by establishment surveys), unpaid family workers, and workers in private households (nannies, cleaners). It includes most wage and salary employees in the private sector plus federal, state, and local government workers. Self-employed and gig workers are largely absent — they appear in the separate household survey that produces the unemployment rate.

Core concepts

Headline NFP — net monthly change in seasonally adjusted nonfarm payroll jobs.
Total private payrolls — excludes government; markets watch this when fiscal hiring distorts the headline.
Goods-producing vs services-providing — manufacturing, construction, and mining vs the much larger services block.
Diffusion index — share of industries adding jobs; breadth indicator beyond the aggregate.
Benchmark revision — annual true-up to unemployment insurance tax records; can rewrite prior-year levels.

Payrolls vs unemployment: two surveys, one release

The Employment Situation combines the establishment survey (payrolls, hours, earnings) and the household survey (unemployment, participation, demographics). They measure different things and routinely diverge in a single month.

Payrolls count jobs; the household survey counts people. Someone holding two part-time jobs adds one to payrolls if both employers respond, but can count as one employed person in the household survey. Payroll data exclude the self-employed; household employment includes them. Payrolls have a smaller relative sampling error for month-to-month changes in total employment, which is why traders anchor on NFP for cyclical momentum — but the unemployment rate captures labor-force dynamics payrolls miss entirely.

A strong NFP print with rising unemployment usually means participation increased: more people started looking for work than payroll growth absorbed. Weak payrolls with falling unemployment can mean discouraged workers left the labor force. Read both surveys in the same breath; our unemployment guide walks through U-3, U-6, and the Sahm rule in detail.

Average hourly earnings and weekly hours

The establishment survey publishes average hourly earnings (AHE) for production and nonsupervisory workers — about four-fifths of private payroll employment. AHE is not a perfect wage index (composition shifts — hiring low-wage retail vs high-wage tech — move the average without anyone getting a raise), but it is timelier than quarterly employment cost indexes and feeds the inflation narrative alongside CPI and services inflation.

Average weekly hours signal labor hoarding vs demand destruction before headcount cuts. In early downturns, employers often trim hours and overtime while keeping bodies on payroll; falling hours preceded negative NFP months in several past cycles. Watch manufacturing overtime hours especially — it correlates with industrial production and goods demand.

Annualizing wage growth

Traders often annualize the month-over-month AHE change: ((1 + m/m_change)^12 - 1), or simply multiply m/m by 12 for a rough estimate. A +0.3% m/m print implies roughly 3.6% annualized wage growth — enough to keep the Fed attentive if productivity growth is only 1–2%. Pair AHE with CPI wages-sensitive components ( shelter lagged, services ex energy ) rather than treating either series as a standalone trigger.

Revisions, seasonality, and the birth-death model

The first NFP print is preliminary. The BLS revises the prior two months each release as late survey responses arrive. A "+200k" headline can become "+160k" after revisions — or the opposite. Markets sometimes fade the initial reaction once revision lines hit the wire thirty seconds later.

Seasonal adjustment

Raw payrolls swing predictably: retailers hire for holidays, schools shed staff in June, construction pauses in winter. Seasonal adjustment removes these patterns so January's unadjusted -2.5 million jobs might translate to +180k seasonally adjusted. When seasonal patterns break (pandemic reopening, unusual weather), adjusted numbers get noisy for several months.

Birth-death adjustment

New businesses form and old ones close between survey updates. The BLS uses a net birth-death model to estimate jobs created by firms too new to appear in the sample. Critics blame the model for overstating jobs late in expansions and understating losses early in downturns. The adjustment is transparent in BLS tables — check whether headline NFP would have been negative without it during soft patches. Annual benchmarks reconcile payroll levels to unemployment insurance counts, occasionally producing large one-time level shifts that do not affect month-to-month changes but rewrite history charts.

Sector breakdown: where jobs are actually added

Headline NFP can mask rotation. A +250k month driven entirely by government and healthcare hiring tells a different story than +250k led by construction and manufacturing. The BLS publishes seasonally adjusted changes by major industry:

Leisure and hospitality — cyclical, wage-sensitive; recovered slowly post-2020.
Healthcare and social assistance — structural growth from aging demographics; often positive even in mild slowdowns.
Professional and business services — includes temp help, a leading indicator of corporate caution when it turns negative.
Manufacturing — ties to capex and export demand; watch with PMI and industrial production.
Construction — residential vs nonresidential; links to housing starts and mortgage rates.
Retail trade — correlates with retail sales but employment lags store closures.
Government — federal, state, local; census and election years add volatility.

The diffusion index for private industries shows what fraction of sectors added jobs. A falling diffusion index with positive headline NFP means narrow, concentrated hiring — less durable than broad-based gains.

ADP, claims, and the run-up to jobs Friday

No single preview predicts NFP, but the calendar builds context:

ADP National Employment Report — private payroll estimate two days before NFP (Wednesday). Methodology differs from BLS; correlation is moderate. Large ADP/BLS divergences make consensus less reliable.
Initial jobless claims — weekly; four-week average trends ahead of monthly payroll turning points. Sub-250k sustained often coincides with +150k+ NFP; rising claims above 280k warn of cooling.
ISM manufacturing and services employment sub-indexes — diffusion-style hiring sentiment; see our PMI guide.
JOLTS — released later in the month; openings and hires confirm or contradict the NFP trend.

Map the full sequence in the economic calendar so you know which data is stale by the time NFP hits.

Worked example: Harbor Logistics release-day read

Harbor Logistics — a fictional freight and warehousing operator in our recurring macro examples — runs a monthly "jobs Friday" desk note for its treasury team. Suppose the BLS releases: NFP +95k (consensus +175k), prior two months revised -58k combined, unemployment 4.2% (+0.1pp), average hourly earnings +0.4% m/m, average weekly hours 34.2 (unchanged). Private payrolls +72k; government +23k. Manufacturing -12k; transportation and warehousing +3k; healthcare +58k.

Harbor's analyst writes:

Large miss with negative revisions — hiring momentum weaker than the prior narrative; three-month average payroll growth now ~120k vs 200k+ earlier in the year.
Earnings beat despite soft jobs — composition effect (healthcare and high-wage professional services adding while low-wage leisure flat) or genuine wage pressure in tight niches; not automatically dovish.
Manufacturing and transport weak — directly relevant to Harbor's end markets; confirms soft freight volumes in their internal data.
Healthcare dominance — narrow breadth; diffusion index likely below 55; cyclical sectors not carrying the expansion.
Policy read — markets price one extra 2026 rate cut; front-end yields fall; equities rally on "Fed put" reflex despite earnings risk. Harbor hedges fuel costs but does not accelerate hiring plans.

The lesson: a single miss matters less than the revision trend, sector mix, and whether wages and hours confirm or contradict the slowdown story.

Indicator decision table

Pattern	What it suggests	Typical market read
NFP beat, positive revisions, rising hours	Broad hiring strength	Higher yields, cyclical equity bid; Fed stays restrictive
NFP miss, negative revisions, hours falling	Demand cooling, possible layoffs ahead	Rate-cut pricing, duration rally; cyclicals underperform
Strong NFP, weak AHE, hours flat	Job growth without wage pressure	"Goldilocks" equity bid; bonds stable
Weak NFP, strong AHE	Stagflationary mix or composition noise	Confused reaction; favor quality, short duration
Two consecutive negative NFP months	Cyclical downturn in employment	Recession playbook; recession indicators in focus
Payrolls strong, unemployment up, participation up	Supply returning to labor force	Moderate risk-on; watch if absorption catches up

Common pitfalls

Trading the first print only — revision lines arrive in the same second; read the full table before sizing risk.
Ignoring private payrolls — government hiring (census, education) can inflate headlines during fiscal expansions.
Annualizing one month of AHE — volatile series; use three-month or year-over-year trends for policy conclusions.
Expecting payrolls to match ADP — different samples and methods; ADP is context, not a lock.
Overweighting birth-death conspiracy — check the actual adjustment magnitude; it is often smaller than social media claims.
One-month NFP for recession calls — weather, strikes (UAW, Hollywood), and pandemic base effects distort single prints; use three-month averages and claims.
Forgetting benchmark years — level revisions can break your spreadsheet models without changing the economic story if trends persist.

Investor checklist

Record headline NFP, private payrolls, unemployment, participation, AHE m/m and y/y, and weekly hours vs consensus.
Note prior-month revisions and compute a three-month moving average of payroll growth.
Scan sector tables and the diffusion index for breadth.
Compare manufacturing and temp-help trends to PMI and corporate guidance in your sectors.
Cross-check claims four-week average and last ADP print for consistency.
Map the result to the Fed reaction function via monetary policy — employment AND inflation mandates both matter.
Update recession dashboards (Sahm, yield curve) if payroll trend breaks below ~100k sustained.

Key takeaways

Nonfarm payrolls count paid jobs on business and government payrolls — not people, and not the self-employed.
Revisions, seasonal adjustment, and the birth-death model mean the first headline is provisional.
Average hourly earnings and weekly hours translate hiring into inflation and margin signals.
Sector detail and diffusion indexes reveal whether growth is broad or narrowly supported.
Professional macro reads combine NFP with unemployment, claims, ADP, and wage data — never one line in isolation.