Every week, people are handed another piece of US jobs data and told it matters for markets.
News platforms, including Invezz, report initial and continuing jobless claims every week, before attention shifts through the month to ADP employment, JOLTS and the closely watched nonfarmpayrolls report.
For anyone who does not spend their day watching economic calendars, it can feel like several versions of the same number.
They are not.
Each report looks at a different part of the labour market. Some tell us whether people are losing jobs.
Others show whether businesses are hiring, how long unemployed workers are struggling to find work, or whether employees feel confident enough to quit.
That distinction matters even if you have no intention of trading the next jobs report.
Employment drives household income and consumer spending. Wage growth can influence inflation. Both feed into Federal Reserve decisions, which in turn affect Treasury yields, the dollar, equities and eventually borrowing costs faced by households and businesses.
The differences are especially relevant now.
Initial jobless claims fell to 203,000 in the week ended August 22, keeping layoffs historically low, while continuing claims stood at 1.778 million in the prior week.
Yet July’s official employment report showed nonfarm payrolls falling by 23,000, while the unemployment rate edged down to 4.1%.
Meanwhile, ADP estimated that private employers added 44,000 jobs in July.
Confusing? Only if we expect all of those numbers to answer the same question.
Claims show whether jobs are disappearing, not whether hiring is healthy
Start with the data investors see most often.
Initial jobless claims are released weekly and count new applications for unemployment benefits. In simple terms, they are one of the quickest ways of spotting a rise in layoffs.
If claims suddenly move sharply higher and remain elevated, it can signal that companies are cutting workers more aggressively.
Low claims tell us something useful too: employers are largely holding on to the people they already have.
When claims fell to 187,000 in July, their lowest level since 1969, Oxford Economics senior US economist Matthew Martin said the extremely low level highlighted the labour market’s “low layoff rate.”
But there is an obvious limitation.
A company deciding not to fire you tells us very little about whether another company wants to hire you.
That is where continuing claims become useful. These measure people who have already made an initial claim and remain on unemployment benefits in subsequent weeks.
They can therefore provide clues about how difficult it is for someone who loses a job to get back into work.
Even here, the headline number needs care.
Research published by the Federal Reserve Bank of Richmond in July found that continuing claims had been declining largely because fewer people were entering unemployment, not because existing claimants were leaving unemployment more quickly.
The weekly exit rate from unemployment insurance was around 11.7% in the first half of 2026, compared with 15.2% in 2022, according to the research. That gives us an important distinction:
Initial claims ask: are more people losing their jobs?
Continuing claims ask: what is happening to those already unemployed?
A labour market can look healthy on the first measure and considerably less comfortable on the second.
Payrolls, ADP and JOLTS look at different stages of the jobs cycle
Then comes the monthly barrage.
Nonfarm payrolls, usually abbreviated to NFP, form part of the Bureau of Labor Statistics’ monthly Employment Situation report and are generally the headline number Wall Street watches most closely.
But even the jobs report itself contains two major surveys.
The establishment survey measures payroll employment, hours and earnings at businesses and government agencies. It produces the nonfarm payrolls number.
The separate household survey measures whether people are employed, unemployed or outside the labour force. That is where the unemployment rate and labour-force participation rate come from.
The two can move differently because they use different samples and definitions. BLS itself says both are needed to obtain a complete picture of the labour market.
This is why a fall in payrolls does not automatically require the unemployment rate to rise in the same month.
And it is why staring only at the big payroll number can be misleading.
Following July’s 23,000 payroll decline, LinkedIn’s head of economics for the Americas, Kory Kantenga, told Yahoo Finance that investors should “scroll past that big headline number,” pointing to unusual seasonal effects in local government education employment.
ADP adds another layer.
Its National Employment Report uses anonymised payroll records covering more than 26 million US employees to estimate changes in private-sector employment.
Crucially, ADP explicitly says the report is an independent measure and is not intended to forecast the BLS nonfarm payroll report.
That point is often lost.
If ADP says private employment rose and NFP subsequently disappoints, it does not necessarily mean ADP “got payrolls wrong.”
The two reports use different datasets and methodologies, while official NFP also includes government employment.
ADP can also provide information that the headline hiring number misses. In July, for example, it reported 44,000 additional private-sector jobs but also found annual pay growth for job-changers accelerating to 7%.
“Job-changers are highly sensitive to real-time economic conditions,” ADP chief economist Nela Richardson said.
Then there is JOLTS: the Job Openings and Labor Turnover Survey.
Instead of asking only how many people are employed, JOLTS looks inside the machinery of the jobs market: vacancies, hires, quits, layoffs and other separations.
June data showed about 7.4 million openings, 5.3 million hires and 3.2 million quits.
Openings provide a measure of employer demand. Hires show whether those intentions are turning into actual jobs.
Quits can provide clues about worker confidence, since people are usually more willing to leave voluntarily when they believe another opportunity is available.
Again, one headline should not dominate the interpretation.
ZipRecruiter labour economist Nicole Bachaud told HR Brew that she would not focus “too heavily on a one-month drop in openings” when other parts of the data were showing relative stability.
Why investors should care when the numbers disagree
The easiest mistake is to decide that one jobs report is right and another is wrong. Often, they are simply describing different stages of the same labour market.
Imagine companies stop advertising as many vacancies but are reluctant to fire existing staff.
JOLTS openings could weaken while initial claims remain low.
Hiring could then slow further, leaving people who do lose their jobs searching for longer. Continuing claims could rise even though layoffs remain subdued.
Payroll growth might not turn negative until later.
None of those signals contradicts the others.
Together, they describe a labour market moving from strong hiring towards what economists often call a low-hire, low-fire environment.
Understanding that sequence is crucial for markets because the Fed has both inflation and employment considerations when setting monetary policy.
A surprisingly strong jobs report, particularly when accompanied by rapid wage growth, can strengthen the argument for tighter monetary policy if investors think labour demand could keep inflation elevated.
Weak employment data can pull the other way by increasing concern about economic growth and the employment side of the Fed’s mandate.
That reassessment can quickly move interest-rate expectations, Treasury yields, the dollar and equity valuations.
After July’s payroll disappointment, Charlie Ripley, senior investment strategist at Allianz Investment Management, told Fortune that the report put the “spotlight back on the employment side of the Fed’s mandate”.
The implications are less abstract for households.
Initial claims can tell you whether job security is deteriorating. Continuing claims can offer clues about how hard finding another role has become.
JOLTS can show whether employers still have an appetite to hire. Wage data tells you how much bargaining power workers retain.
And payrolls pull much of that information together into the broadest monthly snapshot markets receive.
So, if you only have a few minutes when the next round of jobs data lands, do not ask which number is the “real” one.
Ask what question each number is trying to answer.
The post Invezz Explains: why you should closely watch payrolls, claims, ADP and JOLTS data appeared first on Invezz
