The August Jobs Report Lands September 4 and 3 Numbers Job Seekers Should Watch

The August Employment Situation is published on Friday September 4 at 8:30am Eastern, according to the Bureau of Labor Statistics release schedule. It is the single most consequential labour market release of the month, and the headline number is usually the least useful part of it. July payrolls actually fell by 23,000 and the two months before that were revised down by a combined 103,000, which tells you more about hiring conditions than any single month's print. At Metaintro, here are the three numbers worth reading before you react to the headline.

What exactly is being published on Friday?

The Employment Situation combines two separate surveys, which is the source of most public confusion about it. The establishment survey, published through the Current Employment Statistics programme, asks businesses and government agencies how many people are on their payrolls and produces the headline jobs number, along with average hourly earnings and the average workweek. The household survey asks people directly about their labour force status and produces the unemployment rate, the participation rate and the demographic breakdowns.

The two can and regularly do disagree in a given month, because they measure different things using different methods. A person holding two jobs counts twice in the establishment survey and once in the household survey. Someone self employed appears in the household survey and not on any payroll. Knowing which survey a number comes from is the difference between reading the report and repeating a headline, and it matters on Friday because the two have been telling slightly different stories for months. Commentary that treats the report as a single number will usually pick whichever survey supports the argument being made, so knowing the difference is a practical defence against being misled by people who are not deliberately misleading you.

Number one, what happens to the revisions?

This is the number almost nobody leads with and the one that has been most informative this year. In the July report, BLS revised May down by 66,000, from 129,000 to 63,000, and June down by 37,000, from 57,000 to 20,000. Combined, employment in those two months was 103,000 lower than previously reported. A revision of that size does not adjust the picture, it replaces it. Two months that looked like modest growth turned out to be close to flat. Revisions happen because more businesses and government agencies file returns after the first estimate is published, and because seasonal factors are recalculated. They are a normal feature of the data rather than a sign of error, but the direction they consistently run in is informative.

Watch Friday's revisions to June and July before you form any view on August. Persistent downward revisions usually mean the initial estimates are systematically overstating hiring, which is exactly what a turning point looks like in real time. The pattern has been consequential enough that revised data has quietly changed the labour market story more than once this year, and a separate annual benchmark process periodically resets the level entirely, as the preliminary benchmark revision did. If August prints a decent number and June and July are cut again, the cut is the real news. A useful habit is to write down the previous month's figure before the release, then check what it has become an hour later. Very few people do this, and it takes about ten seconds. It is also the fastest way to tell whether a commentator has actually read the release or is reacting to the headline alongside everyone else.

Number two, what is participation doing?

The unemployment rate is the most quoted figure in the report and the easiest to misread, because it can fall for a good reason or a bad one. In July it stood at 4.1 percent with 6.9 million people unemployed, little changed over the month and over the year. That sounds stable. Underneath it, the labour force participation rate was 61.4 percent, down 0.7 percentage points since January, and the employment to population ratio was 58.9 percent, down 0.5 points over the same period.

That combination matters enormously to a job seeker. A steady unemployment rate alongside falling participation means people are leaving the labour force rather than finding work, and someone who stops looking stops being counted as unemployed. It is the mechanism behind the finding that job seekers giving up has pushed participation toward a fifty year low. On Friday, read the participation rate and the employment to population ratio before the headline rate, because those two say whether the labour market is genuinely holding up or simply shedding participants. The July report also counted 5.9 million people not in the labour force who currently wanted a job, plus 1.8 million marginally attached to it and 476,000 discouraged workers who believed no jobs were available for them. None of those people appear in the unemployment rate at all, which is why the rate can look calm while the experience of looking for work does not.

Number three, how long are people staying unemployed?

Duration is the number that describes your actual experience of a job search, and it is buried in the household survey tables. In July the number of long term unemployed, meaning people jobless for 27 weeks or more, edged down to 1.8 million but accounted for 25.5 percent of all unemployed people. Roughly one in four people out of work had been out of work for more than six months. That is the difference between a labour market with a lot of churn and one where getting back in is hard. It also affects how employers read your application, since a long gap invites questions that a short one does not, and that effect compounds the longer a search runs. Anyone approaching the six month mark should treat that threshold as a real one and consider contract or interim work before crossing it, rather than after.

The July report also showed the number of people on temporary layoff rising by 153,000 to 921,000, while permanent job losers were little changed at 1.7 million. Temporary layoffs can reverse. Permanent separations generally do not. Watching whether the long term share rises again on Friday tells you whether to plan for a search measured in weeks or in quarters, and the trend has not been encouraging. Separately, 4.8 million people were working part time for economic reasons in July, meaning they wanted full time work but could not get it or had their hours cut, which is another form of underemployment the headline rate does not capture. The direction has been consistent, with long term unemployment rising through 2026 and a hiring rebound that has largely skipped both new graduates and the long term unemployed.

What did July actually show?

Worth having in mind as your baseline. Total nonfarm payroll employment fell by 23,000 in July, against an average monthly gain of 34,000 over the prior twelve months. Local government education lost 50,000 jobs and retail trade lost 19,000, with declines concentrated in warehouse clubs, supercenters and other general merchandise retailers. Financial activities continued to trend down, losing 14,000 and leaving employment in that sector 121,000 below its recent peak in May 2025. Health care continued to grow, adding 22,000, but at a slower pace than its 36,000 average over the prior year.

On pay, average hourly earnings for all private employees were 37.62 dollars, up 2 cents on the month and 3.2 percent over the year, while the average workweek held at 34.3 hours. Set that 3.2 percent against the cost pressures workers are facing and the picture is of nominal gains that do not travel far. Pay growth of 3.2 percent is neither alarming nor reassuring on its own, and the more revealing detail is that the average workweek has not moved, since employers usually cut hours before they cut people. A stable workweek alongside falling payrolls suggests the adjustment is happening through hiring freezes rather than through reduced schedules. For people already employed that is relatively good news. For people trying to get hired it is the opposite, because a freeze removes the opening before it is ever advertised. The overall shape is a labour market that is not collapsing and is not creating much either, which is why the question of whether it is quietly deteriorating keeps being asked.

Why is the ADP number not a preview of this report?

This is the mistake that gets repeated every month. On Wednesday the ADP National Employment Report showed private employers adding 38,000 jobs in August, the slowest pace since January, and Quartz reported it came in below a consensus estimate of 47,000. That is a genuine and useful data point about private sector hiring. It is not a forecast of Friday's figure.

ADP measures its own payroll records covering private employers. The BLS establishment survey covers government employment as well and uses a different sample and methodology, and the unemployment rate comes from a different survey again. The two routinely diverge in a given month and neither predicts the other. Read both, treat any confident prediction of one from the other with suspicion, and give more weight to the JOLTS data, which put July job openings at 7.3 million against 5.1 million hires with 1.7 million layoffs, describing a market where employers are neither hiring nor firing quickly. That combination is unusual and it is the defining feature of this period. In most slowdowns layoffs rise sharply, and the fact that they have not is why the unemployment rate has stayed near 4 percent while finding work has become measurably harder. If you are wondering why the public commentary and your own experience do not match, that gap is the explanation, and it is a genuine feature of the data rather than a failure of your search.

What does this mean for your job search right now?

Use the report to calibrate expectations rather than to make decisions. If revisions are negative and participation keeps falling, plan for a longer search and start earlier than feels necessary. If the long term share rises, prioritise staying formally attached to the labour market through contract, temporary or part time work, because a continuous record is easier to explain than a gap and keeps you counted among the employed rather than the discouraged. That is not only a statistical point. Recruiters read continuity the same way the survey does, and a contract role that pays less than your last salary still reads better on a record than an unexplained six month gap.

"According to our data, the average number of applications received for any position is in the approximate excess of over 250. Reaching the interview stage could mean that an applicant is in the top 2% of all applications. That in itself is a win," Lacey Kaelani, chief executive of Metaintro, told Upworthy. That is the right frame for a week like this one. A macro report describes the weather, not your chances, and in a frozen market where hiring and firing are both slow the returns come from targeting fewer roles with better evidence rather than from raising your application count. Read Friday's number, adjust your timeline, and then go back to the work that actually moves your search. The report will be discussed loudly for about a day and will change nothing about which specific employers need someone with your skills next week. Those two things are worth keeping separate.

Related Articles

• What a monthly jobs report means for your job search
• Healthcare hiring against tech layoffs in a monthly report
• What jobless claims say about the labor market
• Employee sentiment and the labor market
• Small employers and their hiring challenges
• Zero net job creation and the career impact
• The labor market splitting between office and construction work
• What to watch in a jobs report as a job seeker

People Also Asked

Q: When is the August jobs report released?

A: The Employment Situation for August 2026 is scheduled for Friday September 4 at 8:30am Eastern, according to the Bureau of Labor Statistics release schedule. The date was also confirmed in the July release itself. The report combines an establishment survey covering payrolls, earnings and hours with a household survey covering the unemployment rate and labour force participation.

Q: What should job seekers look at besides the headline number?

A: Three things. The revisions to earlier months, since May and June were revised down by a combined 103,000 in the July report. The labour force participation rate, which was 61.4 percent in July and down 0.7 points since January, because a steady unemployment rate can hide people leaving the workforce. And the long term unemployed share, which was 25.5 percent of all unemployed people.

Q: Does the ADP report predict the BLS jobs report?

A: No. ADP measures its own payroll records for private employers, while the BLS establishment survey includes government employment and uses a different sample and methodology, and the unemployment rate comes from a separate household survey. ADP reported 38,000 private sector jobs added in August, the slowest since January, but the two measures regularly diverge and neither forecasts the other.

Stay ahead of the market rather than reacting to it. Metaintro tracks where hiring is actually happening and matches you with roles that fit the experience you already have. Create your free profile and let the right openings reach you first.