The Bureau of Labor Statistics published the employment report on Friday, 1 August 2025.
My quick summary of the report (see below some numbers from summary table A on page 4):
The monthly increase in the number of jobs is far less than the monthly increase in the number of people in the country. The unemployment rate is approximately unchanged, because non-American workers are leaving the labor force (and may or may not be leaving the country).
The growth almost all comes from health care jobs. The federal government is losing jobs.
Of the newly unemployed people, a lot of them are people who just entered the labor force this month — perhaps because they graduated from either high school or college last month.
The big news was not even in the report; the big news is that the president’s reaction to the report was to immediately fire the person in charge of releasing the report. All economists I have talked to think that having the Bureau of Labor Statistics become a political organization is a bad idea. The Fed depends on objective data for its decision making. Many investors use this data in their decision-making process.
| June 2025 | July 2025 | |
| Civilian non-institutional Population | 273,585,000 | 273,785,000 |
| Employed People | 163,366,000 | 163,106,000 |
| Unemployed People | 7,015,000 | 7,236,000 |
| Unemployment Rate | 4.1% | 4.2% |
| Unemployed / lost job | 3,293,000 | 3,405,000 |
| Unemployed / Quit | 825,000 | 784,000 |
| Unemployed / Reentered | 2,145,000 | 2,180,000 |
| Unemployed / Just entered labor force | 710,000 | 985,000 |
The actual report is currently here: https://www.bls.gov/news.release/empsit.nr0.htm. The report has been coming out every month, on the first Friday of the following month (or a different day if that is a holiday), but I have no idea if that will continue under the new management.
There are two main parts of the report:
The household survey is conducted by going to tens of thousands of people every month and asking them if they were employed during the calendar week that included the twelfth of the month (in this case the week from July 6 to July 12), and if not, were they actively looking for work that they could have done at some point during that week, or during the three weeks before that (in that case, meaning were they looking for work sometime during the period from June 15th to July 12th).
They call the people who answered yes to the first question “employed,” and the people who answered yes to the second question “unemployed.” The sum of those two is called the “labor force.” They make adjustments for seasonality and extrapolate their survey results to the entire country. They also ask other questions to get demographic information, so they can tell you about black women and latino men, etc.
The headline unemployment rate is their estimate of the number of unemployed people divided by their estimate of the labor force.
The Establishment Survey is a monthly survey of employers, asking how many people they were paying for the pay period including July 12th, so that could be for the whole month of July, or for the week, etc. They also apply seasonal adjustments, and extrapolate their findings to the entire country, and publish a report every month with breakdowns by industry.
Each of the surveys uses a sample of thousands of people, which the Bureau of Labor Statistics (BLS) extrapolates to the whole country. The BLS has employed thousands of people to do this, although many have been laid off in the last few months, so they have some challenges. Extrapolating their survey results to the whole country depends on knowing how many employers there are in the country. They update that annually by getting ALL of the records of the unemployment insurance premiums paid by all employers in the whole country. They also have to follow up to get missing data, and revise their reports for a couple of months as data comes in that misses the original deadline. Each month, they typically issue revisions to the previous couple of months.
Revisions for May and June were larger than normal. The change in total nonfarm payroll employment for May was revised down by 125,000, from +144,000 to +19,000, and the change for June was revised down by 133,000, from +147,000 to +14,000. With these revisions, employment in May and June combined is 258,000 lower than previously reported. (Monthly revisions result from additional reports received from businesses and government agencies since the last published estimates and from the recalculation of seasonal factors.)
Seasonal Adjustments are also important. The idea of “seasonally adjusted numbers” is that they want to report changes to employment, NOT COUNTING things that happen because of calendar events. For example, schools employ a lot fewer people in the summer. The BLS wants to report that there are more or fewer jobs because employers are employing more or fewer people NOT because hundreds of thousands of teachers are unemployed due to summer vacation. To continue the example of public school teachers, their estimate of the number of people employed in local government education is (from the end of table B1, which is on page 31 of the report):
May: 8,490,500
June: 8,160,400
July: 7,081,600
Showing a decline of 1,408,900. But the seasonally adjusted number are:
May: 8,204,400
June: 8,204,000
July: 8,193,600
That is just saying that over 1.4 million fewer people are actually employed, but that that only counts as about 11 thousand jobs lost, because it was pretty close to what the BLS thinks is normal for this time of year.