Why the jobs report comes from two employment surveys

Key takeaways

• BLS says the jobs report uses both the Current Employment Statistics survey and the Current Population Survey because both are needed for a complete picture of the labor market.
• BLS says the establishment survey has a smaller margin of error for month-to-month employment change, while the household survey has broader scope because it includes self-employed, agricultural, unpaid family, and private household workers.
• The current BLS FAQ says an over-the-month change of about 122,000 is statistically significant in the establishment survey, versus about 650,000 in the household survey.

The payroll number is not the entire labor market

BLS says the Current Employment Statistics survey, also called the payroll or establishment survey, and the Current Population Survey, also called the household survey, are both needed for a complete picture of the labor market. That is the core fact many fast reactions skip.

The establishment survey is built to measure employment, hours, and earnings in the nonfarm sector with industry and geographic detail. The household survey is built to measure the labor-force status of the civilian noninstitutional population and is where the unemployment rate comes from.

Why the two measures can diverge

BLS says the establishment survey has a smaller margin of error on month-to-month change because its sample is much larger. In the current FAQ, an over-the-month change of about 122,000 is statistically significant in the establishment survey, versus about 650,000 in the household survey.

At the same time, BLS says the household survey has a more expansive scope because it includes self-employed workers whose businesses are unincorporated, unpaid family workers, agricultural workers, and private household workers, all of which are excluded by the establishment survey.

• Payrolls are jobs in the covered nonfarm establishment sample.
• Household employment is people, not payroll positions.
• The unemployment rate comes from the household survey, not the payroll survey.

How to read the release more carefully

A more careful jobs-report read starts by separating the payroll change, the unemployment rate, labor-force participation, and any revisions. BLS notes that the establishment survey revises initial monthly estimates in the following two months and then re-anchors them through an annual benchmark revision.

That is why one monthly release should be read as an update inside a survey system, not as a single clean verdict on the economy.

Source evidence snapshot