Employment up (Non-Farm Payrolls), Unemployment up (Household Survey)

Chief Economist Eugenio J. Alemán discusses current economic conditions.

Sometimes, and February was one of those times, the information on the U.S. labor market doesn’t make sense. Many argue that government agencies are cooking the books. However, the counter argument is: Why wouldn’t they cook the books ‘consistently,’ that is, if they cook the non-farm payrolls survey up to show strength in the labor market they should also cook the books for the household survey. The problem is that to keep the ‘consistency’ over time, especially if you have a particular agenda, it means that statisticians would have to ‘change’ the ‘recipe’ every month (i.e., that is, they will have to change their lie every month), which would mean that they will be easily caught. Thus, there is no conspiracy behind these numbers. These are just two different surveys with different characteristics.

In February, the non-farm payroll survey showed 275,000 new jobs added while the household survey showed a loss of 184,000 jobs. How could that be? Easy, as we said above, these are two very different surveys, and they have very different characteristics. Thus, they give, sometimes, very different views of the labor market. While in the non-farm payroll survey an individual can have more than one payroll job during a specific month, in the household survey, that is not possible. In the household survey you are asked if you worked over the last several weeks or months. If you say no but you have been looking for a job, then you are unemployed. If you say no and you have not been looking for a job, you are out of the labor force and not considered to be unemployed.

In the payroll survey, as long as you are in the payroll of one or more than one firm, then you are employed, and you are counted.

What the February non-farm payroll report showed with the downward revisions to December 2023 and January 2024, down a net of 167,000 jobs, is that the surveys are probably still having issues with seasonal factors, which made them adjust the numbers considerably. Thus, we should also expect some revisions to the February non-farm payroll number.

Household Survey VS Establishment Survey