Devoted readers of Eakinomics and, especially, the monthly Gordon’s Guesstimate leading up to the national employment report from the Bureau of Labor Statistics (BLS) are aware that a key piece of data preceding the BLS report is the ADP National Employment Report. As the press release says: “The jobs report and pay insights use ADP’s fine-grained anonymized and aggregated payroll data of over 25 million U.S. employees to provide a representative picture of the labor market. The report details the current month’s total private employment change, and weekly job data from the previous month.”

That’s terrific and Gordon can talk about it tomorrow if he wants. The news, at least to Eakinomics, is that as part of a reformulated report launched in August, ADP also has a measure of wage growth. Specifically: “ADP’s pay measure uniquely captures the earnings of a cohort of almost 10 million employees over a 12-month period.”

Now that is interesting. Each month, the jobs report provides information on average hourly earnings, but that average is a mixture of earnings across industries, occupations, and attributes of workers. The ADP measure looks at the same workers over a 12-month period, which would seem to be a much cleaner estimate of the wage increase.

Does it matter?

The graph (below) shows the year-over-year percent change in wages every month since January 2021. The blue line represents average hourly earnings for production and non-supervisory workers (“working stiffs”) from the BLS, while the gray line represents the ADP annual pay measure for workers who stay in the same job. There are two things of interest to note.

First, the two measures diverge, and markedly so during 2022. This year, the ADP measure would suggest much more rapid wage growth than the BLS indicator. (Aside: In recent months, the wage growth for workers who change jobs in the ADP survey is roughly double that for those who stay in their job.)

Second, both measures are growing slower than inflation, as measured by the Consumer Price Index (CPI) (orange line). But the recent history on real wage declines is a lot less bleak in the ADP data than in the BLS.

There is no reason to believe that either is “right” or “wrong.” They are simply two different estimates of the movement in aggregate real wages. But to the extent that the reality is closer to the ADP, the resiliency of household spending in the face of the Fed’s tightening is a lot more understandable. Similarly, the more rapid underlying cost-push of wage growth would contribute to the stickiness of inflation the Fed has encountered.