In recent years, the ADP report is less and less likely to be a strong predictor of the offical BLS numbers.
Wall St (and most people including myself) still quote the BLS data as the “offical” numbers.
I am not sure we should do that, (the split came when the quaility of the ADP report improved and the quality of the BLS data slipped.)
Anyway, this month’s ADP report is strongly positive.
That may or may not reflect the official numbers (which come out Friday.)
What do you mean by the quality of the report? What changed to make the ADP report less reliable and BLS data more accurate?
The BLS uses surveys and the surveys are getting declining response rates.
ADP uses actual data from nearly 20% of the US private sector and then uses a multiplier. In 2022, it hired a Stanford Uiv. lab and changed its multiplier based on their more rigorous methods.
.
.
.
.
Per CNN (Dec 2023):
One of the biggest issues the BLS suffers from is falling response rates to their surveys, which reduces the accuracy of their reports, said Erica Groshen, a former commissioner of the Bureau of Labor Statistics who is now at Cornell University.
Per Grok:
ADP uses actual payroll data from over 25 million employees across its client base, adjusted for seasonality and weighted to reflect national trends. . . . (covering about 20% of private-sector employees) . . .
and
In 2022, ADP retooled its National Employment Report in collaboration with the Stanford Digital Economy Lab, moving away from forecasting the BLS report to creating an independent, real-time measure of private-sector employment based on actual payroll data. This change explicitly reduced its focus on predicting BLS numbers, leading to larger discrepancies.
As always thank you.
You are welcome.
I still use the BLS numbers.
so does practically everyone else.
Not sure we should, but . . .. we do.
I believe that you got this backwards?
BLS takes a voluntary survey —> Then uses a multiplier.
ADP uses actual data —> Then uses a multiplier.
In the past both used the same (or nearly the same) multiplier.
A growing bodyof evidence suggested that multiplier was outdated
so in 2022 ADP swiched to the newer better multplier (they made the switch by contrcting with Stanford Univ.)
.
.
.
From time to time things like mutlipliers need to be changed, but the changes don’t come very often because:
