Quarterly employment and wages census data are released today. Recall that some observers have claimed a recession in the first half of 2022 (and/or in the second quarter of 2022), as the household survey and QCEW employment numbers have already flattened. Based on the revised first and second quarter data and the new third quarter data, we have the following picture of nonfarm payrolls and private nonfarm payrolls levels.
figure 1: Non-farm payrolls CES release from January 2023, including base revision (blue), ADP (tan), total workers covered by QCEW, seasonalized using log-transformed Census X-13 (pink) Adjusted, using multiplicative moving average (sky blue), Philadelphia Fed preliminary data benchmark (red squares), all 000, sa light blue shading indicates a hypothetical recession in the first half of 2022. Sources: BLS (various) and ADP via FRED, BLS QCEW, Philadelphia Fed via FRED, and authors’ calculations.
figure 2: Private nonfarm payrolls CES release from January 2023 with baseline revisions (blue), ADP (tan), QCEW private covered workers, seasonalized using log-transformed Census X-13 (pink) Adjusted, using multiplicative moving average (sky blue), Philadelphia Fed preliminary benchmark minus reported government employment (red squares), all 000, sa light blue shading indicates hypothetical recession in 1H 2022. Sources: BLS (various) and ADP via FRED, BLS QCEW, Philadelphia Fed via FRED, and authors’ calculations.
Now, because of the different coverage, it is difficult to see whether employment actually increased throughout the first half of 2022. So I’m showing the cumulative change (in 000s) since December 2021.
image 3: Non-farm payrolls since December 2021 Change since January 2023 CES release includes baseline revisions (blue), ADP (tan), total workers covered by QCEW, using log-transformed Census X- 13 (pink) seasonally adjusted using multiplicative moving average (sky blue), Philadelphia Fed preliminary benchmark (red squares), both 000’s, sa light gray shading indicates a hypothetical recession in the first half of 2022. Sources: BLS (various) and ADP via FRED, BLS QCEW, Philadelphia Fed via FRED, and authors’ calculations.
Figure 4: Change in private nonfarm payrolls since December 2021 CES release since January 2023 Contains baseline revisions (blue), ADP (tan), QCEW private covered workers, using log-transformed Census X-13 (pink) seasonally adjusted using multiplicative moving average (sky blue), Philadelphia Fed preliminary benchmark minus reported government employment (red squares), all 000, sa light gray shading represents hypothetical upper 2022 half-year recession. Sources: BLS (various) and ADP via FRED, BLS QCEW, Philadelphia Fed via FRED, and authors’ calculations.
Keen observers will notice Every The sequence in Figure 3 is higher at the end of 2022H1 than at the end of 2021H2. each series, Save the Philly Fed’s early benchmarksin 2022M06 is higher than in 2022M03.
In Figure 4 (private non-farm payrolls), each series at the end of 1H22 is again higher than at the end of 2H21. Both series – the Philadelphia Fed’s earlier benchmark minus measured government employment and the seasonally adjusted QCEW using a moving average – show a slight decline from March to June. But all other series (including the ADP series which relies on actual payroll processing data and thus independent of CES or CPS) are up.
For the question of future trends, every series in 2022M09 is higher than 2022M03.
So, when evaluating this January’s comments :
The first has to do with the credibility of the source. Which should we believe? CES or HH survey? Menzie argues for CES. That seems to be a bit of a problem in the first half, because we’re seeing lower productivity and lower GDP. If we added so many full-time jobs, why did productivity collapse and GDP drop?
In contrast, the HH (CPS) survey showed 1) employment was flat after March and 2) more than 100% of job growth came from part-time work or multiple jobs, suggesting a surge in low-wage jobs that negatively impacts productivity. This seems more reasonable to me.
At the same time, I think it’s possible that both surveys are actually correct but give a misleading impression because we’ve misinterpreted the data. That still seems possible, although I’ve read others that CES was rigged to provide a more rosy picture going into the election. Regardless, multiple staff members may be able to reconcile the discrepancy if one feels that the two surveys may be true in some sense. But it turns out that multiple jobs accounted for only 314,000 of the 2.7 million jobs at CES from March to November. Period, I think we are debating period. So this assumption turned out to be incorrect, as Menzie pointed out, and I admit.
Then we learned that CES was fundamentally incorrect and the Fed cut March-June job growth from 1.1 million to 10,500. During that time, it made the whole issue of the settlement moot by asserting that CES had been producing phantom jobs. No settlement required.
Thus, the HH survey seems to be the more reliable source and will certainly continue into June and probably most of the rest of the year. This suggests that the growth in employment came entirely from part-time and multiple jobs.
It can be argued that almost all statements are false.exist part time job. exist CPS and CES. On CES operations, there are No Evidence (unless you find it on an Italian satellite that changes vote counts). It’s wrong that low wages lead to a clear drop in productivity, I can’t even say anything about it.Regarding the comment: “Then we learned that CES was fundamentally incorrect and the Fed cut March-June job growth from 1.1 million to 10,500.” , I would observe that it is Philadelphia Fed researchers Reduced the number of estimated job gains, not the Fed as an institution.
