2024 ADP Jobs Created Overstated by Near 550K?
Updated
Recently, the September ADP Employment Report was published. (You can download historical data from the link above.)
After the report was released,
DXY, 
US02Y, US10Y, and US30Y rose, suggesting that the market perceived the report as strong. However, the details of the report tell me the opposite.
Note, the data being published is seasonally adjusted (SA). However, it is possible to obtain the raw, non-seasonally adjusted (non-SA) data from the website above. I calculated the number of jobs created from the beginning of the year until September (inclusive) for both non-SA and SA data and determined the differences between these two values. You can find my spreadsheet here: icloud.com/numbers/0acGeMCM_n9bybTq_mLBk3jjA A screenshot of the results is also shown in the chart.
As you can see, in typical years, the difference between jobs created from the start of the year through September for non-SA and SA is around 1.1M. Non-SA figures are usually higher because the last quarter tends to be weak for job creation. However, 2024 is quite different. The 2024 SA total jobs created is larger than expected by about 550K jobs. If we adjust by removing 550K reported SA jobs from 2024, the difference between non-SA and SA jobs would become approximately 1.1M, which is typical for a regular year.
Why is this significant? Many indicators suggest that the U.S. economy is nearing a recession. Thus, this unusual job creation pattern is very suspicious. The published SA ADP employment numbers may be masking underlying economic weakness.
Even with rate cut(s), I expect that the last quarter of 2024 will be weaker for job creation compared to a typical year. Therefore, I anticipate significant revisions to ADP employment data around December or January.
After the report was released,
DXY, US02Y, US10Y, and US30Y rose, suggesting that the market perceived the report as strong. However, the details of the report tell me the opposite.Note, the data being published is seasonally adjusted (SA). However, it is possible to obtain the raw, non-seasonally adjusted (non-SA) data from the website above. I calculated the number of jobs created from the beginning of the year until September (inclusive) for both non-SA and SA data and determined the differences between these two values. You can find my spreadsheet here: icloud.com/numbers/0acGeMCM_n9bybTq_mLBk3jjA A screenshot of the results is also shown in the chart.
As you can see, in typical years, the difference between jobs created from the start of the year through September for non-SA and SA is around 1.1M. Non-SA figures are usually higher because the last quarter tends to be weak for job creation. However, 2024 is quite different. The 2024 SA total jobs created is larger than expected by about 550K jobs. If we adjust by removing 550K reported SA jobs from 2024, the difference between non-SA and SA jobs would become approximately 1.1M, which is typical for a regular year.
Why is this significant? Many indicators suggest that the U.S. economy is nearing a recession. Thus, this unusual job creation pattern is very suspicious. The published SA ADP employment numbers may be masking underlying economic weakness.
Even with rate cut(s), I expect that the last quarter of 2024 will be weaker for job creation compared to a typical year. Therefore, I anticipate significant revisions to ADP employment data around December or January.
Note
Today, October ADP report was published. I have downloaded the latest data and updated my analysis: tradingview.com/x/bHQFyHhh/.
This is the spreadsheet: icloud.com/numbers/0c1mBDH9WukdPb-nM7qq8KWRA
Average difference for non-SA and SA jobs created since the beginning of the year to October is 1.252M. This year, it is 376K. Even if we compare with 2023 data, this year SA data is overstated by nearly 1.04M - 376K = 664K. (We need to reduce SA jobs created this year by 664K, so the difference between non-SA and SA data aligns with previous year.) If we compare with the average from all previous years, we see an even larger discrepancy.
Therefore, the situation since September has only worsened, despite the published ADP number being significantly higher than expectations.
Trade active
Today, November ADP report was published. I have downloaded the latest data and updated my analysis: tradingview.com/x/ye6VFUgl/.
This is the spreadsheet: icloud.com/numbers/038wyKouRekei6vcP7n7lHSPQ
Average difference for non-SA and SA jobs created since the beginning of the year to November is 1.192M. This year, it is 445K. Even if we compare with 2023 data, this year SA data is overstated by nearly 1.02M - 445K = 574K. (We need to reduce SA jobs created this year by 574K, so the difference between non-SA and SA data aligns with previous year.) If we compare with the average from all previous years, we see an even larger discrepancy.
Therefore, the situation since October has slightly improved, but still 574K jobs are missing.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.