Will the September non-farm payrolls see another "significant downward revision" and open the door to a "50 basis point rate cut"?

　　Wallstreetcn

　　Goldman Sachs and Standard Chartered have warned that the monthly non-farm employment data may have been overstated by 40,000 to 70,000 jobs. On September 9, the U.S. Department of Labor will release the annual benchmark revision of non-farm employment data. This revision may result in a one-time downward adjustment of 550,000 to 800,000 inflated jobs, potentially prompting the Federal Reserve to follow last September’s example and opt for another significant 50 basis point rate cut.

　　U.S. employment data may once again face a “face-slapping” significant downward revision, paving the way for a 50 basis point rate cut in September.

　　On September 9, the U.S. Department of Labor (BLS) will release the annual benchmark revision of non-farm employment data. According to data projections by Goldman Sachs, Standard Chartered, and others, this could be a massive “employment fraud” of 550,000–800,000 jobs, which will directly impact market confidence in the U.S. labor market and may force the Federal Reserve to conduct a 50 basis point rate cut, just like last September.

　　There are two main reasons for the significant downward revision. First, the birth-death model is distorted, as it overestimates the jobs created by new businesses. Second, the sharp reduction in illegal immigration has led to a systematic overestimation of the labor force population. It is estimated that these biases may have caused actual employment to be overestimated by 40,000 to 70,000 jobs per month, amounting to an annual cumulative inflation of 550,000 to 800,000 jobs.

　　The implications behind this are very significant. A senior trader at Goldman Sachs stated that the key factor determining Powell’s pace now is not inflation, but employment. If this revision is similar to last September (when BLS also revised down 800,000 jobs and the Federal Reserve immediately cut rates by 50 basis points), Powell may once again face the choice of “whether to cut rates by 50 basis points in one go,” even if only to “prove innocence”—that last year’s rate cut was not a political compromise, but based on real economic slowdown.

　　Standard Chartered estimates that the NFP published by BLS is 70,000 jobs higher than the actual number each month

　　Goldman Sachs pointed out that the biggest source of distortion in employment data is the BLS’s long-term use of the “birth-death model.” This model is used to estimate the number of jobs created by new businesses, but it is not based on actual business registration or tax data, rather on model estimates, which tend to systematically overestimate employment growth. In contrast, QCEW (Quarterly Census of Employment and Wages) and BDM (Business Employment Dynamics) are based on records of companies actually paying unemployment insurance and are considered the more reliable “gold standard.”

　　Goldman Sachs, using its own model combined with BED data and higher-frequency business dynamics information, found that in the second half of 2024, the BLS model indeed overestimated employment growth, with an average monthly overestimation of 45,000 jobs. Although BLS has slightly adjusted model parameters in recent months and reflected the stabilization in the number of new businesses, the deviation remains significant.

　　Steven Englander of Standard Chartered more directly called the birth-death model a “fig leaf for the data.” He estimates that the NFP published by BLS is 70,000 jobs higher than the actual number each month.

　　According to his analysis, since the beginning of 2024, old companies have only added 25,000 jobs per month, while BLS estimates that “new companies” contribute more than 100,000 jobs per month. But BDM data shows that new companies actually account for only 20% of all new jobs, far below the BLS assumption. More seriously, the number of jobs created by new businesses in 2024 is less than 20% of that in 2022. If the model reflects this reality, NFP would be at least 70,000 jobs lower per month.

　　Englander further pointed out that to maintain basic labor market equilibrium, the “reasonable level” of non-farm employment data should be 170,000 jobs per month—100,000 from real natural growth, and 70,000 from the model’s overestimation.

　　It is worth noting that although BDM is lagged (latest only up to 2024), like QCEW, it is the data basis used by the U.S. Department of Labor for the mid-year benchmark revision, and its authority far exceeds the sample-based non-farm employment data. The employment benchmark revision to be released by BLS on September 9 is based on these data. Once revised according to the real trends reflected by BDM, non-farm employment may be adjusted downward by 550,000 to 800,000 in one go, which will have a huge impact on market confidence and policy outlook.

　　Five major signals: Signs of inflated employment data have long existed

　　Goldman Sachs pointed out that in addition to the virtual increase caused by the birth-death model, there are at least five additional reasons further indicating serious problems with the data.

　　1. Decrease in illegal immigration

　　Goldman Sachs estimates that the number of illegal immigrants has dropped significantly in recent months. Illegal immigration has a significant impact on labor supply. The “immigration wave” from 2022 to 2024 brought a surge in employment demand, but now that immigration has slowed, the actual need for new jobs has also decreased. If BLS continues to estimate employment demand based on old immigration assumptions, it will obviously be too high.

　　2. Seasonal adjustment models may misjudge trends

　　Seasonal adjustment models often initially mistake real trend changes for seasonal fluctuations. Only when it is later confirmed that the trend has indeed worsened will the model revise previous data downward.

　　3. Historically, raw data is always revised downward during economic slowdowns

　　Historical experience shows that during periods of economic slowdown, the initial raw employment data is often revised downward later. This phenomenon has occurred in every recession since 1979 (except once).

　　4. ADP data questions BLS’s exaggeration of the healthcare industry

　　As a major U.S. payroll data provider, ADP’s data shows that employment growth in the healthcare industry is far less robust than reported by BLS. In the past three months, new jobs in the healthcare industry accounted for more than all non-farm employment growth. Both ADP and industry analysts believe it is not as exaggerated as BLS claims, and the real situation may be somewhere in between.

　　5. Household survey overestimates immigration and employment

　　The household survey may currently overestimate U.S. population growth and employment growth. The immigration estimates used at the beginning of the year were reasonable at the time, but are now far too high. The current model assumes that the U.S. population grows by about 1 million per year, which may be an overestimate. This could cause the employment growth data in the “household survey” to be overestimated by about 50,000 jobs per month.