Significant downward revision? Nomura: The annual benchmark revision for non-farm payrolls will be released, with expectations from both the market and the Federal Reserve

The U.S. Bureau of Labor Statistics is about to release the annual benchmark revision data for non-farm employment. Nomura Securities stated that both the market and the Federal Reserve have anticipated a significant downward revision.

On September 3rd, according to news from the Chasing Wind Trading Desk, Nomura Securities in its latest research report stated that it expects the non-farm payroll annual benchmark revision to be released by the U.S. Bureau of Labor Statistics (BLS) on September 9th to be significantly lowered, with a revision expected to reach between 600,000 to 900,000 jobs.

This revision will cover the 12-month period from April 2024 to March 2025, meaning that monthly job growth will be revised down by 50,000 to 75,000 jobs. Nomura analysts derived this prediction based on data from the Quarterly Census of Employment and Wages (QCEW) and other sources.

The research report stated that the estimation bias of undocumented immigrants and new businesses has become the main reason for the data revision. In addition, the Bureau of Labor Statistics will also release industry breakdown data for the annual benchmark revision. Nomura expects that job growth will further concentrate in a few industries.

Nomura believes that although this revision may have a dovish impact on the market, it is a dovish signal within market expectations, and since Federal Reserve officials have already anticipated this adjustment, the actual impact on monetary policy may be limited.

Nomura maintains its expectation that the Federal Reserve will begin to cut interest rates by 25 basis points quarterly starting in September, and will not implement a significant cut of 50 basis points or consecutive meeting cuts unless there are large-scale layoffs or severe financial pressures.

Multiple Data Support Significant Downward Revision Expectations

Nomura's analyst team found based on QCEW data that data from the second to fourth quarters of 2024 indicate that the non-farm employment report has overestimated job growth by 857,000 jobs (about 95,000 per month). If simply extrapolated, this would reduce the average monthly non-farm employment growth for the 12 months ending in March 2025 from the current official figure of 146,000 to about 50,000.

However, the QCEW data for the second to fourth quarters is usually revised upward, which means the gap between non-farm employment and QCEW may narrow. Conversely, the QCEW data for the first quarter often tends to be weaker than non-farm employment data, which may partially offset the positive impact of the upward revisions in the second to fourth quarters.

It is worth noting that QCEW data covers over 95% of employment positions in the U.S., and although it is released with a lag of 6 to 9 months, it is more comprehensive and accurate than monthly non-farm data.

In addition to QCEW data, Nomura also analyzed other related data (such as unemployment claims and tax data), which also suggest that non-farm employment will be significantly revised down.

Furthermore, as of March 2025, non-farm employment appears to be high relative to household employment and other employment indicators like ADP, further confirming the necessity of the revision

Multiple Factors Lead to Employment Data Discrepancies

Nomura analysts believe that the significant gap between QCEW and non-farm data may stem from two main factors:

First, QCEW data fails to capture the employment situation of undocumented immigrants, while non-farm data may reflect some employment status of this group.

Second, the "birth-death" adjustment embedded in the monthly non-farm data by the Bureau of Labor Statistics may overestimate actual employment growth.

This adjustment aims to reflect the net employment changes brought about by business openings and closures, but forecast errors often account for a large proportion of annual revisions.

Historical data shows that the forecast errors of the "birth-death" adjustment are typically an important component of annual non-farm data revisions, and this significant downward revision is likely related to this.

Rising Concentration of Employment Growth Raises Concerns

The Bureau of Labor Statistics' annual benchmark revision will also reveal a more concentrated distribution of employment growth across industries. Nomura expects employment growth to become more concentrated in a few sectors, as only a handful of industries have driven employment growth in recent months. The research report states:

Based on QCEW data from the second to fourth quarters of 2024, it is expected that professional services, construction, information technology, and manufacturing will see significant downward revisions in employment. Among these, the information technology sector is likely to experience the largest downward revision, reflecting the impact of AI.

Even before the preliminary benchmark revision, Nomura's calculations have shown that the proportion of industries reducing employment monthly has risen from 25% in March 2024 to about 45% in July 2025. The further deterioration in the concentration of employment growth should increase the downside risks to future employment growth.

Maintaining the Federal Reserve's Expectation of 25 Basis Points Rate Cuts Each Quarter

Nomura states that although the significant downward adjustment constitutes a dovish signal for the market, the actual impact on monetary policy is expected to be limited for the following reasons:

Federal Reserve officials are already anticipating this: Powell mentioned in his Jackson Hole speech that based on QCEW data, non-farm data "will be significantly revised downward." Governor Waller has also stated that monthly job creation will average about 60,000 fewer, corresponding to an annual downward adjustment of 720,000

Unemployment Rate Unaffected: This revision does not involve adjustments to the unemployment rate, which is a key indicator repeatedly emphasized by Powell and New York Federal Reserve President Williams.

Many members of the Federal Open Market Committee believe that the unemployment rate reflects labor market conditions better than non-farm data due to a reduction in labor supply caused by immigration policies.

Nomura continues to expect that the Federal Reserve will begin to cut interest rates by 25 basis points on a quarterly basis starting in September, rather than a significant cut of 50 basis points or consecutive meeting cuts, unless there is a sharp increase in layoffs or severe financial stress