U.S. employment data "inflated" for the second consecutive year? Major revisions may trigger interest rate cuts and political storms

According to Zhitong Finance APP, the employment growth data currently released by the U.S. government may significantly overestimate actual performance. The latest analysis shows that the momentum of U.S. job growth over the 12 months ending in March of this year is far weaker than the strong figures indicated by official data, suggesting that signs of a slowdown in the labor market had already emerged before the hiring weakness this summer.

Economists from Wells Fargo, Comerica Bank, and Pantheon Macroeconomics predict that the preliminary benchmark revision data to be released by the U.S. Bureau of Labor Statistics (BLS) on Tuesday will show that the non-farm payrolls as of March are nearly 800,000 less than the current estimate, translating to an average monthly shortfall of about 67,000 jobs. Nomura Securities, Bank of America, and Royal Bank of Canada believe that the downward revision could even approach 1 million.

Although this revision involves historical data, a significant downward adjustment will confirm that last year's labor market momentum was far weaker than expected, further strengthening market expectations for the Federal Reserve to initiate a series of interest rate cuts. This will also mark the second consecutive year of substantial downward revisions to employment data, potentially triggering a new round of questioning from Trump regarding the accuracy of the statistics—he has previously questioned the accuracy of BLS data.

As part of the annual standard procedure, the U.S. Bureau of Labor Statistics will conduct a benchmark revision of employment levels as of March, referencing more accurate but lagging "Quarterly Census of Employment and Wages" (QCEW) data. This census is based on state unemployment insurance tax records and covers nearly the entire U.S. workforce. This revision complements the regular adjustments made to the monthly employment report, enhancing the accuracy of the data.

Bill Adams, Chief Economist at Comerica Bank, stated: "Compared to the recent downward revision of employment growth data, the significant downward revision of employment data before March 2025 has a smaller impact on monetary policy, but it provides a more comprehensive context for assessing the overall economic situation. Moreover, all else being equal, a downward revision of employment growth data will increase the pressure on the Federal Reserve to ease policy."

This also supports the view that the Federal Reserve should have started cutting rates months ago. Federal Reserve Governor Waller indicated that the benchmark revision is expected to lead to a downward adjustment of about 60,000 in the average monthly employment growth data. He supported a rate cut at the July monetary policy meeting, but the final decision was to maintain interest rates. The market widely expects the Federal Reserve to announce a rate cut at next week's meeting.

Political Impact

Although this revision will not change the understanding of the current state of the labor market, it means that the recent trend of hiring slowdown actually began much earlier. The preliminary estimate report covering data up to March 2025 may be used by the Trump administration as evidence to demonstrate that job growth had significantly weakened before he took office. The final revised data will be released early next year.

Samuel Tombs, Chief U.S. Economist at Pantheon Macroeconomics, pointed out: "These data primarily reflect the job creation situation before Trump's administration. Therefore, he can fully claim that this indicates the economic conditions he inherited were actually weaker than everyone expected." Just a month ago, the monthly employment data underwent a rare and significant downward revision, triggering outrage from the White House, leading Trump to fire the head of the Bureau of Labor Statistics. At that time, he not only criticized these monthly revised data but also questioned last year's preliminary benchmark revision — which indicated that the downward revision of non-farm employment data would reach the highest level since 2009.

Despite Trump's ongoing criticism of data revisions, both monthly and benchmark revisions are a routine process that updates estimates as more data becomes available. The increase in the magnitude of revisions in recent years is partly due to a continuous decline in survey response rates.

Birth-Death Model

In recent years, the employment growth indicated by the monthly non-farm employment data has generally been higher than that of the QCEW data. Some economists believe that the difference between the two partly stems from the so-called "birth-death model" — an adjustment made by the Bureau of Labor Statistics to account for the net employment changes brought about by new and closed businesses. Since the pandemic, the difficulty of this calculation has significantly increased.

There is also a viewpoint that another reason behind the difference is immigration. The monthly non-farm employment report does not survey respondents' citizenship status, while the QCEW data is based on unemployment insurance records, which illegal immigrants cannot apply for, and thus are not included in the statistics.

Ultimately, while waiting for the complete data for 2025 (to be released next February along with the government's final revision report), economists and policymakers will rely on preliminary benchmark data to assess the pace of the labor market slowdown.

Carrie Freestone, an economist at the Royal Bank of Canada, stated: "If there are significant revisions to last year's data, it will help us re-establish the benchmark starting point. But I think what Federal Reserve officials are most concerned about is that the economy is losing momentum — we are likely at a turning point in the labor market."