On Wednesday, we will gain initial insights into the condition of the US job market as 2026 begins, along with a more defined understanding of hiring trends in 2025. The Bureau of Labor Statistics will release the January jobs report at 8:30 am on Wednesday. The important employment report is experiencing a slight delay due to the recent government shutdown and will indicate whether there has been an improvement in the US labor market, which has been in a state of low hiring and low firing. In the previous year, the economy recorded its lowest level of job growth outside of a recession since 2003. The year concluded with the economy adding 50,000 jobs in December, which closely aligns with the average monthly gain for the year, and unemployment decreased to 4.4%, as reported. “Many workers feel stuck in their careers or feel frozen out of the job market,” stated Daniel Zhao. The crucial churn necessary for a robust labor market has significantly diminished, resulting in a greater number of individuals seeking employment than there are positions available. The January jobs report will feature a series of essential revisions (specifically, the annual benchmark revision) and statistical modeling adjustments that will not only offer a more comprehensive perspective on past employment trends but also have the potential to influence our current and future understanding of the labor market.
The latest labor market data, encompassing both public and private sources, suggests a strong possibility that job growth has been modest, unemployment levels have stayed low, and health care continues to be a key factor in overall hiring trends. There’s a possibility that seasonal and weather-related factors could result in a stronger-than-expected reading for January: “Weaker holiday hiring has meant fewer post-holiday layoffs, and unseasonably warm weather during the early part of last month could have bolstered employment in industries like construction.” According to a source, economists’ consensus estimates indicate job gains of 80,000 last month, with the jobless rate expected to remain at 4.4%. On the supply side, Baby Boomers are aging and retiring, population growth has slowed, and there has been a significant reduction in immigration alongside an increase in deportations. On the demand side: Large employers are reducing their workforce after over-hiring during the pandemic; significant uncertainty – especially regarding the Trump administration’s abrupt and sweeping domestic policy changes – has obscured businesses’ decision-making and hindered hiring; companies have redirected some investments from hiring to equipment and technology (including artificial intelligence) to enhance productivity; and a high-cost environment, coupled with steep tariffs, federal funding cutbacks, and stringent immigration enforcement, has adversely impacted certain businesses.
Joe Brusuelas emphasized several factors while countering White House economic adviser Kevin Hassett’s assertion on Monday that subdued job gains are mainly due to declining population figures and increased productivity. “The notion that slower hiring is merely a result of long-term demographics is both inadequate and serves as a diversion from the issues surrounding immigration and trade policies – consider the 72,000 drop in manufacturing jobs last year, which is expected to appear even more severe after the forthcoming benchmark revision,” Brusuelas stated. Federal data is dynamic and often revised as more precise and comprehensive information emerges. The BLS’ monthly jobs report aims to deliver a more frequent perspective on employment trends; however, this immediacy may compromise accuracy. The monthly employment snapshot is derived from a survey conducted by the BLS, which engages approximately 121,000 US employers across 631,000 work sites, representing over a quarter of total employment. Respondents are provided with three chances to report their payroll gains and losses for any specific month. Each year, the BLS engages in a meticulous process designed to deliver a nearly comprehensive employment count by reconciling the monthly survey estimates with data sourced from the Quarterly Census of Employment and Wages program, which encompasses approximately 95% of jobs in the United States.
The QCEW offers a more thorough and precise understanding of the number of businesses, employees, and wages across the nation, as this information is sourced from state unemployment insurance tax records that most employers are mandated to submit. Considering that process, the QCEW experiences a notable delay: The release of the data for the third quarter of last year is scheduled for next month. The preliminary benchmark revision was an annual first-look estimate that coincided with the release of the first-quarter QCEW data. In September, the preliminary revision indicated that the US economy probably added approximately 911,000 fewer jobs than the initial estimates provided in the jobs reports for the 12-month period spanning from April 2024 to March 2025. When distributed, that amounts to approximately 76,000 fewer jobs each month. If the preliminary estimate were to materialize, it would effectively reduce the reported job gains for that period by fifty percent. No, the process of benchmarking and these significant adjustments to past employment do not serve as evidence of any malicious data-related activities, contrary to the unfounded assertions made by President Donald Trump and others.
This is a process that has been carried out by the BLS in various forms for 90 years. As former BLS commissioner Erica Groshen stated, it’s “not a bug; it’s a feature.” These and other revisions illustrate how a transparent and rules-driven organization accounts for and adjusts to new information as it becomes available, said Groshen and other former BLS officials. If it were to hold – and history has shown that the final revision is smaller – it would represent the largest downward revision on record, according to data dating back to 1979. Economists anticipate that the ultimate adjustment may result in a downward revision of 700,000 jobs. This time last year, the final annual benchmarked figure for the 12 months ending in March 2024 stood at a negative 589,000 jobs seasonally adjusted (-598,000 not accounting for seasonality). That’s considerably narrower than the initial estimate of -818,000 jobs, which continues to linger in some minds, despite not being the final figure. The final tally of nearly negative 600,000 marked the largest downward revision since March 2009, which had previously held the record at minus 902,000. This revision was also steeper than the downward adjustment of 489,000 jobs for the 12-month period ending in March 2019, during Trump’s first term. For context, the adjustments represent a small fraction (tenths of a percentage point) of total employment. Nonetheless, significant fluctuations, whether positive or negative, usually happen during periods when the economy is undergoing rapid changes that even the most finely calibrated models struggle to detect promptly.
The factors likely contributing to the upcoming downward revision include: declining survey response rates; the BLS’ modeling of business creation (known as the birth-death model) being disrupted by the pandemic and overestimating job gains; and gaps in measurement related to immigration. “This has been a half-decade with enormous changes to the economy – both the pandemic and the beginning and end of the immigration surge,” stated Jed Kolko. The benchmark revision – which ultimately impacts 21 months of not-seasonally adjusted data from April of the previous year through the subsequent year’s December – is not the sole adjustment in this forthcoming release. The BLS establishment surveys existing employers, which results in the omission of businesses that have recently opened or closed. BLS developed the “birth-death” model to encapsulate these dynamics. BLS has refined its birth-death model and will revise historical data generated under the previous model. The agency traditionally updates its seasonal adjustment models with each benchmark revision, which results in the past five years of seasonally adjusted data being impacted.