Viewpoint

November 21, 2025 | 14:41

Fed Policy: Old Data, New Concerns

Economist

Deputy Chief Economist and Managing Director
Economics

Michael is part of the team responsible for forecasting and analyzing the North American economy and financial m... (Read more)

Priscilla Thiagamoorthy

Senior Economist and Vice President
Economics

Priscilla joined BMO Capital Markets in 2016 and plays a key role in analyzing Canadian household trends and the... (Read more)

Michael Gregory , CFA

Deputy Chief Economist and Managing Director
Economics

Michael is part of the team responsible for forecasting and analyzing the North American economy and financial m... (Read more)

Priscilla Thiagamoorthy

Senior Economist and Vice President
Economics

Priscilla joined BMO Capital Markets in 2016 and plays a key role in analyzing Canadian household trends and the... (Read more)

Our Thoughts

Fed Policy: Old Data, New Concerns

September’s report on the Employment Situation was delayed by 48 days because of the government shutdown. Although the data were dated, they were still useful in gauging the labor market’s autumnal trajectory. Nonfarm payrolls increased by 119k, which was stronger than expected, but there were 33k worth of net downward revisions. August’s print is now -4k, marking the second negative figure over the past four months. (Note: repeated negative employment prints are historically the stuff of recessions.) The three-month trend is now +62k, which continues to register the weakest results since the start of the pandemic and, before then, since the aftermath of the Great Recession.

Meanwhile, the unemployment rate rose one-tenth to 4.4%, which is the highest since October 2021. And it was a ‘high’ 4.4% at 4.440% unrounded. This marks the third straight one-tenth rise in a row, as employment growth has not kept pace with workers net entering the labor force. Hiring caution amid economic uncertainty is likely a factor. And we reckon the proliferation of AI applications in the workplace is also playing a role, amid the broader push to boost productivity and control costs.

On balance, the September data revealed a weak trend on the ground (at the time). And for the Fed, as next month’s policy decision approaches, a key question is: Has the trend worsened, improved, or stabilized? Unfortunately, answering the question has again become problematic.

This week, the Bureau of Labor Statistics (BLS) announced that the employment report for November would be published on December 16 instead of the scheduled December 5. It arrives after the next Fed confab on December 10. And a partial report for October will be published alongside, with establishment survey figures only (so no unemployment rates, for example). Thus, policymakers are going to have to employ two months’ worth of private sector and other surrogate data to fill in the labor market blanks for October and November. The Fed did this before, with the absence of the September data for the October 29 meeting. At the time, the alternative data were a mixed bag. In their own contexts, some deteriorated and others improved, but all were sitting on the ‘weak’ sides of their ledgers.

However, having to fill in for two months’ worth of missing figures instead of one does lift the risk of misinterpreting the state of the labor market as defined by the official data. This should make some Fed officials more cautious, other things equal. In last month’s FOMC meeting, there was already one dissenting vote in favor of not reducing policy rates. And participants’ post-meeting public statements point to others sporting a greater sense of caution.

Indeed, the Minutes from the meeting (released this week) said “many participants suggested that, under their economic outlooks, it would likely be appropriate to keep the target range unchanged for the rest of the year” (note that ‘many’ is still a minority). But this should not be a surprise. As far back as September’s ‘dot plot’, it was clear that the FOMC’s conviction about an October-December rate cut couplet was not that strong to begin with. There were 10 participants in the two-or-more moves camp and nine in the one-or-none group.

And there is a new wrinkle for the Committee to contend with: the availability of official inflation data. This turned out not to be an issue for last month’s confab. The CPI for September arrived in time for the October 29 meeting (and the PCEPI wasn’t scheduled until two days after anyway). BLS employees were temporarily recalled from furlough to produce the data needed for annual Social Security cost-of-living adjustments. Note that 40% of the observations were imputed (meaning local price changes were replaced by broader regional moves) which was the highest share so far this year (and as far back as 2019), a share that has been rising due to labor shortages at the BLS.

Apart from concerns about data accuracy, the CPI results were also a little concerning from a price stability perspective. For the total, core, core goods, and core services excluding rents metrics, nearly all the three- and six-month changes accelerated from their August paces (nothing slowed). And all the three-month moves outpaced their six-month changes. The policy dilemma posed by a weak labor market and sticky inflation becomes much more complicated when there is an absence of official data on both fronts.

The Bureau of Economic Analysis (BEA) has not yet published a date for September’s PCEPI, and the October release (due next week) has officially been postponed (no re-scheduled date given). The BLS has canceled October’s CPI and the November release, due originally on December 10 (Fed decision day), has been delayed to December 18. Come December 10, and with respect to the data, it appears that the Fed will be flying even more blind than it was last month.

With policy rates still on the restrictive side (slightly), and our judgment that the Fed’s risk management approach still has the weakness in the labor market trumping the stickiness of inflation (slightly), we reckon the odds still favor another rate cut next month (slightly). The market is currently near 70% odds, but it has been as low as under 35%. For the Fed, it’s going to be close call.