October and November jobs data came in cooler than expected, but the long government shutdown muddies the waters. The jobs report alone won’t push the Fed to make more interest-rate cuts than previously communicated.
Takeaway: Mortgage rates are likely to remain largely unchanged following the release of mixed October and November jobs data. The data shows a cooling labor market that is staying afloat, but remains at risk of further deterioration. This data, on its own, is unlikely to change the expected trajectory of the Fed’s rate policy.
Private sector employment remained positive as the federal government shed jobs due to the deferred resignation program. 64,000 jobs were created in November, and 105,000 jobs were lost in October. However, the sharp decline in October was because 162,000 federal government jobs were lost. This was expected, and represented workers who accepted the deferred resignation program offered by DOGE earlier in the year.
Total private sector employment increased by 69,000 in November, and 52,000 in October.
These job creation numbers are expected to revise downwards in the coming year when the Bureau of Labor Statistics (BLS) benchmarks the more timely monthly survey with slower, but more comprehensive, data. Fed Chair Powell commented last week that Fed economists estimate job creation is being overstated by as much as 60,000 per month.
The only sectors that created jobs to any significant degree in October and November were health care and social assistance. All other sectors hovered around zero.
While job creation has fallen dramatically over the past year, much of the change is due to a slowdown in immigration. Economists estimate that the monthly “breakeven” rate of job creation given these demographic changes is around 50,000 per month.
The unemployment rate inched up slightly more than expected in November, but the government shutdown introduced some statistical noise into this measure. The unemployment rate increased to 4.6% in November, from 4.4% in September (October data was not collected). Economists and the Fed had projected an increase to 4.5%. However, the unrounded numbers show that the increase is smaller than it appears at first blush. The previous unemployment rate was 4.44% and it has increased to 4.56%, just barely rounding to 4.6% instead of 4.5%.
In interpreting the increase in the unemployment rate, however, the labor force participation rate rose, meaning more people wanted to work. The rate of employed workers to the population for people of prime working age held steady.
The federal workers furloughed by the shutdown should have been counted as employed for the purposes of calculating the unemployment rate.
Casting a shadow over all of the unemployment rate data, however, is additional statistical noise that has been introduced by the government shutdown. While the BLS was still able to collect October job creation data using its survey of firms, it wasn’t able to collect October unemployment rate data because that relies on a survey of households who are likely to suffer from greater recall bias. The BLS states: “The November 2025 estimates are associated with slightly higher than usual standard errors. This is due to multiple reasons: lower survey response, composite weighting changes, and the use of a 2-month period of analysis rather than a 1-month period. For example, the November unemployment rate required a 0.26 percentage point change to be statistically significant compared with a required change in September of 0.21 percentage point.”
Typically, the BLS uses October’s data to compute weights for November. The absence of that data changed their weighting procedure.
The response rate was lower than usual because the November data was collected later than usual. The shutdown ended November 12, during the “reference week” the BLS uses for this survey.
The Fed has a high bar for further rate cuts. Today’s data, with all of its uncertainties, is unlikely to push the Fed toward cutting again in the near future. Mortgage rates will remain steady for the time being.
