EJ Antoni mangles the data:

Given sampling error, it really doesn’t make sense to look at one month’s m/m change in January 2025 (why not December 2024?) against one month’s (preliminary) m/m change in January 2026. Here, I plot the 2024-2026 post-benchmark private employment.

Figure 1: Private nonfarm payroll employment (blue), and 2024 stochastic trend (red), 000’s, s.a. Source: BLS via FRED, and author’s calculations.

Using y/y changes, in the last year of Biden both private and government employment were increasing. In contrast, in the first year of Trump through December, private employment grew at a slower pace than under Biden, while government employment growth was negative.

And this guy wanted to be our latex salesman…