Tales from the Employment Situation release for July: (1) July Establishment employment change below consensus; (2) Revisions make trends slower; (3) Adds data indicating a slowdown.
Figure 1: Nonfarm Payroll Employment, July release (bold black, right log scale), implied Bloomberg consensus (gray +, right log scale), June release (teal, right log scale), Private NFP, July release (blue, left hand log scale), implied Bloomberg consensus (blue +, left hand log scale), June release (tan, left hand log scale), all in 000’s, s.a. Implied Bloomberg estimates calculated by iterating consensus change on top of June release figures. Source: BLS, Bloomberg, and author’s calculations.
While one observation doesn’t make a trend (+73K consensus vs. +106K for NFP), three observations might. We don’t make judgments on the basis of one preliminary observation because the Mean Absolute Revision going from first to third revision over the 2022-24 period is about 40K. Hence, it’s very possible that in two month’s time, the 3rd release for July’s data could end up being above consensus. Still, given the revisions in the previous months data, it looks like a deceleration.
Here’s a picture of several series depicting the slowdown along other dimensions of the labor market, and other economic indicators:
Figure 2: Monthly GDP (black), private nonfarm payroll employment-ADP (blue), civilian employment w/smoothed population controls (light blue), aggregate hours (light green), consumption (tan), personal income ex-transfers (pink), real retail sales (purple), all in logs 2025M01=0. Real retail sales is 3 month centered moving average of retail sales, divided by chained CPI. Source: S&P Global, ADP-Stanford, BLS, BEA, Census, and author’s calculations.
This seems to have been taken by the market as signaling a slowdown. The CME implied Fed funds rate dropped from 4.3% to 4.2% going from yesterday to today at 10AM CT (although admittedly the Trump tariff announcements might have had some impact as well). Betting markets are sending the same message:
Source: Kalshi, accessed 8/1/2025 10AM CT.
Assuming the greater than 25bps cut group is indicating 50bps, then the implied drop in September is 0.2 percentage points.
Total payroll employment rose an average of 35k in the 3 months to July. In the 3 months to April, the average monthly rise was 127k, nearly 4 times faster.
Private payroll employment rose an average of 52k per month in the 3 months to July, versus 118k in the 3 months to April.
Households reported employment FELL an average of 74k in the 3 months to July versus an average gain of 25k jobs In the 3 months to April.
ADP reports an average monthly job gain if 47k since May, 139k in the 3 months to April, nearly three time faster.
There’s a pattern here, methinks. What changed in April?
Perhaps the bigger story is what Trump did in response:
“President Trump is firing the commissioner in charge of the Bureau of Labor Statistics (BLS), which handles the monthly jobs report, after a bad report showing the economy in trouble was released on Friday.”
” “I have directed my Team to fire this Biden Political Appointee, IMMEDIATELY. She will be replaced with someone much more competent and qualified. Important numbers like this must be fair and accurate, they can’t be manipulated for political purposes,” he added. ”
https://thehill.com/homenews/administration/5430914-live-updates-jobs-economy-trump-tariffs/