• Nonfarm payrolls have fully recovered from the pandemic’s shock…
  • ...as job growth smashed expectations again
  • Wage growth accelerated as the workforce shrank
  • Jobs and hours worked keep piling higher…
  • ...but productivity stinks, which often doesn’t end well
  • Another 75bps may be delivered at the September FOMC…
  • ...but with a lot of other ground to cover by then


  • US nonfarm payrolls m/m 000s // UR % // wages m/m %, SA, July
  • Actual: 528 / 3.5 / 0.5
  • Scotia: 290 / 3.6 / 0.3
  • Consensus: 250 / 3.6 / 0.3
  • Prior: 398 / 3.6 / 0.4 (revised from 372 / 3.6 / 0.3)

What rudely landed in the laps of market participants this morning was a resoundingly strong jobs report with one or two ongoing flies in the ointment. Jobs and wages say hello to 75bps in September for a data-dependent Fed that is obsessed with looking over its shoulder at the latest prints. Markets reacted by shoving the 2-year Treasury yield about 15bps higher and the 10s yield about 14bps higher while the USD appreciated and so far the S&P 500 is about ¾% lower. September is mostly priced for a 75 move by the Fed.

The US is amazingly 32k above the level of nonfarm payrolls from just before the pandemic in what is now a full job market recovery. I'm admittedly surprised by this, but here we are and they have indeed done so. The household survey is only 576k below pre-pandemic levels which for all its wonkiness is at good as even. Both levels of employment have essentially returned the US job counts to where they were (chart 1).

Chart 1: US Employment Levels

Revisions were minor and only added 28k to the prior two months of jobs estimates with most of that affecting the June estimates.

Wages accelerated. The 0.5% m/m SA rise in July and upward revision to 0.4% m/m SA in June turn in annualized wage gains of 5.8% and 5.4% respectively on a m/m SAAR basis. Wages have been running at this kind of trend pace pretty consistently since about April 2021 (chart 2). Our view remains that real wage growth will turn positive later in the year into next and that in the meantime consumers are bridging the effects of inflation through tapping wealth and liquid cash and near-cash holdings.

Chart 2: US Hourly Earnings

Hours worked are tracking a 2.7% q/q SAAR gain in Q3 after gains of about 3½% in each of Q1 and Q2 (chart 3). Hours worked were up by 0.4% m/m SA after a string of 0.3% monthly gains over the prior three months. Since GDP is an identity defined as hours times labour productivity (output per hour worked), the strength in hours worked in isolation of activity readings is much stronger than the GDP numbers suggest. That’s because labour productivity is tanking after a 7.3% q/q SAAR drop in Q1 and probably another ~5% drop in Q2 that we’ll get next week (chart 4). I’m still bothered as an economist by the fact that hours and bodies keep going up while productivity stinks. That often doesn’t end well and so be on guard toward the real possibility that jobs will eventually get hit even if they are currently exceeding expectations.

Chart 3: Total Hours Worked; Chart 4: US Labour Productivity

The labour force participation rate fell again by a tick but it has fallen since the pandemic-era peak in March of this year (chart 5). That’s derived from the household survey that showed the labour force fell by 63k after a prior drop of 353k. The US is having difficulty feeding the beast.

Chart 5: Participation Rate

The unemployment rate also fell a tick to 3.5% to tie September 2019 for the lowest unemployment rate since 1969 (chart 6).

Chart 6: US Unemployment Rate

Chart 7 shows high breadth to the gain in nonfarm payrolls. The private sector added 471k jobs and is on fire over recent months. Government jobs were also up by 57k and this time led by the state and local governments. 

Chart 7: July Changes in US Non-Farm Payroll Employment

Service sector jobs were up 402k with goods sector jobs up 69k. Within services, the leader were education and health (+122k) followed by leisure/hospitality (96k) and business services (89k with only a 10k assist from temp help).

Chart 8 shows how it is still a somewhat uneven job market recovery at least on a sector basis. Leisure and hospitality industries are still down during the pandemic to date with smaller drops in a couple of other categories, but those workers have been absorbed elsewhere.

Chart 8: US Employment Recovery to Pre-Pandemic Levels

Overall, I think numbers like these support the narrative that the Fed is going at least 4-handled on the policy rate and could be there by Q4. They also continue to lean against market pricing for rate cuts next year that have never made sense in my view. A lot more damage has to be incurred in order to credibly stand at chance at licking inflation at least cyclically, though probably not in long wave fashion as I’ll argue in the Global Week Ahead.