• Payrolls topped expectations as revisions were more negative than expected
  • The unemployment rate dipped
  • Markets barely reacted to stale, weather- and strike-driven effects with strong negative revision risk
 
  • Nonfarm payrolls m/m 000s / UR %, SA, March:
  • Actual: 178 / 4.3
  • Scotia: 80 / 4.3
  • Consensus: 65 / 4.4
  • Prior: -133 / 4.4 (revised from –92k / 4.4)

US nonfarm payrolls beat expectations and the unemployment rate edged lower in March.

The market reaction was muted because a) it’s backward looking, b) the nonfarm gain is likely to be heavily revised lower, c) positive weather and strike cessation effects boosted the number temporarily, and d) most of the rest of the world is shut for Good Friday—all of which means this will be a short note! The US 2-year yield was up by only about 3bps post-data with the 10s yield up by the same. S&P equity futures ignored the release.

Breadth was ok, not great, as shown in chart 1. Education and health sector hiring added 91k jobs in March which was driven by the end of the nurses’ strike in NY plus added momentum. Leisure/hospitality and construction benefited from what I had argued would be a positive weather effect. Most other categories were relatively little changed.

Chart 1: March Changes in US Non-Farm Payroll Employment

Private sector payrolls ex-health care remain volatile but have been up in two of the past three months which may be encouraging after a long string of disappointments (chart 2).

Chart 2: US Private Sector Jobs ex Health Care

Still, revision risk will remain high and negative and I continue to have low confidence in US data on jobs and inflation. Chart 3 shows that the collection rate on first pass remains very low due to budget cuts and staffing issues. I wouldn’t be surprised to see March payrolls get revised to under 100k next time. Revisions (chart 4) removed 7k jobs from the prior two months but 41k from February.

Chart 3: How the BLS Collects & Publishes Nonfarm Payrolls; Chart 4: US Non-Farm Payroll Revisions

The unemployment rate—derived from the companion household survey—edged a tick lower to 4.3% matching my deeply offside consensus call and halting the longstanding deterioration (chart 5). The participation rate (chart 6) shows why; the labour force shrank by ~400k in March, far worse than the 65k drop household survey employment.

Chart 5: US Unemployment Rate; Chart 6: US Participation Rate

Hours worked fell by –0.15% m/m SA which is the second consecutive drop. Q1 hours are tracking 0.9% q/q SAAR, extending the multi-quarter soft patch that means GDP growth relies mostly upon hopes for continued productivity gains.

Wages grew by 0.2% m/m SA, a slowdown from prior months (chart 7).

Chart 7: US Hourly Wages

Government hiring subtracted less than I had expected (-8k). Federal government losses (-18k) were smaller as hiring offset DHS no-shows. State/local governments added 20k. See the next page for charts on strike effects, hours, plus NSA and SA comparisons.

Chart 8: US Workers On Strike; Chart 9: US Total Hours Worked; Chart 10: Comparing US Payroll NSA for All Months of March; Chart 11: Comparing US Payroll SA Factor for All Months of March