ON DECK FOR FRIDAY, SEPTEMBER 4th

KEY POINTS:

  • Markets await dual US, Canadian jobs reports
  • Nonfarm payrolls expectations
  • Canadian jobs to test strong recent momentum
  • German factories are doing very well

It’s barely even worth noting how global markets are performing thus far this morning since a lot could change once US payrolls arrive with a side order of Canadian jobs. Just for giggles, sovereign yields are rather flat across most markets except a fairly strong bull flattener in JGBs overnight with the longer-end down by about 10bps. Stocks are mixed with small losses across most European cash markets and either flat to small gains in US futures. Currencies are mixed with small moves other than mild outliers to yen softness and won strength. Oil prices are exhibiting a touch of weakness this morning.

Overnight developments were very light. Germany’s factories are doing very well of late with three consecutive increases in the volume of factory orders including +2.5% m/m SA in July and 3.7% in June. There was more tough trade talk from the unaccountable and unelected folks in Trump’s cabinet.

CANADIAN JOBS—TESTING MOMENTUM

Statcan’s Labour Force Survey for August will be released at 8:30amET with a hopefully fixed website. It arrives against the backdrop of solid recent momentum with 181k jobs created in the past three months while the unemployment rate has fallen from 6.8% at the end of last year to 6.4% in July.

Estimates for job growth range from roughly zero to 25k (Scotia 20k) with outliers once again coming from foreign firms that submit late. The unemployment rate is forecast to slip to 6.3% with consensus at an unchanged 6.4%.

Any surprises may be taken in stride by the BoC that generally looks to trends versus the latest spot data. That’s the way it should be for a noisy report with a +/-57k 95% confidence band.

Scenarios for the change in jobs if the only things we focus upon are seasonal adjustment factors and seasonally unchanged changes in jobs point to gains (chart 1). There has been steady upward drift in the SA factors for like months of August coming out of the pandemic due to the recency bias in the way they are calculated that overstates job gains. August is normally a down-month for seasonally unadjusted jobs.

A second consideration is the panel rotation methodology employed by Statcan that rotates out the first month and replaces it with the latest month in a rolling six-month target population of households. This can lead to momentum in hirings or firings by basically asking similar groups of individuals across similar panels. Chart 2 shows that when job growth has exceeded 150k or the prior three months (181k this time), the next month is usually up again.

There are other drivers and considerations that were flagged in my weekly’s preview (here), but one other thing to monitor is hours worked that have been on a tear of late. Based on the Q2 hand-off and July, hours have been tracking a gain of nearly 4% q/q SAAR. August’s reading could driver a revision to this estimate in either direction, but it’s important because GDP is hours times labour productivity and hence the figures are signalling momentum in the economy.

NONFARM PAYROLLS—WARSH ALREADY SAID TO IGNORE IT

It might take a lot for nonfarm payrolls to matter to Federal Reserve Chair Warsh. That’s because he has basically indicated everything is peachy with the US labour market. In regular writings I’ve tended to disagree with that and feel he’s imposing an unhealthy filter over the data and should be more balanced in how he reads development across both sides of the dual mandate.

Estimates for the change in payrolls run from roughly zero (one negative outlier) to +125k. Scotia’s estimate is +30k (ranked 3rd of 76 forecasters). Most estimates are within around 20k to 100k. The 90% confidence band of +/- 122k.

My weekly offered a full preview (here) along with an explanation of why I think Chair Warsh’s indifference to the state of the job market is misleading.

There is a wide range of possible outcomes this time. Chart 3 shows a range of possibilities if we apply last August’s SA factor to the range of seasonally unadjusted changes that are usually gains across like months of August in history. I think the middle scenarios around small losses and small gains is the most likely outcome within the noisy bands.

That shifts the focus to other considerations. Will local government education jobs that fell by about 50k in July post another drop as the expiration of ESSER funding has begun to hit education jobs? More school boards across the US have been reporting further job losses into the start of the new school year. Will health sector hiring continue to dampen somewhat as the lagging effects of ACA subsidy reductions work through? Will the leisure/hospitality sectors lose jobs again in light of readings like plunging air travel.

Other considerations were raised in my weekly preview but what I would mostly default to is a bottom-up deconstruction of the estimate by sector.