Daily Points

ON DECK FOR FRIDAY, NOVEMBER 21

KEY POINTS:

• Markets continue to play defence after nonfarm fallout
• The one thing that everyone should be asking the BLS about nonfarm
• Global PMIs signal weaker growth in Japan, Australia…
• …cooler growth in the UK and India…
• …and no change in the Eurozone with the US on tap
• UK consumers retreated on eve of Starmer’s budget
• Don’t look for a Blue Jays effect in today’s Canadian retail numbers

Sovereign yields continue to move moderately lower this morning. Equities continue to be soft. Fallout from yesterday’s US job readings continues as the nonfarm details were soft, the SA factors appeared cooked (see below), the unemployment rate increased and continuing claims increased which signals further increases in the unemployment rate. Yet Fed-speak was mixed, although it’s still important to weight the perspectives by voting status. Overnight releases didn’t help via a mixed picture for global PMIs and soft UK data ahead of US PMIs and Canadian retail sales.

NONFARM FALLOUT—IT’S ALL IN THE CHOSEN SEASONAL ADJUSTMENTS

I think the BLS has to account for why it went with a record high seasonal adjustment factor in yesterday’s nonfarm payrolls report. The SA factor was off the charts compared to any prior month of September on record (chart 1). This distorted payrolls as argued.

There is never any explanation other than hiding behind an opaque statistical estimation. Yet it was the single reason why payrolls remained positive despite one of the weakest seasonally unadjusted payroll numbers on record for like months of September. Had the BLS opted for last year’s SA factor or continued to drift lower given greater distance from the pandemic given the recency bias to how the SA factors are calculated, then payrolls would have declined (chart 2). 

The chart shows how seasonally adjusted payrolls changed as reported, how seasonally unadjusted payrolls changed as reported, and then what seasonally adjusted payrolls would have been at SA factors less than record highs such as using the 2024 September SA factor or continuing the downward drift in SA factors as distance from the distorted pandemic era yields less influence upon the recency bias to how they are calculated.

The number that was revealed was sure to curry favour with Trump. Sure to avoid his criticism had it shown job losses. Perhaps it lessened the risk that another obviously unqualified candidate like Antoni would be nominated to head the BLS, or perhaps it raised the odds that an internal candidate would be chosen.

In short, my trust for this data is pretty low.

PMIs REVEAL A MIXED PICTURE

The monthly parade of purchasing managers’ indices marched on overnight with the US readings still on tap. Quicker growth was signalled in Japan and Australia, cooler growth was indicated by readings from the UK and India, and the Eurozone was unchanged and still signalling moderate growth. Composite PMI trends are shown in chart 3.

• Eurozone: The composite PMI was flat (52.4, 52.5 prior) given little change in services (53.1, 53.0 prior) and manufacturing (49.7, 50.0 prior). Germany’s composite PMI decelerated mainly due to services but also manufacturing while France accelerated because of services.
• UK: The composite PMI fell significantly to 50.5 (52.2 prior) which signals a stalling economy. The decline was entirely due to services (50.5 from 52.3) as manufacturing accelerated but to still no real growth (50.2, 49.7 prior).
• Japan: The composite PMI moved up half a point to 52.0 entirely due to a 0.6 gain in manufacturing to 48.8 which signalled a slower pace of contraction while services remained at 53.1 which continues to signal moderate growth.
• India: The composite PMI slipped half a point to 59.9, signalling solid ongoing growth. The decline was entirely due to manufacturing (57.4 from 59.2) that nevertheless continues to post solid growth while services accelerated (59.5, 58.9 prior).
• Australia: Quicker growth was signalled by a 52.6 composite PMI (52.1 prior) as manufacturing accelerated (51.6, 49.7 prior) and services were little changed (52.7, 52.5 prior).
• US: The November PMIs are due out this morning (9:45amET). They’ve been signalling moderate overall growth that is led by services.

UK DATA SOURS BEFORE BUDGET

Macro data soured on the eve of the Starmer administration’s second budget that is coming next week. The result contributed to lower yields across the curve. UK retail sales fell sharply in October. Total sales volumes were down by 1.1% m/m (-0.2% consensus) as the prior month was revised up a touch (0.7% from 0.5%). Ex-fuel sales volumes were down 1% m/m, or double consensus, with minor revisions.

CANADA REFRESHES RETAIL SALES FOR SEPTEMBER–OCTOBER

Retail sales are expected to dip when September’s final estimate arrives, but key will be October’s fresh guidance. Statcan had advised a month ago that September’s reading would decline with consensus estimates around -0.7% m/m SA, but that guidance can be materially revised. October’s reading will be the first guidance and will give us a better feel for how the consumer is performing in Q3 into Q4. Current pre-data tracking is shown in chart 4 up to Q3.

A big caveat is always that Canadian retail sales are a fraction of consumer spending. They include no services unlike, say, the US that at least includes restaurants and bars. Ergo, don’t look to retail sales to show much of any impact of the “world” series for one thing.