FED: SF's Daly Eyes "One Or Two More" Cuts To Counteract Labor Weakness: Reuters

The BEA has issued a statement clarifying its approach to October PCE prices given the lack of CPI data collection for the month - where needed (which will be much of the report) it will average September and November CPI data for its estimates:

• "The Bureau of Labor Statistics did not produce its full suite of Consumer Price Index data for October. For those CPIs that are not available for October, BEA will use an average of BLS’ September and November CPI data for its estimates included in the personal income and outlays report for October. BEA uses the CPIs to adjust current-dollar consumer spending statistics for the effects of price change and to produce the Personal Consumption Expenditures Price Index."
• Recall that the BLS used a carry-forward imputation for large swathes of the CPI data in order to produce November price indices, which effectively meant they assumed October's price levels were the same as September's, an approach that was believed to have biased the data lower.
• That's especially so in the housing inflation categories which were clearly understated, but the 2-month change in the level for categories between September and November in many categories should largely have captured the "correct" inflation rate between the two months, though with the caveat that November's collection period amid holiday discounting was seen to have unduly depressed prices in certain categories.
• There will be October and November PPI out on Jan 14 which will have multiple PCE-readthrough categories that don't come from CPI however, including on airfares, financial services and healthcare.
• While the BEA will be able to cobble together October PCE as best it can, and this is a very defensible approach under the circumstances, we would expect the overall report to be taken with a grain of salt given the distorted CPI data. 

• Average hourly earnings are seen increasing 0.3% M/M in December which would be enough for a one tenth acceleration in the Y/Y to 3.6%, against a backdrop of steady average weekly hours worked at 34.3.
• Primary dealer analysts see a clear skew towards a dovish surprise though, with estimates in a 0.1-0.3% M/M range.
• The two months of average hourly earnings data in last month’s release were a mixed bag rather than the outright weakness that the headlines suggested, with non-supervisory employee wage growth running firmer and hours worked also increasing in November.
• Overall AHE growth of 0.14% M/M in November was clearly softer than the 0.3% M/M widely expected although it was countered by a stronger than expected 0.44% M/M in Oct (we had seen limited estimates with a median 0.3 but with risks skewed lower). Still, September was also revised lower to 0.19% M/M vs the previously estimated 0.25% M/M. The combination meant the Y/Y rate surprised lower, with 3.51% Y/Y (cons 3.6) after 3.75% in Oct for a fresh low since May 2021.
• Non-supervisory earnings painted a stronger picture however, at 0.35% M/M in Nov after 0.41% M/M in Oct and only a marginally downward revised 0.22% (initial 0.25%) in Sep. This typically less volatile category that captures about 80% of employees accelerated to a three-month high of 3.86% Y/Y from 3.81%.

• RES 4: 1.3950 61.8% retracement of the Nov 5 - Dec 26 bear leg     
• RES 3: 1.3925 Low Dec 4
• RES 2: 1.3892 50.0% retracement of the Nov 5 - Dec 26 bear leg   
• RES 1: 1.3844 50-day EMA 
• PRICE: 1.3809 @ 15:34 GMT Jan 7
• SUP 1: 1.3701/3643 Low Jan 2 / Low Dec 26 and the bear trigger
• SUP 2: 1.3637 Low Jul 25
• SUP 3: 1.3576 Low Jul 23
• SUP 4: 1.3540 Low Jun 16 and a key M/T support   

A bear theme in USDCAD remains intact and the latest recovery is considered corrective - for now. The pair has traded through the 20-day EMA, and this signals scope for a stronger corrective bounce. Attention is on the next important resistance at 1.3844, the 50-day EMA. A clear break of the 50-day average would highlight a stronger reversal. Key short-term support and the bear trigger has been defined at 1.3643, the Dec 26 low.