US DATA
Retail sales started Q4 on a stronger footing than expected, with revisions pointing to solid consumption dynamics. But growth was largely driven in October by auto sales, and a slightly soft sequential Control Group (GDP Input) reading and weak breadth of growth across categories took the sheen off the report.
- Headline retail sales accelerated unexpectedly to 0.4% M/M (vs 0.3% expected), with the beat even more impressive due to the prior month being revised up (to 0.8% M/M vs 0.4% initial). Ex-Auto/Gas sales were a little softer on the month than per survey, but that's after large upward revisions to prior too (0.1% vs 0.3% expected, but prior 1.2% vs 0.7% initial).
- However a weaker than expected sequential Control Group reading of -0.1% M/M (0.3% expected) takes the sheen off slightly, even if prior was revised up to 1.2% from 0.7% initial. It suggests a more solid Q3 than previously expected, with higher levels of activity going into Q4 - though likewise that could hurt the sequential growth comparison for the Q4 data. 3M/3M annualized control group was a very solid 4.6% in October, though that's a deceleration from 5.7% in October (ie Q3) - that could prove slightly diminished given higher-than-expected CPI in October as well (retail sales are reported nominally).
- Category-by-category, breadth in retail activity was not strong: motor vehicles/parts - the largest category of retail sales - accelerated as had been expected, up 1.6% vs 0.2% prior. Electronics stores were the only other category to accelerate from September. Multiple categories saw contraction, including furniture, department stores, clothing, sporting goods,health/personal care, and miscellaneous store retailers.
- That could point to some hurricane impact, but as the Census Bureau notes in their methodology: "While a few individual firms can report large increases or decreases in their sales because of the effects of weather related events, this additional variation is not typically large enough to substantially affect the reliability of the published estimates."
319 words