Retail sales decreased 0.2% in August according to the advance Census Bureau report. This was well below expectations, which called for a rise of the same magnitude. Moreover, the previous month’s surge in retail sales was cut in half, with the July gain now only reported as 0.3% m/m.
Sales at motor vehicle & parts dealers (-1.6%) subtracted from the headline – not surprising given the pullback in August auto sales to 16 million annualized. This decline was more than offset by the surge in gasoline stations spending , which rose by 2.5% on the month with the ex. autos and gas measure down 0.1% on the month.
Excluding gas, autos, building materials (-0.5%), and food services (+0.3%), the so-called ‘control group’ used in calculating GDP was down 0.2% on the month – below the 0.2% gain expected by economists. Gains in the control group were relatively broad, with miscellaneous (+1.4%), furniture (+0.4%), food & beverage (+0.3%) and general merchandise (+0.2%) seeing gains. Unusually, e-commerce sales were down 1.1%, alongside declines in clothing (-1.0%), electronics (-0.7%), and building materials (-0.5%).
Key Implications
Last month’s advance report made us giddy with excitement at the prospect that American consumers were back in full force. And then came this morning’s report, which effectively erased much of the outsized gains we believed occurred last month. Worse still, this morning’s report also indicated that consumers were reluctant to spend in the month of August, despite the strong job and income figures as of late. Overall, the new information suggests that consumption in the third quarter was not as strong as otherwise thought, with real PCE growth likely to be near the 2.5% mark – down from the 3.2% during the second quarter.
Still, as far as the decline in retail sales during August, we believe that most is likely related to the disruption in economic activity in southeast Texas. While some retailers likely experienced pre-emptive buying ahead of landfall, others likely faced prolonged closures due to weather conditions and lack of inventory. The Census Bureau indicated that reports suggested that the "hurricane had both positive and negative effects" on sales while others indicated no impact at all. Moreover, some firms reported due to "permanent or temporary store closures and stores having reduced business due to damage, shipment delays, etc."
Ultimately, the decline in July makes sense given the disruption in economic activity related to the storm. We expect that both Harvey and Irma are likely to drag down GDP growth in the third quarter by about half a point (in annualized terms) with a boost of a similar magnitude during the fourth quarter. As such, expect further volatility in the data going forward. Having said that, we expect the Fed to largely see through the volatility given its transitory nature.
