GDPNow has had a hot hand in accuracy vs Nowcast for at least two consecutive quarters. Its estimate of fourth-quarter GDP jumped to 3.0% with real final sales, the true bottom line estimate of GDP at 3.3%.

The retail sales numbers were interesting.

Yesterday, I asked Core Group Retail Sales Jump 0.9% in Nov vs 0.2% Overall: Hurricane Distortions?

Retail sales only increased by 0.2% in Nov. but control group sales which exclude gas and autos jumped 0.9%. It’s the control group that feeds many GDP estimates but I do not know if that is true for GDPNow.

Control Group Exclusions

​

Exclusion Synopsis

1. If you exclude 44% of retail sales, then retail sales rose by 0.9%
2. Otherwise, retail sales rose by 0.2%
3. The former drives many GDP estimates.

Hurricane Distortions?

​

Florence was a killer. But people had plenty of advance notice coupled with lots of fear of what happened to others who did not flee recent hurricanes.

Florence Q&A

• What did people do? They drove away, saving their cars.
• What was ruined? Stuff left behind, especially appliances, electronics, and furniture,

Retail Sales Spotlight​​

• Gasoline sales fell 1.2% That’s easy to explain: The price of crude plunged.
• Furniture sales rose 1.2%. Really? With existing and home sales falling? Why?
• Electronic and appliances rose 1.4%. Really? With existing and home sales falling? Why?
• Nonstore retailers (think Amazon) rose 2.3% but this is part of an overall trend. It may also have been exacerbated by hurricane item replacements.
• The only strange item in this group is health and person care sales which rose 0.9%, but that is one of the smaller breakdown components.

Hurricane distortions or not, it seems pretty silly to exclude 44% of retail sales while bragging about retail sales.

Mike “Mish” Shedlock

​