So earlier today, following the NFP number (which was of course a beat), the Atlanta Fed lowered their GDPNow tracking forecast for Q3 growth to 3.7% from 4%.
On August 4, the #GDPNow model forecast for real GDP growth in Q3 2017 is 3.7% https://t.co/ATnmStLj32 pic.twitter.com/sAeHdUy6K3
— Atlanta Fed (@AtlantaFed) August 4, 2017
Here was the rationale (assuming you care, which I’m reasonably sure you do not):
The forecasts of third-quarter real consumer spending growth and real fixed investment growth declined from 3.0 percent and 5.2 percent to 2.8 percent and 4.1 percent, respectively, after this morning’s employment report from the U.S. Bureau of Labor Statistics. The model’s estimate of the dynamic factor for July–normalized to have mean 0 and standard deviation 1 and used to forecast yet-to-be released monthly GDP source data–decreased from 0.64 to 0.27 after the report.
Ok, whatever, right?
Well here’s the thing about that 3.7% forecast: it’s still pretty damn robust, wouldn’t you say?
Now see if you can spot the problem in the following chart from Citi:
Let’s go to Jimmy McMillan to explain:
The most recent quarter will revise to sub 2% and that is with an excessively overshooting model. Speaking of models, a productive discussion might follow from complete publication of how the current 3.6% number was derived. But the FED is not required to hand in its homework, or even show its work.