On October 5, I commented GDPNow Forecast Rises to 2.7 Percent Annualized Despite Weak Economic Data
Today, despite weak retail sales data, especially inflation-adjusted, the model held firm.
I have commented many times that it’s not the data that matters but what the model expected vs the data, but on at least two occasions recently, the model anticipated weak data, even expecting worse than happened.
Base Forecast vs Real Final Sales
The real final sales (RFS) number is the one to watch, not baseline GDP. RFS ignores changes in inventories which net to zero over time.
The base GDPNow forecast is 2.8 percent and the RFS forecast is currently 2.9 percent.
If this is in the ballpark toss aside recession calls.
October 14 GDPNow Forecast
Please consider the latest GDPNow Forecast for the third quarter of 2022, emphasis mine.
The GDPNow model estimate for real GDP growth (seasonally adjusted annual rate) in the third quarter of 2022 is 2.8 percent on October 14, down from 2.9 percent on October 7. After recent releases from the US Bureau of Labor Statistics and the US Census Bureau, the nowcast of third-quarter real personal consumption expenditures growth decreased from 1.3 percent to 1.2 percent.
I am struggling with that paragraph and the huge surge earlier this month.
The GDPNow model expects real personal consumption expenditures of 1.2 percent despite very weak retail sales.
Factoring in Inflation, Retail Sales Remain Very Weak in September
In July, real retail sales fell 0.4 percent. In August, they rose 0.3 percent, and in September they fell another 0.4 percent.
Yet GDPNow estimates a 1.2 percent rise.
Scroll to Continue
For discussion, please see Factoring in Inflation, Retail Sales Remain Very Weak in September
Question on GDP Deflator
A reader asked me what deflator the GDPNow model uses for inflation in its forecast.
I have a reply from Pat Higgins, the GDPNow model creator, emphasis mine.
Hi Mish – there is not an implied GDP deflator, or at least not one that gets more accurate as data is accumulated over the quarter. For the monthly spending data that has not been released, the model only forecasts the real (price deflated) series, not the nominal series and its deflator.
For the nominal spending data the model does have, price deflators are constructed, when feasible, using standard monthly price releases like the CPI, PPI, and import/export price releases. The model does forecast quarterly GDP subcomponent price deflators using lagged quarterly data and a BVAR model so that it can put something into the chain-weighting formulas. It doesn’t use PPI/CPI/import+export price releases to improve those forecasts.
Using the correct deflators isn’t usually a first order issue if you are only forecasting one quarter into the future; i.e. you generally get close to the same thing forecast of the real series if you assume prices are unchanged from the previous quarter as you do if you plug in a reasonable forecast of the price deflators. So we don’t put those quarterly subcomponent price forecasts into the spreadsheet because those aren’t designed to be as accurate as possible, especially near the end of the quarter.
Best regards,
Pat
Blue Chip Forecast
I am not the only one struggling with the GDPNow surge. The Blue Chip professional forecast consensus is much lower and trending lower. Its data is as of October 6, thus not including today’s retail sales numbers.
Typically, the GDPNow model starts out very volatile then tends to be close to the mark by the end of the quarter.
As it stands now, I believe GDPNow is way high, but we still will have skirted recession if the Blue Chip assessment is correct.
Still, the Blue Chip range is very wide, from less than zero to about 2.7 percent.
This post originated at MishTalk.Com
Please Subscribe!
Like these reports? I hope so, and if you do, please Subscribe to MishTalk Email Alerts.
Subscribers get an email alert of each post as they happen. Read the ones you like and you can unsubscribe at any time.
If you have subscribed and do not get email alerts, please check your spam folder.
Mish
Source link
Author Mish