GDPNow Forecast an Unbelievable 5.4%: I’ll Take the Under (Way Under)

The Atlanta Fed GDPNow Model forecast of first-quarter GDP rose to 5.5% today on the strength of ISM and construction reports.

Latest forecast: 5.4 percent — February 1, 2018

The GDPNow model forecast for real GDP growth (seasonally adjusted annual rate) in the first quarter of 2018 is 5.4 percent on February 1, up from 4.2 percent on January 29. The forecast of real consumer spending growth increased from 3.1 percent to 4.0 percent after this morning’s Manufacturing ISM Report On Business from the Institute for Supply Management, while the forecast of real private fixed-investment growth increased from 5.2 percent to 9.2 percent after the ISM report and this morning’s construction spending release from the U.S. Census Bureau. The model’s estimate of the dynamic factor for January—normalized to have mean 0 and standard deviation 1 and used to forecast the yet-to-be released monthly GDP source data—increased from 0.42 to 1.37 after the ISM report.

I’ll Take the Under (Way Under)

This is the third or fourth time that GDPNow kicked off with unusually high forecasts. Over time, those previous ridiculously- looking estimates came down. I expect the same to happen again.

For starters, ISM has been nothing but noise for at least a year. It has not matched actual factory output. Perhaps it does this time, but ISM missed the mark so often, I stopped reporting on it. Odds are the number is noise once again.

I do follow construction reports. However, construction reports are the most revised and volatile of the economic reports. I did not look at today’s report yet, but my previous comments apply: Both housing and autos got a one-quarter boost from hurricanes.

Neither will jump start the economy. Interest rates are rising and payback time is in the near future. In addition, consumers overspent at Christmas and credit card debt has soared. Again, I expect reversals.

Blow-Off Top

Could this be a blow-off top? Yes. But a blow-off top in sentiment, not GDP, is far more likely, as is a housing bust.

For discussion, please see “This Isn’t a Drill” Mortgage Rates Hit Highest Level Since May 2014.

Here’s the problem with models: They cannot think.

Mike “Mish” Shedlock