(Click on image to enlarge)

The GDPNow model forecast jumped to 5.4% on February 1, only to come crashing down the next day.

Pat Higgins, creator of the GDPNow model addresses the question GDPNow's Forecast: Why Did It Spike Recently? on today's Atlanta Fed Macroblog.

If you felt whipsawed by GDPNow recently, it's understandable. On February 1, the Atlanta Fed's GDPNow model estimate of first-quarter real gross domestic product (GDP) growth surged from 4.2 percent to 5.4 percent (annualized rates) after a manufacturing report from the Institute for Supply Management. GDPNow's estimate then fell to 4.0 percent on February 2 after the employment report from the U.S. Bureau of Labor Statistics. GDPNow displayed a similar undulating pattern early in the forecast cycle for fourth-quarter GDP growth.

What accounted for these sawtooth patterns? The answer lies in the treatment of the ISM manufacturing release.

In the Atlanta Fed's GDPNow factor model, the last month of ISM manufacturing data have large weights when calculating the terminal factor value right after the ISM report. These ISM weights decrease significantly after the employment report, when about 50 of the indicators have reported values for the last month of data. [Mish note: the next day]

A possible shortcoming of the GDPNow factor model is that it does not account for the previous month's forecast errors when forecasting the 127 indicators. For example, the predicted composite ISM PMI reading of 54.4 in December 2017 was nearly 5 points lower than the actual value.

If we decide to incorporate adjustments to GDPNow's factor model, we will do so at an initial forecast of quarterly GDP growth and note the change here. [Mish Note: "Here" is download of a PDF of GDPNow Model Modifications]

Would the adjustment have made a big difference in the initial first-quarter GDP forecast? The February 1 GDP growth forecast of GDPNow with the adjusted factor model was "only" 4.7 percent. Its current (February 9) forecast of first-quarter GDP growth was the same as the standard version of GDPNow: 4.0 percent. These estimates are still much higher than both the recent trend in GDP growth and the median forecast of 3.0 percent from the Philadelphia Fed's Survey of Professional Forecasters (SPF).

Most of the difference between the GDPNow and SPF forecasts of GDP growth is the result of inventories. GDPNow anticipates inventories will contribute 1.2 percentage points to first-quarter growth, and the median SPF projection implies an inventory contribution of only 0.4 percentage points. It's not unusual to see some disagreement between these inventory forecasts and it wouldn't be surprising if one—or both—of them turn out to be off the mark.

Not Whipsawed

On February 1, I commented GDPNow Forecast an Unbelievable 5.4%: I'll Take the Under (Way Under).

I'll Take the Under (Way Under)

The GDPNow model rose to 5.4% today on today's economic reports. Color me skeptical, extremely skeptical.

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.

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

Unthinking Models

My comment about models not being able to think does not mean I find no use in the GDPNow forecasts. Nor is it an indictment of either Pat Higgins or his model.

I have high respect for Pat Higgins and he is always generous with his time. I have learned a lot from him.

Rather, my comment is simply that people need to do a bit of thinking on their own. Of course, people do not always think clearly, including me. This is the point of models in the first place.

When I disagree with the models, I give my reasons. And in this case, we have seen the ISM problem before.

We have also seen the FRBNY Nowcast model go haywire for similar reasons. For example, in December of 2016 Industrial Production numbers surged but it was not business expansion. Instead, utility production surged because that December was one of the coldest Decembers on record. The models failed to spot the difference and forecasts surged.

In January, when Industrial Production declined, the models took it back. I wrote about this in Industrial Production “Unexpectedly” Declines Due to Weather.

Models do not evaluate why, they just see numbers and extrapolate them.

Pat Higgins is reevaluating his model, noting "A possible shortcoming of the GDPNow factor model is that it does not account for the previous month's forecast errors when forecasting the 127 indicators."

I also wonder if the dynamic values of GDPNow are simply too high. ISM has not tracked reality for quite some time.