On Friday, I emailed Pat Higgins a couple of questions.
"Hi Pat, can you isolate the impact on GDP today, distinguishing the decline due to employment vs the factory report? And regarding employment, was it the anemic wage growth as opposed to the decline in employment that impacted PCE and residential investment?"
Pat Higgins' Reply
The revision in the model’s dynamic factor led to most of the declines in the PCE, equipment, nonresidential structures and residential investment contributions. If one simply replaces today’s estimates of the factors from July – December [the November and December values are forecasts] with the values from Wednesday’s published update, then the growth rates of these components are all within a few tenths of their values from yesterday. The October value of the dynamic factor is going to be heavily influenced by the employment report, rather than the Census manufacturing report, because the Census manufacturing data only goes through September. Here is what we wrote on our public blurbs for the last two updates:
The GDPNow model forecast for real GDP growth (seasonally adjusted annual rate) in the fourth quarter of 2017 is 3.3 percent on November 3, down from 4.5 percent on November 1. The forecasts of real consumer spending growth and real private fixed investment growth declined from 3.9 percent and 8.7 percent, respectively, to 3.2 percent and 6.0 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 October—normalized to have mean 0 and standard deviation 1 and used to forecast the yet-to-be released monthly GDP source data—declined from 1.50 to 0.63 after the employment report. The forecast of the contribution of inventory investment to fourth-quarter real GDP growth declined from 0.14 percentage points to -0.03 percentage points after this morning’s manufacturing report from the U.S. Census Bureau.
The GDPNow model forecast for real GDP growth (seasonally adjusted annual rate) in the fourth quarter of 2017 is 4.5 percent on November 1, up from 2.9 percent on October 30. The forecasts of real consumer spending growth and real private fixed investment growth increased from 2.8 percent and 4.4 percent, respectively, to 4.1 percent and 8.8 percent, respectively, after this morning’s Manufacturing ISM Report On Business from the Institute for Supply Management. The model’s estimate of the dynamic factor for October—normalized to have mean 0 and standard deviation 1 and used to forecast the yet-to-be released monthly GDP source data—increased from 0.04 to 1.43 after the ISM report.
Along with roughly 100 other series, the model uses something like 22 payroll employment series from different industries at various levels of aggregation and two from the household survey to estimate the dynamic factor. It doesn’t use any wage data from today’s employment release.
Most of the downward revision was in the inventory contribution to growth was, mechanically, due to an upward revision the model’s estimate of change in private inventories for the third quarter after today’s manufacturing report. See cells E26:F27 in the tab TrackingHistory of the online spreadsheet.
The dynamic model reacted most to the employment data. But in contrast to my expectation, the anemic wage growth was not the precise culprit.
Once again, Pat Higgins was very generous with his time in explaining his model.
Mike "Mish" Shedlock