Last year’s fourth-quarter GDP posted a small decline of 0.1%, BEA reported last week. The slight contraction came as a surprise to many analysts, including yours truly. The final Q4 nowcast on these pages, posted a few days before the government’s report was published, anticipated a 2.0% gain in 2012’s final quarter. What happened? Based on BEA numbers, an extraordinarily large decline in defense spending dragged down GDP in last year’s fourth quarter, as I discussed previously. Citing BEA numbers, national defense spending tumbled a hefty 22.2% in Q4 vs. a 12.9% rise in Q3. Quarterly declines of that magnitude are unusual by the standards of the past several decades. Unusual or not, the red ink caught most analysts by surprise. For some context, non-defense federal spending rose 1.4% in Q4, down from 3.0% in Q3.
Exect A Closer Match
The Capital Spectator’s nowcasting models focus on private-sector data. That focus isn’t usually a problem, unless there’s an unexpectedly steep drop in government spending, which was the case in Q4. Assuming that the fall in government spending was a one-time event, the GDP nowcasts going forward should offer a closer match with the final data. Regardless, the intention on these pages is one of evaluating the ebb and flow of the private sector, the dominant driver of the business cycle... usually.
Keep in mind that the relatively upbeat GDP nowcast for last year's Q4 is a sign that the private sector continues to grow, despite what the headline GDP number implies. Then again, factoring in changes in government spending can alter the big picture, as we learned the hard way last week. That said, it's an open debate if a similar glitch awaits in this year’s first quarter.
In any case, the current nowcast for this year’s Q1 continues to focus on private-sector data, and on that front the initial outlook is comfortably in the black. The usual caveats apply, of course, starting with the fact that several months of Q1 updates are yet to come and so today's nowcast will be revised as new numbers arrive. As usual, I’ll publish the updates and compare the revisions with the previous nowcasts.
For now, here’s how the current nowcast compares with several recent forecasts for Q1 from other sources:
GDP: Q1 preview
Also, here’s a look at the individual nowcasts:
Finally, here's a brief profile for each of The Capital Spectator's nowcasts:
R-4: This estimate is based on a multiple regression in R of historical GDP data vs. quarterly changes for four key economic indicators: real personal consumption expenditures, real personal income less government transfers, industrial production, and private non-farm payrolls. The model estimates the statistical relationships from the early 1970s to the present. The estimates are revised as new data is published.
R-10: This model also uses a multiple regression framework based on numbers dating to the early 1970s and updates the estimates as new data arrives. The methodology is identical to the 4-factor model above, except that R-10 uses additional factors -- 10 in all -- to nowcast GDP. In addition to the data quartet in the 4-factor model, the 10-factor nowcast also incorporates the following six series:
- ISM Manufacturing PMI Composite Index housing starts
- initial jobless claims
- the stock market (S&P 500)
- crude oil prices (spot price for West Texas Intermediate)
- the Treasury yield curve spread (10-year Note less 3-month T-bill)
ARIMA 4: This model is similar to the ARIMA technique above in terms of the econometric application, but with a key difference. Instead of using historical GDP data as a lone input, the ARIMA 4 model analyzes four historical data sets to predict GDP: real personal consumption expenditures, real personal income less government transfers, industrial production, and private non-farm payrolls.
VAR-4: This vector autoregression model uses four data series in search of interdependent relationships for estimating GDP. The historical data sets in the R-4 and ARIMA 4 models above are also used in VAR-4, albeit with a different econometric engine. As new data is published, so too is the VAR-4 nowcast. The data sets range from the early 1970s to the present, using the "vars" package in R to crunch the numbers.