GDP surprised on the upside at 2.8% q/q AR vs. Bloomberg consensus of 2.0%. One purported financial adviser writes:

2.8% Fake GDP is number are reported by biased govt agencies. What a sham…..believe nothing question everything. Does anyone believe economy is same now as it was 6 yrs ago?

I’ll note that GDPNow was at 2.6% on the day before the release, as was Goldman Sachs tracking. Here’s a picture of GDP in levels, versus nowcasts and forecasts.

Figure 1: GDP as reported (bold black), Bloomberg consensus (pink triangle), GDPNow as of 7/24 (teal square), and WSJ mean survey (blue), all in bn.Ch.2017$, SAAR. Source: BEA 2024Q2 advance, Atlanta Fed, Bloomberg, WSJ, and author’s calculations.

Hence insofar as the growth rate is concerned, advance GDP was only slightly above GDPNow. 2.8% does seem far above the Bloomberg consensus, but it’s important to remember that the advance release is just that — advance, with lots of forecasted components. The standard error of revisions going from advance to third release is 0.5 ppts (the standard error from advance to final is 1.2 ppts, which is part of the reason why NBER does not place primary importance on GDP).

As an aside, 2.8% is still less than the 3.4% forecasted by the irrepressible James F. Smith in the WSJ July survey, (almost) always the leader of the pack.

In one respect, skeptics might have some reason to question — not the motives but the accuracy — of estimated GDP, insofar as alternative indicators suggest lower levels of economic activity.

Figure 2: GDP (bold black), GDO (tan), GDP+ (green), all in bn.Ch2017$ SAAR. GDI for 2024Q2 calculated by predicting 2024Q2 net operating surplus using GDP, lagged surplus, lagged differenced surplus, 2021Q1-2024Q1.  GDP+ level calculated iterating growth rates on 2019Q4 GDP level. Source: BEA 2024Q2 advance release, Philadelphia Fed, and author’s calculations.

In 2024Q1, reported GDP is 1.1% above GDO.

More By This Author: