This note provides a brief assessment of how to interpret the consensus guess for the fourth quarter GDP release, due on Thursday. The main contention here is that the consensus has a stronger view of the economy than the consensus realizes, although this is slightly less the case for my single favorite measure of underlying demand.
The screen consensus has the GDP rising 2% (ar) during the fourth quarter. Built into that is just over 50 bps of drag from net exports and (especially) inventories collectively. And government spending is roughly neutral. So, the implied consensus for final sales to private domestic purchasers (FSPDP) is a gain of 2.5%. That would bring the 2-quarter growth rate there to 2.8%, which would be the strongest 2-quarter gain since the peak of the stimulus-fueled mid-Covid bounce. See the right panel of the chart below for a picture of that in level terms.
Conventional measures of aggregate demand
Source: BEA, FH calculations and estimates
Data are actual to Q3 and implied consensus for Q4.
The demand bulls, among whom I tend to include myself these days, now seem to have forgotten about the GDP – GDI distinction that was a focus two to three quarters ago. But the case for using the average of GDP and GDI (i.e., GDO) as the better proxy of aggregate demand has not gone away. There was recently a technical reason for being suspicious about that because of a weird handling of Fed portfolio losses, but the bean counters have fixed that issue, so far as I understand. So, let’s take a look at GDO and then an analogue of final sales to private domestic purchasers adjusted for the same consideration, which I concede involves taking some liberties. As often, I am going for less wrong here.
We have no reason to believe that the bean count for GDO should be any different from that for GDP. When the data come out, the growth rates there will almost certainly be different, but we have no basis at this point for saying which way. With GDI below GDP in level terms, we might expect some catch up, which would redound to the favor of the demand bulls. But the evidence does not suggest a catch-up effect, according to analysts who have actually looked into this formally. So, we have GDO up 2% in Q4 and at an annualized rate of 2.6% over the past two quarters, as the left panel of the chart below shows.
Incorporating the GDP – GDI distinction, including with a minor kludge
Source: BEA, FH calculations and estimates
Data are actual to Q3 and implied consensus for Q4.
We can work up an analogue for FSPDP by taking the standard measure and multiplying it by the ratio of GDO to GDP, admittedly a minor kludge. That would be up 2.5% in Q4 and at an annualized rate of 1.9% during the two quarters to Q2. In fairness, 1.9% is not that strong, and yet it is my single best guess of the underlying trend of aggregate demand growth, because I do not discount the GDI vs GDP distinction. Still, the pattern is a quickening one. These would be the strongest results for the 1- and 2-q growth rates there in six quarters.
I had a discussion with colleagues Peter and Kevin about the consensus’s very weak outlook for growth in Q1. We were wondering if the Bloomberg numbers are actually stale or if they are a current read of estimators who just update the priors slowly. I have been assured by Bloomberg staff that the numbers are current in the sense of reflecting what the guessers are actually saying, but I concede that the gap between what is on ECO (2%) and ECFC (1.3%) for even Q4 GDP is a fly in the ointment. Whatever the case, what is on ECFC does seem low. I badly assert that the typical guesser there is leaning heavily on priors and maybe a bit less so on the more typical approach of just applying a sense of momentum overlaid with whatever technical issues in the GDP bean count might be relevant. For example, after the inventory drag, we all have slated in for Q4, the flow of inventory investment would be “normal,” that is, inclined neither to accelerate nor decelerate on the basis of mean reversion considerations, which do in fact work.
Where my own take would be exposed here would be around the January employment report, due February. So far, the screen consensus of a 145k gain in private jobs, is fully with me. But we shall see.
We might be wrong ourselves, but we were wondering if the survey is stale or the submitters sluggish
Source: Bloomberg ECFC
