The 2010Q2 advance GDP release has been covered by Jim, as well as others. [RTE/Izzo] [CEA] [FreeExchange/RA] [CR] [MA] The release was accompanied by an annual revision of data extending back to data for 2007Q1. This revision alters our understanding (or lack of understanding in the cases of certain people) of the evolution of this recession. Here are the points I gleaned.
Figure 1: Log real GDP from 2010Q2 advance release (blue), from 2010Q1 3rd release (red), and potential GDP from CBO (January 2010), all Ch.2005$, SAAR. NBER defined recession shaded gray, assumes trough at 2009Q2. 6’6% is the implied output gap (in log terms) as of 2010Q2. Source: BEA, various releases, CBO, NBER, and author’s calculations. First, it is generally unwise to make definitive statements such as Donald Luskin’s September 14, 2008 assertion that “…anyone who says we’re in a recession, or heading into one — especially the worst one since the Great Depression — is making up his own private definition of “recession.” Figure 2 below shows the impact of the data revisions on growth rates.
Figure 2: Quarter-on-quarter growth rates SAAR, for 2010Q2 advance release (blue), 2010Q1 3rd release (red) and 2008Q2 preliminary release (green). 2008Q2 series in Ch.2000$. NBER defined recession shaded gray, assumes trough at 2009Q2. Source: BEA, various releases, NBER, author’s calculations. Then CEA Chair Ed Lazear’s statement “The data are pretty clear that we are not in a recession” is slightly less egregious because he was speaking on May 8, 2008, before a lot of the weak data had been reported. And implicit in his statement is the point that he was referring to the data at hand. But of course he knew the data were going to be revised … repeatedly. Anyway, you can see here previous revisions changing our understanding of the economy’s course [1] [2]
Second, this is indeed the deepest recession since the Great Depression, notwithstanding Professor Casey Mulligan’s assertions [3] [4]. Below in Figure 3 I plot the last recession against Mulligan’s conjoined 1980Q1-1980Q3 and 1981QIII-1982Q4 recessions; I’ve normalized on peak dates rather than trough dates as Professor Mulligan does, but the change is not essential to the message [5] (you can replicate the Mulligan graph using the NBER dates here).
Figure 3: GDP normalized on peak dates/recession begin dates for last recession (blue) and C. Mulligan’s 1980-82 recession (red). Source: BEA, NBER, and author’s calculations. To me, it is clear that this last recession was much, much worse than this 1980-82 “recession” Professor Mulligan alludes to. In addition, we came close to the 11 trillion (Ch.2000$) floor that he insisted we wouldn’t breach (I calculate 2009Q1 GDP when expressed in Ch.2000$ was 11.378 trillion, using a conversion factor of 0.88648, as explained in this post.) And Professor Mulligan’s October 2008 forecast was not conditioned on any stimulus package (It might have been conditioned on credit easing — not sure on this point). In any case, it is possible that a future comprehensive revision will drive the GDP figure experienced in this last recession very close to his threshold. (By the way, we long ago blew through his employment floor of 134 million, in January 2009.
Finally, the revisions provide additional information regarding the amount of slack in the economy: namely the output gap is now bigger than we were given to understand before.
Figure 4: Log output gap from 2010Q2 advance release (blue), from 2010Q1 3rd release (red). NBER defined recession shaded gray, assumes trough at 2009Q2. Source: BEA, various releases, CBO, NBER, and author’s calculations. In 2010Q1, the output gap was 6% (in log terms) using the pre-annual revision data. Using the post-revision data, the 2010Q1 output gap is now 6.8%, using CBO’s January 2010 estimate of potential GDP. (Presumably, CBO will revise its measure of potential GDP in light of BEA’s annual revision, but I expect the impact will still be there). Given this, it is no wonder that inflation indicators are muted. [6] As I said, we should’ve had a bigger stimulus.
My last point is directly related to the impact of revisions. Imports were a “negative contribution” to GDP growth in an accounting sense. However, higher imports are suggestive of higher economic activity, contemporaneously or in the future. In particular, one might think of capital imports as being very much related to expectations related to satisfying future demand (either domestic or foreign). In this regard, I think it’s important to observe that 46.5 bn (Ch.05$, SAAR) of the 92.5 billion increase in non-petroleum imports was attributable to capital goods imports.
Figure 5: Quarter-on-quarter real growth rates (Ch.2005$, SAAR) for nonresidential fixed investment (blue), equipment investment (red) and capital goods imports (green), calculated as log-differences. NBER defined recessions shaded gray; assumes last recession ends 2009Q2. Source: BEA, NBER, and author’s calculations. To the extent that investment is rebounding that’s a sign of some optimism about the future on the part of the private sector; increased capital goods imports is a reflection of that phenonenon — that’s about the only positive I saw in the report. However, even that positive is tempered by the fact that equipment investment remains 8.7% (log terms) below peak levels in 08Q1. (Nonresidential investment is 16.9% below peak.). Of course, hewing to my thesis, there’s a big caveat — the import series are based on two months worth of trade data. When the June release comes out, we may have a different take on how imports — including capital goods — have grown.
Similar Posts:
- More Data: Debt, and the Origins of Debt
- UK economy suffers blow as tepid growth confirmed
- Recession over for UK
- Will a Tax Repatration Holiday Spur Investment?
- Citi says number of credit cards hacked was actually 360,000
Recent Comments