Third-quarter GDP growth was revised up more than expected, to 2.2%, from the original estimate of 1.6% (we and consensus had anticipated a revision to 1.8%).

Growth in final sales to domestic purchasers—that is, spending by U.S. residents—was revised down slightly, from 2.2% to 2.1%. Consumer spending growth was revised down to 2.9% (from 3.1%), while the decline in residential investment was steepened to 18.0% (instead of down 17.4%). These were roughly offset by stronger gains in business fixed investment (up 10.0%, instead of 8.6%) and government consumption (up 2.2%, instead of 2.0%).

Although U.S. spending was little changed, the revised foreign trade data showed that more of this spending was on domestic output, and less on imports, than first thought. Import growth was revised down from 7.8% to 5.3%. As a result, growth in final sales of U.S. output was revised up to 2.1%, from 1.7%.

The rest of the revision to GDP came from inventories, which added 0.2 percentage point to growth, instead of subtracting 0.1 percentage point. While the stronger growth in final sales is good news, it is counterbalanced by the faster accumulation of inventories, some of which may be unwanted.

On the inflation front, the revision brought a touch of good news in the core personal consumption price index—the Fed's favorite measure of inflation—whose annual rate of increase was revised down to 2.2%, from 2.3% (although that is still above the Fed's 1–2% comfort zone).

But there was more significant news on the labor cost front. The estimates of labor compensation for the second and third quarters have been revised down— massively—based on newly available evidence from the Quarterly Census of Employment and Wages (QCEW). The original estimates for the second quarter showed total employee compensation rising at a 7.4% annual rate. That has now been revised down to 1.4%, knocking $108 billion (at an annual rate) off employee compensation. Compensation growth in the third quarter (at 4.7%, instead of 4.9%) is little changed from before, leaving the third-quarter level of labor compensation now $113 billion below previous estimates.

How can such a large revision happen, and what does it mean? The second-quarter slowdown appears to be a correction after the first-quarter surge—when compensation growth reached a 12.6% annual rate. When this surge was first announced, it seemed likely that much of it was accounted for by temporary jumps in stock options and bonuses. If so, a correction should have followed in the second quarter. The initial estimates of second-quarter compensation growth did not incorporate such a correction, but now that hard data are available from the QCEW, it is clear that a correction occurred and it is now incorporated in the data.

What are the implications? First, household finances look worse than before, since with less income but roughly the same level of spending, the saving rate in the third quarter has been revised down to minus 1.3%, from minus 0.5%.

Second, and importantly for the Fed, the unit labor cost picture does not look so threatening. The compensation revisions will flow straight through into the revisions of unit labor costs (to be announced on December 5). On a year-on-year basis, unit labor costs probably rose at a 3.0–3.5% rate in the second and third quarters, instead of the 5.1–5.2% pace originally announced. While still a bit high for comfort, those would be far less worrisome increases, and would look less threatening for price inflation. The lower the increase, the more plausible the idea that it will be absorbed in profit margins, rather than passed on in higher prices.

And by the evidence in today's GDP report, profit margins are still extremely healthy. Pre-tax profits from current production are the best underlying measure of profits in the national accounts. Today's estimates saw a 30.9% year-over-year gain in profits on that basis. True, profits were depressed by hurricane-related write-offs in the third quarter of 2005, so that gain is somewhat inflated, but the quarter-on-quarter gain of 4.2% (simple rate) is not hurricane-distorted and is another impressive increase. On this basis, profits hit 12.4% of GDP in the third quarter, their highest share in nearly 56 years.

Our view remains that profits growth will slow into single digits during 2007, given our view that GDP growth will remain below trend, but the third-quarter performance means a stronger starting point than we had expected.

The upward revision to third-quarter GDP growth does not imply that we are now more optimistic about future growth. This is partly because some of the revision came in inventories rather than final sales, but more because indicators of consumer spending, residential construction, and business fixed investment have all been weaker than we had expected last month, when we projected fourth-quarter growth at 2.2%. It now seems more likely that growth will be in the 1.5–2.0% range.

Looking forward to 2007, the question remains whether the economy will bounce back to trend growth (3%) or above as the decline in residential construction becomes less steep, or whether the weakness in housing will spill over to the rest of the economy—e.g., consumer spending and retail construction. We expect some spillovers keeping growth subpar in 2007 (i.e., below 3%) and helping core inflation to ease lower, eventually paving the way for the Federal Reserve to lower interest rates (beginning in May 2007). But as Chairman Bernanke made clear on November 28, for now the Fed is still debating whether or not interest rates should go higher, based on continuing worries about inflation, not whether they should be cut.

by Nigel Gault

www.wmrc.com

www.globalinsight.com