U.S. Economy Returns to Growth in the Third Quarter

• The economy is growing again, after four quarters of decline, and across most major categories of spending.
• Growth was not solely driven by the inventory cycle. Real final sales rose 2.5%, compared with a 0.7% increase in the second quarter.
• Inventories added 0.9 percentage point to growth, simply because they were falling less steeply in the second quarter. But they were still falling fast. That means there is plenty of support still to come from the inventory cycle in coming quarters.
• Consumer spending, business equipment and software spending, residential fixed investment, and exports all returned to growth in the third quarter. Consumer spending rose 3.4%, after falling 0.9% in the third quarter. Spending on motor vehicles and parts rose at a 56.3% annualized rate, due to cash-for-clunkers incentives, but that was not the whole story. Motor vehicles contributed less than half of the overall consumer spending increase. There were widespread improvements across all major consumption categories. Overall, consumer spending growth will slow in the fourth quarter, but not go into reverse.
• Exports returned to growth (up 14.7%), as did imports (up 16.4%). That is a good sign of a revival in world trade activity, even though trade was overall a 0.5-percentage-point drag on growth. We should expect trade to remain a drag in coming quarters, as the swing in the inventory cycle will pull in more imports.
• Inflation was very low. The GDP price index rose just 0.8%.

Caveats

• Growth was still heavily dependent on the federal government. Federal spending added 0.6 percentage point to growth. Some of that reflected stimulus spending, although most of the increase was in defense and is probably not stimulus-related.
• The surge in residential investment reflects increased construction and sales of single-family homes (sales generate GDP via brokers' commissions). There will be a temporary setback, at least, given the scheduled expiry of the new homebuyer tax credit at the end of November. Even if the credit is extended, as seems likely, its impact will fade.
• Nonresidential structures spending continued to decline, and still has a long way to fall.
• The increase in business equipment spending was very tentative (just 1.1% at an annual rate), and fell short of our expectations.
• Growth was heavily influenced by a surge in vehicle output. If we look at GDP from the production side, the contribution of vehicles becomes very clear. Real GDP excluding motor vehicle output rose 1.9%, meaning that rising motor vehicle production added 1.7 percentage points to GDP. We will not see similar contributions from vehicles in future—but this is not just a "cash-for-clunkers" effect that will reverse in the fourth quarter. Most of the increase in vehicle output would have happened even without cash for clunkers, since the level of production had fallen so far that inventories were being depleted rapidly in any case. On average, vehicle output will be higher in the fourth quarter than the third, so vehicles will again add to GDP.

Conclusions

• The GDP report does not dramatically change our view of the outlook for coming quarters. On the positive side, the inventory adjustment was less than we expected, which means more of an inventory boost to come in subsequent quarters. And consumer spending excluding vehicles is entering the fourth quarter with more momentum than we had anticipated.
• But there is less upward momentum in business equipment spending than we had expected, there may be some payback due in defense spending after two sharp increases in a row, and the initial strong momentum in residential investment generated by the tax credit will weaken. And even though we do not see a payback in vehicle output, it will not be adding to GDP as strongly as it did in the third quarter.
• We have been projecting fourth-quarter GDP growth at 2.6% and first-quarter growth at 1.8%, and anticipate that 2.5–3.0% for the fourth quarter and 1.75–2.0% for the first remain the most likely ranges.