The big story in today's GDP report is consumer spending—weaker than expected in the second quarter, and revised down in previous quarters. Consumers have tightened their belts more than previously estimated, taking the saving rate up to 6.2%. A weak employment market, high debt levels, a weak housing market, and tight credit are holding consumers back.

Consumer spending on services is now estimated to have fallen 0.4% over the four quarters from the first quarter of 2009 to the first quarter of 2010, instead of rising 0.9% as previously published. This is a major shift, considering that consumer services spending accounts for almost half of GDP. It is a reminder that initial estimates of spending on services are based on incomplete information and must be treated with caution.

Although second-quarter consumer spending only increased 1.6%, total final sales to domestic purchasers increased 4.1%, its strongest increase in more than four years. This represents all spending on final goods and services in the United States (it includes imports, and excludes exports and inventories). Unfortunately, this strong rise does not mark an acceleration in the recovery, since much of it just sucked in imports, and since some of it represents temporary factors.

The best news in the report was the second 20%-plus annualized increase in business equipment and software spending in a row (led by high tech). Businesses are much readier to spend on capital equipment and software than they are to rehire workers. The not-so-good news is that much of this extra equipment was imported. The increase in capital goods imports was roughly the same size as the increase in capital equipment and software spending.

There was also an increase in business structures spending, of 5.2% annualized, its first advance in two years. This was not a surprise, based on monthly indicators, and mostly reflected another strong increase in oil and gas drilling activity, although other spending appears to have roughly stabilized. We are doubtful, though, that business structures spending has hit bottom. Drilling activity will be hurt by the deep-water moratorium, and building construction indicators (like the architecture billings index) are still pointing down.

Residential structures were also a plus for growth, rising 27.9% in the second quarter, albeit from a low base. Unfortunately, this sector will now turn into a drag on growth again as we go through the home sales' and housing starts' paybacks for the homebuyers' tax credit that expired on April 30.

Government spending also added to growth, rising 4.4% in total on a bounce in defense spending, a surge in other federal spending (linked to the Census), and the first increase in a year in state and local government spending (probably driven by federal support for capital spending). Government spending will not necessarily decline in the third quarter, but at a minimum it will rise more slowly.

Outside domestic spending, we saw another increase in inventory accumulation. Inventories rose, and did so more rapidly than in the first quarter, so that inventories added 1.1 percentage points to GDP growth. Much of this was probably intentional restocking, but some was probably unwanted, resulting from disappointment in consumer spending. We cannot expect inventories to keep adding to growth in the second half of the year.

The boost from inventories was more than wiped out by a huge deterioration in net exports, which subtracted a massive 2.8 percentage points from growth. Exports rose 10.3%, similar to their first-quarter pace of 11.4%, but imports surged 28.8%. This is the biggest one-quarter increase in import volumes in 26 years. We must treat these initial estimates with some caution, though, because there is no trade data yet for June, so the quarterly estimates from the Bureau of Economic Analysis are based on assumptions for June (the BEA has assumed a big widening in the trade deficit).

What does seem clear is that much of the burst of inventory accumulation has been met by imports (which implies that as inventory accumulation stabilizes or perhaps eases back, so will imports). Additionally, as we have already noted, imports were pulled in by stronger business equipment spending.

Historical revisions to GDP in today's report stretch back to 2007, and they have reduced growth by an average of 0.2 percentage point per year. The biggest shift is in 2008 growth (now zero instead of 0.4%). The peak-to-trough decline in GDP during the recession is now 4.1%, instead of 3.8%, further underlining the status of this recession as the worst in the postwar era. For those interested in the minutiae, the peak GDP quarter is now the fourth quarter of 2007; the rise in GDP during the second quarter of 2008 is now insufficient to reverse its first-quarter drop. As a result, the peak GDP quarter now lines up with the official monthly cycle peak of December 2007, identified by the NBER's Business Cycle Dating Committee.

Of more immediate interest is the question of whether we have truly exited the recession or are about to dip back down. The growth pattern now shows a peak growth rate of 5.0% in the fourth quarter of 2009, but a loss of growth momentum since then as the boost from the inventory cycle has faded. Final sales growth (i.e., excluding inventories) is positive, but very sluggish.

The economy entered the second quarter with plenty of momentum, and left it with very little. The implication is that growth in the third quarter will be slower than in the second. In particular, residential construction will drop due to the payback for the homebuyers' tax credit, and government spending will be much less supportive as Census spending declines. A full reversal into a double-dip recession remains a possibility, but is still not our base case. Overall, 2010 now looks set to generate GDP growth of just below 3.0% for the calendar year, and we expect growth to stay in the 2.5–3.0% range during 2011.