Rethinking the US Economic Outlook
By Joseph G. Carson, Senior Vice President and Director—Global Economic Research
While we still expect real US GDP growth to average 1% to 2% in each of the next several quarters, a spate of recent data poses a big challenge to this view. At a minimum, the new data indicate that the economic slowdown might prove to be a lot bumpier and somewhat weaker than we had originally thought.
One surprise in the third-quarter real GDP data was a 25.7% annualized gain in motor-vehicle output, driven primarily by a huge increase in truck production. The gain was suspicious because other measures showed a huge decline. According to the Federal Reserve’s industrial-production data, motor-vehicle assemblies contracted at an annual rate of 24%, while the Census Bureau’s manufacturing shipments and inventory data showed a 22% annualized contraction.
Based on our analysis of the GDP data, there was no procedure or measurement error. The Bureau of Economic Analysis uses a different methodology to estimate the real-dollar value of motor-vehicle output, taking into consideration not only the number of motor vehicles assembled in the period, but also the value added at the wholesale and retail levels. Moreover, our research found that over time, while all three measures of output move in the same direction, sizeable differences can occur from quarter to quarter.
Nonetheless, the gap in the third-quarter output measures was twice as big as anything we’ve seen in the past decade, indicating that the correction in the GDP motor-vehicle output could be relatively large in the fourth quarter. In fact, in looking at fourth-quarter light-duty truck production schedules, we estimate that the GDP measure of real truck output could decline as much as $50 billion annualized for the quarter—subtracting as much as 1.5 percentage points from the fourth-quarter growth rate. That seems large. But given that motor-vehicle output added a 0.7 percentage point to the third quarter (when other measures indicated it should have subtracted from growth), in our view the adjustment this quarter will have to be relatively large to realign the GDP output measure with those of the Federal Reserve and the Census Bureau.
While our fourth-quarter GDP estimate of 1.4% annualized assumed additional weakness in motor-vehicle output, we now think this weakness will be much greater. Assuming that we are close to the mark in estimating the other components, odds are relatively high that fourth-quarter GDP growth could be less than 1%.
Meanwhile, the October survey of manufacturing purchasing agents was soft, creating additional concern about the near-term path of the economy. The manufacturing index fell 1.7 points to 51.2, the weakest reading since the spring of 2003. Importantly, the fall was driven by a decline in the new-orders and production indices. At the same time, inventories rose—not a good development after the huge inventory build seen in the third-quarter GDP numbers. Purchasing agents are also surveyed about their customer inventory holdings. This series rose to its highest level since the last recession, which probably implies that there will be additional weakness in orders and in manufacturing generally in the months ahead as key customers work off excess inventory.
The spread between the new-orders and inventory indices has been an important indicator of the future direction of the manufacturing sector. At this time, the narrow gap between orders and inventories points to more weakness—i.e., additional production cuts—that could also weigh on GDP growth in the fourth quarter or beyond. The inventory adjustment plays a very important role in the length and depth of an economic slowdown. As a rule, during slowdowns the rate of inventory accumulation drops to nearly zero or slightly below for several quarters. In longer and deeper economic corrections, inventory liquidation can last a full year. We expect an economic slowdown in which the pace of inventory building drops from its third-quarter rate of $50 billion annualized to close to zero over the next few quarters—taking a slice from overall GDP growth.
Lastly, the Census Bureau released third-quarter data on housing inventory showing a further large rise in the number of vacant homes for sale. Indeed, during the quarter, an additional 206,000 vacant homes went on the market, bringing the total to a record 1.935 million. Vacant supply now equals at 2.6%, of occupied (or owned) housing stock, more than a full percentage point above its long-run average. It is not fully clear whether this vacant inventory represents speculative homes coming back on the market, individuals selling second homes, or recent purchasers having not yet put their existing homes up for sale.
Whatever the situation, it seems clear that the inventory overhang is still growing—a factor that will likely lengthen the housing correction and perhaps deepen it.
If this happens, we think the Fed would entertain the thought of lowering official rates in early 2007.
For an illustrated discussion with charts and graphs, please download the pdf at the top of the page.
alliancebernstein.com |
