Waiting for Q4 GDP

January 26, 2006 – It looks like growth in real GDP (coming Friday) slowed to a 2% to 2.5% annual rate in Q4 following a 4.1% gain in Q3 (Chart 1). Weaker Q4 consumption growth should account for virtually all of the GDP slowing. This consumer slowdown seems to be temporary – and narrowly based in the auto sector.

Chart 1. Percent change in real GDP. Q1 2000 through Q3 2005 and Q4 forecast.

Car and light truck sales rose by a strong 1.4M units to a 17.1M unit annual rate in December, after a big 1M unit November increase. However, light vehicle sales fell from August through October, in the wake of July’s huge incentive-driven spike to 20.7M.

Reflecting this earlier (post-July) weakness, Q4 light vehicle sales plunged by 2.1M to an average 15.8M units (Chart 2). The related decline in motor vehicle & parts spending could reduce real consumption growth by about 3 percentage points in Q4.

Chart 2. Light vehicle sales. Q1 2000 through Q4 2005.

In contrast to autos, consumer demand for non-auto goods seems to have strengthened in Q4. Real non-auto retail sales rebounded to a hefty 7.8% annual rate in Q4 after slowing to 2.3% in Q3 (Chart 3).

Chart 3. Percent change in real non-auto retail sales. Q1 2000 through Q4 2005.

Assuming that consumers’ real services purchases rose at a 2% to 2.5% annual rate in Q4 (in line with their October-November average), real consumption growth may have slowed to a mere 1% annual rate in Q4 after rising by 4.1% in Q3. This slowing in consumption growth would reduce Q4 real GDP growth by about 2 percentage points, compared to Q3.

Beyond consumption, real non-defense capital good shipments excluding aircraft rose at a brisk 15% annual rate in Q4 (+1.8% in Q3, Chart 4). This suggests that Q4 growth in real business equipment & software investment may have accelerated somewhat, compared to Q3’s already-strong 10.6% increase.

Chart 4. Real non-defense capital good shipments excluding aircraft and real business fixed investment in equipment and software. Percent change. Q1 2000 to date.

However, based on two months’ data, real construction spending (residential plus non-residential) seems to have slowed a bit in Q4. Thus, the guess here is that total real fixed investment growth roughly matched Q3’s 8% gain in Q4.

Data through November also suggest a decline of about $30B in Q4 real net exports (measured in chained 2000 dollars), partially offset by firmer (less negative) real inventory growth. In Q3, the real trade deficit was roughly flat at $617.5B and real inventories fell by $13.3B.

Real government spending (always a wild card) rose at a relatively strong 2.9% annual rate in Q3. This forecast assumes a comparable Q4 increase.

Suzanne Rizzo