First on industrial production, which was generally regarded as a good number:
A seemingly positive economic report was the July industrial production report, which blew away consensus expectations, coming in at 1.0% MoM versus 0.5% expected. Production in the manufacturing sector rose 1.1% on a spectacular 9.9% jump in motor vehicle production.
Here’s why we called it a “seemingly” positive report. Remember that July has been a strange month for seasonal factors, especially related to the auto shutdowns (GM kept most of its plants open this year and the government’s seasonal-adjustment models don’t account for this). We saw wild swings in the weekly initial jobless claims partly due to seasonal-factor adjustments (claims dropped by 10% in the first two weeks only to increase by 10% in the third week of July, to give you a sense of the volatility).
We think something similar was going on with the seasonally-adjusted motor vehicle production numbers — in other words, the seasonal factors may have artificially boosted the data. In fact, when we adjust the seasonal factors, total IP would have been 0.4%, according to our calculations, not the 1.0% reported. We could see a significant pullback in August as part of the July increase is reversed. On top of this, the back revisions to the actual data were negative — the +0.1% gain in June was turned into a -0.1%. So the underlying increase over June and July would be in the 0.2% range, hardly inspiring.
And as for housing starts, which were not so hot:
Yesterday’s U.S. housing starts report was disappointing, especially in the wake of the soft NAHB report on Monday. Total starts rose 1.7% MoM in July, slightly missing the 2.0% increase expected by the consensus. That was probably the best news of the report.
Downward revisions to the June data were harsh with the -5.0% initial estimate being taken down to -8.7%. Another way of putting it, if it weren’t for the downward revisions, the headline would have been negative.
The increase in total starts was driven by a huge jump in the very volatile multi- family component — up 33% after a 33% decline in June. A better gauge for underlying demand is single-family housing starts, which dropped 4.2%, the third monthly decline in a row, falling to the lowest since May 2009. The YoY rate fell to -14%, the second month in negative territory and the weakest print since August 2009 — it really doesn’t seem like things are getting any better. To put it into perspective, single-family starts has been essentially range-bound since the end of 2008, languishing near record lows— even after all the government stimulus.
Forward-looking components of the report were weak as well, with building permits falling 3.1%, the third decline in four months. Note that total permits go into the index of leading economic indicators and our tally so far suggests a flat July monthly print for LEI, at best.
Again, single-family permits were soft, down 1.2%, the fourth consecutive monthly decline — pointing to more weakness in starts in the coming months. Multi-family permits fell 8% suggesting that the 33% increase in multi-starts will be partially reversed in coming months.
Join us on our
Share this page with your friends
on your favorite social network: