Via Jeff Snider of Alhambra Investment Partners,
I don't think there was much in the GDP report that wasn't expected, except durable goods. The decline in durable goods was comparable to Q2 2011, right down to the primary driver of that weakness - motor vehicles. However, there was no earthquake in Japan this year to disrupt supply chains, production schedules and brand availability. Just like last year, marginal economic growth overall seems to be backfilled with a tide of inventory. The trouble with inventory at the margins of growth is that it is essentially a build-up of forward demand, and therefore susceptible to reversal should overdone production move out of alignment with final demand. Both monetary and fiscal policies actively seek to pull forward demand, meaning this inventory-driven activity conforms to policy goals. This is true in perhaps more ways than expected since the heavy arm of fiscal activism is readily apparent in an important sector of the marginal manufacturing economy.
That brings up a bit of a curiosity since durable goods "sales" of motor vehicles should be largely in sync with durable goods "output" of motor vehicles:
"Sales of MV", contributions to GDP past 5 quarters -.53 .05 .63 .31 -.29
"Output of MV", contributions to GDP past 5 quarters .05 .03 .55 .72 .13
That's a pretty big deviation in the past two quarters. The “sales” figures include the sales of motor vehicle parts, while the output figures do not, but I highly doubt the marginal difference is solely a collapse in parts purchases. It might be a statistical anomaly that gets revised away but it might also explain some of the recent weakness of the regional Fed surveys and the ISM manufacturing survey.
Clearly weakness in Q2 2012 PCE is attributed to the change in the direction of durable goods (from +13.9% and +11.5% in the preceding quarters to -1% in Q2 2012); durable goods added .85 to GDP in Q1, but subtracted .08 in Q2. It is also pretty clear that durable goods weakness itself can be mostly attributed to weak motor vehicle sales (and parts). Spending on nondurable goods was largely unchanged from the preceding quarter, while spending on personal services was slightly higher (due largely to the economically “beneficial” pickup in spending on “housing and utilities”).
MV sales are highly correlated to income growth, but there is also a credit availability component. I haven’t seen anything recently in the consumer credit numbers that suggests there are any financing-related issues. Is this perhaps a price issue? Both sales and inventory figures in the GDP release are reported in dollar terms, so it is possible this is a function of price markdowns after leaving the factory. That still ends with the same result, however, slowing future production in response to an imbalance of supply and demand.
According to Briefing.com in February of this year, inventory levels are still below historical run rates for the domestic manufactures, though they have risen steadily at GM and Chrysler (as pointed out on ZeroHedge http://www.zerohedge.com/news/class-action-lawsuit-filed-against-gm-channel-stuffing). Ford, on the other hand, has actually trimmed its inventory/sales ratio since 2009, perhaps suggesting that inventories are being more robustly managed to concur with a far different environment (and within the industry, inventories are obviously being managed quite differently with regard to government involvement). Total car sales in the US peaked at over 17 million units, and even optimistic forecasts call for, at best, near 14 million units in 2012.
This below peak buying rate has been used to build the optimistic case for auto demand since the overall age of the domestic car and truck fleet has increased from about 10.5 years in 2009 to about 11.5 years. However, that trend toward “aging” vehicles has been in place for more than a decade – average fleet age in 2002 was 9 years. If anything, the trend in the aging of the fleet has actually been relatively consistent since 2005 (with light trucks being held onto more than cars). Since this trend pre-dates the Great Recession it should not be a surprise that turnover is reduced, meaning that sales growth marginally depends more on households buying additional units. That’s an expensive proposition in the post-crisis period, especially in comparison to the housing bubble that preceded it.
In 2007, the number of motor vehicles per licensed driver, according to Briefing.com, was about 1.21. That rate of usage has fallen off to about 1.17 since. For a frame of reference, in 2003, just as the housing bubble was swinging into full gear, there were about 1.15 motor vehicles per licensed driver.
Optimists see pent up demand in these numbers, but lack of per capita income growth may be keeping a lid on that potential demand. In my opinion, it is likely that the amount of cars and trucks produced and sold in 2007 was another artifact of monetarism that cannot be repeated in the current environment. After all, if there is so much pent up demand, then the contraction of motor vehicle sales in the GDP accounts is an empirical result that runs perfectly contrary to that optimistic expectation. Given that there does not seem to be a contemporary problem with available financing (thanks in large part to Ally Financial, formerly GMAC and still sporting that hefty federal government investment) that may mean the industry as it existed pre-crisis is itself an anachronism, and perhaps much more sensitive to changes in real economic variables at the margins. The big question for the economy is whether inventory and production levels are as sensitive.
Obviously, for the overall GDP accounts, the difference between sales and output is inventory.
Demand for durable goods and motor vehicles join a downward trend in the wider economy, showing up in gross domestic purchases and real final sales, and highlighting the disconnect with inventory:
A diverging trend between final sales and overall inventories, especially as they both intersect through the sensitive motor vehicle sector, is not an indication of future strength to me. If the trend in the first chart is valid, we should ultimately expect the second chart to follow. Maybe the economy is devolving faster than previously thought in the aftermath of financial dysfunction and the largely disabled/ineffective wealth effect. We cannot ignore, however, that the intent behind increasing production may not simply be a function of pure market forces that will be solely respond to demand. After all, GM proclaims it’s in a record-setting run despite the noticeable increase in inventory (in sharp contrast to Ford inventories), in what just might be a shadow stimulus. How long can this trend continue before production overruns prove too costly? Maybe GM’s stock price is an indication.
It's almost like the 1960's and 70's, with motor vehicles and government spending driving the marginal economy again. All that’s missing is for Ralph Nader to show up and write about how cars are dangerous.