“Visible shivers running down my spine.” – Elvis Costello, Watching the Detectives
Two months to go and the stock market is up over 20% with nothing to worry about, because there’s always next year. Such was the case in 2009, 2010, 2011, 2012 – the economy was about to accelerate the following year. That’s the prediction for 2014, too: nearly five years into the expansion leg, the recovery is going to pick up steam. No, we really mean it this time.
On the nearer horizon, a sign that stocks seem to be readying the general retreat I’ve been writing about was the price action of last week – the usual FOMC meeting rally was spent almost immediately after the market open on Wednesday, and the first-day-of-the-month rally on Friday was a tepid affair, despite the ISM manufacturing reading coming in above consensus (see below).
On the other hand, a sign that the general stock market has been getting into a bubble overall is the growing number of assertions that stocks are not in a bubble. The is-there-or-isn’t-there bubbly dispute is getting to be one of the dominant threads of conversation in the media of late. If it’s any comfort, the argument about whether or not there is one generally precedes the bursting of the bubble by not a few days or weeks, but by some number of months – anywhere from six to eighteen, as a rule.
But aren’t earnings are at record levels? Yes, and they will be until the next recession begins, because it would be quite unusual for earnings to decline before such a time. But with about two-thirds of the S&P 500 reporting, the composite earnings growth rate is at 4.2%, according to Zacks Investment Research, on revenue growth of 2.4% – just barely above the rate of inflation.
That leaves the S&P prices at a not-at-all inexpensive 19.4 times, as opposed to the lower “forward” earnings multiple that is mostly fantasy – the fourth quarter estimates that were for 10.4% growth a couple of weeks ago are now down to 8.4% and, guess what? They’ll be down to about 2% by the time the season begins in mid-January. After all, we need to guarantee another cinema of surprise when by golly, most companies are beating estimates. As to the price-to-sales ratio, that is now at a rather lofty 1.59, the highest it’s been since January of 2001. The all-time record is 1.77 in December of 2000, which could be eclipsed if stocks manage not to fade much before the traditional December rally (not even the crash of 2008 could stop that one).
Many seem to have become convinced that neither the earnings nor the economy matter anymore, that as long as the Fed maintains its QE policy, stocks are predestined to rise. I happen to disagree – earnings and the economy do matter, with the latter being like gravity – one may escape its pull for a long time, but it tends to win out in the end.
I do see some sort of cooling off in the stock market’s immediate future; I don’t yet see the end of the current bull, though as we move higher the market gets more vulnerable to upset. It could end in January, or it might be next March, even the end of next year, but we are way ahead of earnings and way ahead of valuations. If they keep marching higher, there’s going to be a terrific payback.
Boston Red Sox fans don’t have to wait for next year – congratulations to the team on their unexpected and most cool championship. Finally, a reminder to all that the U.S. turns back the clocks the morning of Sunday, November 3rd.
The Economic Beat
There were several major releases last week, beginning with the delayed September retail sales report and ending with the ISM manufacturing report, the Fed’s policy statement and ADP payrolls coming in between.
The retail sales report showed a decline of (-0.1%), pulled down by auto sales; excluding those, they were up 0.4%. That result looks set to reverse in October, as Friday’s auto sales reports for October (which admittedly don’t always line up with the Commerce Department estimate) were coming in with decent numbers, while weekly chain store sales data was indicating a down month for the rest of the sector. Year-on-year sales are tentatively up 3.2% for the month, though the quarterly result improved to about +5%, thanks to a strong July comparison.
The ISM manufacturing index reported a result of 56.4 for October, a sound result, about the same as September (thanks to seasonal adjustments) and about a point ahead of consensus. It’s quite likely that new orders are getting a boost from the post-government shutdown, though it hasn’t really shown up yet in the survey. The growth-contraction score for industries was 14-4, also a good number, and while the responder comments were somewhat mixed, on balance they were positive.
The national report followed a big Chicago PMI number the day before of 65.9. So, manufacturing is roaring, right? The MNI people that now sell the report boasted that it was “evidence that the recovery is gaining traction.”
It’s a good number, to be sure, but neither the Chicago nor the ISM numbers are great leading indicators anymore, not of the stock market, nor of the economy. The last time the Chicago was up over 65 was the spring of 2011, and you’d have done well to fade the market at the time. The ISM reading for October 1999 was 57.2; a year later the market was flat, in the early innings of an historic meltdown. Had you followed a strategy of buying stocks after the index has been in the basement for a few months and selling them after a series of historically high readings, you’d have made out like a bandit the last 15 years..
Curiously the Markit Economics’ PMI reading, a separate, more recent measure, came in much lower at 51.8, less than the previous month. It’s supposed to cover more small businesses than the ISM, which is tilted towards large companies. The Dallas Fed also reported a lower number for its region in October, 3.6 vs. 9.0, perhaps reflecting some weakness in the price of oil.
In any case, the surveys are diffusion numbers that are not well understood. The best way to understand them is as part of longer trends – one should look at ISM behavior for six to twelve months, rather than one or two, although abrupt changes should usually be respected.
The surveys measure breadth rather than volume – if everyone’s orders went up by $1 in a given month, the new orders index would be at 100 with no meaningful increase in business. If orders in two sectors are up a dollar and the third is down 100 dollars, the diffusion result is very good (not quite 66.6, depending on the adjustment, but still strongly positive) while the sum economic impact is contraction. The media tends to look at these numbers like basketball scores, saying manufacturing is growing “the fastest since” some date, but the truth is the results don’t tell us how fast it’s growing. One only has to look at the evidence of the September industrial production report, which showed manufacturing production increasing a very modest 0.1% in a month that featured an ISM result of 56.2 on top of an August reading of 55.7.
The headline number for total industrial production in September was deceptive, showing a strong increase of 0.6%, but it was dominated by the very lumpy category of utility production, which soared 4.4%. Over the last twelve months, the total index is up 3.2% with the manufacturing index up 2.6%.
The ADP report was soft, coming in at 130,000 versus expectations for something a little higher, with a downward revision to August to boot – 145K, down from 166K originally. The softness might have been of some help to the Fed’s decision not to do anything, though the FOMC committee minds were probably already made up. One aspect of the committee statement that seemed to matter to the bond market was the exclusion of any more reference to tightening financial market conditions; the yield on the ten-year bond has been rising ever since.
If the recovery is gaining traction, it isn’t evident in pricing. The producer price index (PPI) declined 0.1% in September and is up only 0.3% over the last year, while the consumer index (CPI) was up 0.2% for the month and 1.2% the last twelve months (1.7% excluding food and energy). Import-export prices declined on balance over the same period.
Home prices were up though, perhaps explaining why both existing and pending home sales were down. According to the Case-Shiller people, prices rose 12.8% over the year ending in August. Pending home sales fell 5.6% on top of a decline in the existing sales rate last month. As I’ve written previously, there seems to be something of a stand-off developing between buyers and sellers. Pending applications were probably also hurt by a lack of access to the IRS tax-form data that is now virtually obligatory for a home loan.
Consumer confidence was down sharply, as measured by the Conference Board and Bloomberg, but these sentiment swings have little correlation with spending.
Next week will feature the delayed October jobs report on Friday, making it a nine-day lag from the ADP report instead of the customary two. The Labor Department’s report includes a preliminary benchmark revision (in effect, re-estimating everything since at least February), so there could be some fireworks. It will be accompanied by September personal income and spending (delayed), but that will be mostly known from the first estimate of third-quarter GDP the day before. Those two reports should heavily tilt the price action towards the back end of the week.
Factory orders come out Monday, but is less apt to surprise after the release of the durable goods report. The ISM non-manufacturing report comes out Tuesday, but despite being two-thirds of the economy doesn’t get as much respect as the manufacturing report unless it’s a big surprise. A delayed leading-indicators report comes out Wednesday, along with the Challenger layoffs report, but the week should belong to GDP and employment.