The Luau of the Lemmings


“There will be a great confusion as to where things really are.” – Monty Python’s The Life of Brian

Goofy action in the stock market is nothing new, but last week’s luau was something special. Monday was normal enough – the last day of November often sees some selling – but the rest of the week was a pack of lemmings running from one side of the boat to the other (or if you like, running from one cliff to the next). While the first trading day of the month usually has a mildly positive bias, there was no good explanation for the scope of Tuesday’s rally other than the fear of being left behind if the market makes a big December move. January brings the annual obsession with calendar-year returns, and managers are right to fear the wrath of fickle clients using that metric in search of the best deal. With the broad indices nearly flat on the year in 2015, performance anxiety is going to be with us the rest of the month, so expect more volatility as the constant running back and forth amplifies trading moves. There isn’t much investing being done this month.

What the market is really lacking these days is a good story for going higher. The last two quarters have seen negative earnings growth for the S&P 500, and estimates for the fourth quarter are already well into the red. Yet the declines have been mild – neither sharp nor broad-based, they aren’t the kind of thing that provokes widespread selling. Instead we are told to ignore a few errant sectors and concentrate on better times ahead that will redeem the current lofty valuation. It’s a tune they play at the end of every cycle.

Lacking earnings growth, the natural alternative (Wall Street natural, that is) is to obsess even more about central bank policy. Most of the Street suffers from the delusion that the business cycle is controlled by the Fed, so they are apprehensive about what the long-delayed tiny bump to rates (to a range of 25-50 basis points) might mean. Some correctly point out that on balance the markets have usually risen for many months after an initial rate increase. True, but the Fed doesn’t usually wait until the end of the cycle to begin raising rates. Others point out (also correctly) that a range of 0.25%-0.50% cost of Federal funds should hardly be an impediment to lending. Besides, isn’t European Central Bank President Mario Draghi still ready to do “whatever it takes?” Just don’t ask any impertinent questions about how that affects our own economy.

An increase (or lack thereof) in rates will matter for short-term stock price action, but not for the business cycle. It’s going to end anyway, and is highly unlikely to end like the last two, with a massive financial crisis or a tech bubble. It will have one key characteristic in common, though – a highly valued stock market. At the ends of cycles, highly-valued stock markets tend to go sideways for long episodes, with lots of volatility as earnings growth flattens, then disappears. You know, kind of like 2015.

The November jobs report sparked quite a relief rally that was itself a warning – beware of big rallies with thin volume and poor breadth. The result of 211,000 new jobs (about which much more below) was slightly ahead of consensus, but the markets had come to fear something worse. Many were positively rejoicing over a number that was only half that of November 2014 (423,000). How quickly they forget. The weak earnings numbers and the destruction in commodity prices have nonetheless fanned fears about economic growth, and not all are ready to believe the usual Wall Street assurance that earnings growth will unquestionably be better next year, sir.

Are the financial markets being complacent about terrorism? Perhaps. Like many others, I myself worry about more attempts as we approach the holidays, and like most others, have nothing else to go on but anxiety and the recent events. As I’ve written before, the stock market is not a place where one should look for clues about the import of political events. They really don’t know any more than you do and are just as subject to ideological blind spots as the next person. All I can tell you is that there is little fear of a major event in current prices, and that could prove to be a problem. Hopefully, it won’t turn out that way.

The Economic Beat

It’s hard to say what the highlight of the week was – one could make a case for the employment report, the ISM manufacturing report, or Yellen’s appearance before Congress.

If you’re an optimist, you would probably pick the employment report. Though the jobs number of 211,000 (seasonally adjusted) was neither unusual nor a surprise – consensus had been in the 190K-200K range, and the ADP report two days earlier had predicted 217K – it seemed to come as quite a relief to equity traders and a green light to equity trading AI programs. A very good example can be found in the Econoday website comments, which followed up its lead-in sentence description of payroll growth as “solid” with a switch to “very solid” (emphasis added) in the very next sentence. Given that the number is actually about 10% below the 12-month average of 237K, one might wonder where the “very solid” enthusiasm was coming from – unless you were worried about something much worse. The November 2015 result is only half the size of the November 2014 report (423K).

The reaction was without a doubt anxiety-induced, with a large measure of the blame going to the ISM manufacturing survey released on Tuesday. The survey index of 48.6 was the lowest reading since June of 2009. Note that I didn’t write the “weakest” reading, since the survey is a diffusion measure and the current level of activity is well above the level it was six and a half years ago. But the short-term trend is unmistakable. The weakness in the number was signaled by last month’s report, which skirted the neutral line of 50 with a reading of 50.1 yet reported a weak composite score of seven sectors growing versus nine contracting, with two unchanged. The composite worsened in the November survey to only five growing versus ten contracting (three unchanged), with new orders also moving into mild contraction (48.9).

The ISM report led a brokerage house to note that such readings had historically been precursors to recession about two-thirds of the time, a remark that actually helped lace up the equity market buying shoes on the bromide that the Fed “never” raises rates with a manufacturing ISM below 50 – a hope that was clearly dashed by Yellen’s testimony and speeches over the ensuing two days (she was clearly leaning towards a hike). Indeed, the CEO of US Steel (X) was on CNBC Friday heartily endorsing the notion of an “industrial” recession. In that vein, the Dallas regional survey was negative for the 11th month in a row and the Chicago private-sector purchasing survey went negative (below 50) again at 48.7, ending its one-month winning streak.

Yet the manufacturing data got the usual “it doesn’t really matter this time because” treatment that unwelcome numbers get near the end of business cycles, usually with an explanation that the damage is limited to one or two narrow sectors – as if the entire economy ever rolls over in a single month. Keep in mind the bit about two-thirds of the time, however – ISM manufacturing is indeed not an infallible indicator, and has in the past had weak episodes that were not lead-ups to recession. A good reason to worry this time is that the false readings have usually been pauses occurring mid-way through the business cycle, a claim that cannot be made this time, not with the last cycle ending eight years ago.

Employment is not only a lagging indicator, the BLS understanding of changes in hiring tend to lag as well as its statisticians ponder changes in limited samples. It would be unusual behavior indeed for hiring to collapse at this time of the year from the full-employment phase of the cycle that we occupy now (note again, however, the difference from November 2014). December might be different, but we’ll have to wait for clues from the claims data. Jobs growth continues to feature lower-paying service jobs that have kept a lid on broader hourly earnings growth – in November it fell back to only 0.2% and the year-on-year gain to 2.3%. Indeed, the number of involuntary part-timers rose by 319K in the household survey. The household survey is a small-sample report with a lot more volatility in monthly numbers, but the number is nevertheless a good indication of a substantial increase, perhaps spread out over more than one month. Trouble in the employment data usually starts slowly and I don’t expect any definitive changes to begin before 2016, but the growth in the last few months is definitely off from a year ago. Jobless claims remain in their current peak trend and should remain so until the turn of the year.

There wasn’t much remarkable elsewhere in the jobs report. October and September were revised higher, though an earlier downward revision meant that September ended up about where it started, at a less-than robust 145K (originally 142K). September might be a better indicator than the holiday-related surge in year-end hiring (cf. November 2014), but time will tell. The average workweek edged down a tick, which would appear to substantiate the household survey’s increase in part-time workers; the participation rate edged up a tick to 62.5, still lower than the year-ago 62.9 that explains at least part of the decrease in the unemployment rate from 5.8% a year ago to the current 5.0%. Rounded down, it was the same as last month and nearly the same as September (5.1%) I would not expect this rate to change before the January report, nor would I be surprised to see a four-handle (e.g., 4.9%) for December.

Trouble in ISM manufacturing? Not to worry, say the apologists, manufacturing is smaller now and we’re really a service economy now. There is no disputing the first point, as manufacturing employment is less than half of what it was 20 years ago – when we were also a service economy. The manufacturing number owes its elevated profile to its status as a more-sensitive cyclical indicator, rather than as a guide to the next GDP number (currently tracking quite low for now, according to the Atlanta Fed).

Alas, there was a noticeable decline in the non-manufacturing survey as well, from 59.1 to 55.9. The composite score was a still-reasonable 12-6 growth vs. contraction, yet one that is not in the range of stronger reports that typically show two sectors or less in contraction. Participant comments in the report certainly showed an uptick in anxiety, and a drop in the employment measure (still only a diffusion number) from 59.2 to 55 probably fueled some of the fears about a big miss in the jobs report. Prices, however, were sideways and they are the best leading indicator for the services report. Like its manufacturing brother, one usually needs to see several months of similar behavior in the surveys to have confidence about a change in the underlying trend.

“Very solid” also made its way into Econoday’s characterization of the October factory orders report, which showed a 1.5% increase in new orders (seasonally adjusted), though a less-robust 0.2% when excluding transportation (the numbers were inflated by air-show orders, a category where shipments often come up well short of the original order). Very solid indeed: total shipments declined again, putting downward pressure on GDP, and while new orders for business capital goods did stage a 1.3% rebound, they were negative on a year-ago basis for the tenth month in a row. Year-to-date, new orders are down overall by a very solid 7.1% (unadjusted).

Putting further pressure on GDP for both the current and last quarters was the report on international trade for October. A negative September revision will mean a subtraction for the last quarter, while a larger-than-expected deficit for October will lower estimates for the current quarter. Both exports and imports have been declining all year, with exports off 4.3% year-to-date and imports down 2.6%.

The warm October that hurt apparel sales helped construction spending, up a provisional 1% for the month and up 13% year-on-year. It didn’t seem to help pending home sales, which were up a much smaller than expected 0.2%. Both numbers are subject to substantial revisions.

Looking ahead, the highlight of next week should be the November retail sales report. Despite much of the anxiety about indifferent Black Friday sales, I do expect a decent holiday quarter, given the low unemployment rate. Retail sales growth has been weakening all year long, but the fourth quarter should see a last-gasp rebound (much like the ones in 2000 and 2007). The labor market checks in with its turnover report (JOLTS) on Tuesday, and I have been highlighting the fact that while job openings are up substantially on a year-ago basis, hires have been down, as they were in the November jobs report. Inventory data throughout the week will tell us something about its start to fourth-quarter GDP – the contribution is not expected to be positive – and another report on declining import-export prices is due on Thursday.

share this blog post
Facebook Twitter Plusone Linkedin Digg Delicious Reddit Stumbleupon Tumblr Posterous