Avalon's MarketWeek

A Curious Incident


Inspector Gregory: “The dog did nothing in the night-time.”
Sherlock Holmes: “That is the curious incident.”

- Sir Arthur Conan Doyle, Silver Blaze

To use the vernacular, what was that all about? After the release of a distinctly run-of-the-mill jobs report for January on Friday, the stock market cratered again, producing yet another dramatic day of losses in the 2% range.

With such events, the usual thing to do is to blame the data. Deeper analysis usually consists of rounding up some quotes from Street strategists and economists about one or two catchy details. The general line of thinking was that as the number was well below consensus (true), it must have reignited fears about global growth.

There are plenty of grounds for worry when it comes to global growth, but none of them were in the January 2016 jobs report. In that respect, it resembles the dog of Holmes’s famous remark – it did nothing at all, and that is the curious part.

In 2013, the first estimate for January job creation was 157,000 (seasonally adjusted, or SA), compared with the initial estimate of 151,000 for January 2016. In 2014 the estimate fell to 113,000 (113K), then rose to 257K last January. That puts January 2016 in the middle of the pack over recent years. It is very middle of the pack, in fact, as the initial estimate of the year-on-year gain in the unadjusted January data is 1.89%, compared with the trailing five-year average of – wait for it – 1.85%.

So yes, the number was below a consensus guess of 190K-200K (depending on the reporter), but keep in mind that pundits were predicting the day before that such a shortfall might be good for stocks, as it could convince the Fed that further tightening was off the table for the near future. Obviously, that reaction wasn’t what ensued.

Why not? The report (about which more below) was very much in line with recent January reports, with the unemployment rate ticking down another notch to a quite low 4.9% and hourly earnings growth maintaining an above-trend clip of 2.5%. There are certainly some flaws in the picture, but just about every jobs report of the last five years has had flaws. I have been writing frequently of late about the end of the current business cycle, yet the January jobs report had no such evidence – except, perhaps, to those who are reading doom into every passing event. The initial estimate may not have been a gangbusters number, to be sure, but it was very much an average report.

So why the sell-off? One good explanation would be that we have entered a bear market, or at least the mentality of one. Up until about the middle of last year, one of the leading market characteristics of the current cycle had been the jobs report rally. Regardless of what it said, stocks rallied. If the number was above consensus, it was repeated that the economy was reaching escape velocity, while below consensus would mean the Fed was on hold. It was all good.

But for some months now, the stock market has been consistently selling off on the jobs report and the narrative has turned: above consensus and the Fed will tighten, below consensus and the economy is disintegrating. That kind of thinking is bear-market behavior. The jobs number was fine, it was typical, but it missed consensus, and just as companies get slaughtered for missing estimates these days, the stock market did too.

Looking ahead, the market is getting very oversold on an intermediate basis and remains oversold on a short-term basis. The latter isn’t quite severe enough to step in yet for a rebound trade, but it’s reasonable to expect some breathing room soon, though at what level I couldn’t say. Five weeks ago you could have looked all day in vain for anyone to say that either a bear market or a recession might be at hand, but now five weeks of selling has those words on many lips these days. CNBC has nearly acquiesced to the former and many worry openly about the latter. Some relief may come from Fed chair Janet Yellen’s midweek testimony to Congress, but I worry about the January retail sales report on Friday, as the weekly reports were not encouraging. It’s going to be a rough year – whether the dog barks or not (the basis of Holmes’s fictional remark), you can count on hearing more from the bear.

The Economic Beat

So the report of the week was the jobs report, and as noted above, it really wasn’t anything out of the ordinary. The headline number of 151K is just a seasonally adjusted (SA) estimate, of course, as the payroll count doesn’t actually go up in January. We lose about 2% of the insured workforce (the source of the payroll estimate) to separations every January – in a bad year, a little more, in a good year, a little less. The real strength of the market is how many months it takes for the 2% to be back on the official roles again. It usually happens by April or May.

This year’s churn is 2.07% (that is, the unadjusted January total is 2.07% lower than the December estimate), not as good as last year’s 1.99% but in line with the trailing five-year average of 2.06%. Those totals are likely to change somewhat upon revision, but not dramatically so. As noted above, the year-on-year January comparison shows an improvement of 1.89%, versus the five-year average of 1.85%.

Regular readers know that I have been saying for some time now that the business cycle is ending and that we are in recession territory. I don’t believe we are quite there yet, but one may be starting this quarter (it will take years to find out). However, don’t look at the current jobs report for evidence. The unemployment rate ticked down, the household survey showed a gain of 615K jobs, and while one may reasonably argue with the seasonal adjustments, the printed totals of 40K for new goods-producing jobs and 29K for new manufacturing jobs aren’t indicative of a slowdown. We didn’t really add them, just lost fewer (at first glance) than the expected January churn, but for now those numbers are fine. The ADP payroll report beat consensus with a 205K estimate versus consensus for 190K, so maybe expectations got a little high. That said, a 210K print would probably have rattled the markets just as much, sparking the conviction that the Fed would inevitably raise rates in March. It’s been that kind of market.

Some pointed to the December revision as evidence of a bad report, because the total was revised down by 30K – and pushed into November, which gained 28K. Such adjustments are typical at year-end and the revisions should be treated as a wash until proven otherwise – unless, of course, you are already living in deep anxiety and every miss is The Big One. We will almost certainly get the big one this year, so far as jobs are concerned, but it wasn’t in January, not by any means but the tape.

The weak got off to a rough start with an ISM manufacturing number that came in on the light side again, this time with a below-neutral reading of 48.2, the fourth such reading in a row (50 is neutral). Not good. The growth vs. contraction sector score actually improved a bit to 8 growth vs. 10 contraction (December was 6 to 12). New orders did claw back to neutral (50.2), but the markets were no doubt hoping for better after last week’s Chicago PMI. The ISM’s non-manufacturing sibling disappointed as well two days later with a reading of 53.5, down from 55.8 in December and below expectations for about the same reading. The growth-contraction score was an unrobust ten vs. eight, and the business activity index actually fell sharply, from 59.5 in December to 53.9. That’s a big move for these numbers. Prices, which are the best leading indicator in the service sector (forget new orders), also suffered a significant drop, from 51.0 to 46.4. Frankly, this was the most disappointing report of the week, though it got little airplay.

Construction spending fell by 0.6% in November – maybe – and might have risen by 0.1% in December. I don’t trust the first estimates for this series at all, so take them with a grain of salt. Factory orders were revised down to (-0.7%) in November after an initial estimate of (-0.2%), and then fell some more in December, (-2.9%) by the initial estimate. It was not a pretty report, save that the business cap-ex categories got tiny revisions higher. New orders fell an estimated 6.6% in 2015. It was weak both ex-defense (-6.7%) and ex-transportation (-6.8%), but it wasn’t recession across-the-board weakness, as some categories did fare better (e.g., electronic component orders were up a whopping 26.8%).

Exports fell in December and imports rose, according to the trade report. Rising import demand is actually a good sign of domestic demand, but that’s a number that will push down on the next estimate of Q4 GDP. Personal income was up a healthy 0.3% in December, but the spending estimate was for no change. At least November spending got a bump up, from 0.3% to 0.5%. The categories are not leading indicators, as they tend to follow employment.

Next week will start off with the labor market conditions index, a not-widely followed number that the Fed is supposed to use (though nobody is really sure what the Fed uses anymore). It’ll be followed by the labor turnover report (JOLTS) the next day, though I will be paying more attention to the more-sensitive wholesale trade report. The big report is retail sales on Friday, and there the market is looking for a gain of 0.2%. If they don’t get it, there could be hell to pay. Janet Yellen’s testimony before Congress on Wednesday and Thursday should dominate the week, but time will tell. Import-export prices on Friday round out the week, along with business inventories and a consumer sentiment reading later that morning.

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