Goodbye to July

“Sometime too hot the eye of heaven shines.” – William Shakespeare, Sonnets (16th)

Regular readers know that I have been writing all year that the business cycle is slowly running out of gas. In recent weeks I have re-emphasized that point (perhaps to the point of being boring) in the face of a stock market rally that was more than a little contrived. I closed last week’s market note with the sentence that we would soon see “astonishing bits of news that will show that gosh darn it, growth isn’t as good as the June/July choir made it out to be.”

The wait wasn’t long. Friday’s release of the first estimate for second-quarter GDP (+1.2%, seasonally adjusted and annualized, or SAAR) caught the market by surprise, though you wouldn’t guess it from the muted tape. Other reactions were varied – the Wall Street Journal blared the next day in page one banner headlines that the recovery was the slowest since 1949 (the Journal doesn’t usually throw that much water on the economy – maybe, just maybe the headline was written with an eye on the election?). Over on Wall Street, the reaction has largely been the very predictable kind of denial that makes people hate the investment business. It’s been mostly two-fold – on the one hand, beatific fund managers assuring us that this is really good news, since it keeps the Fed on hold; and on the other, a host of voices trying to figure out what is wrong with the way that gosh-darn GDP number is calculated, since it clearly can’t be so.

The Journal ran a sloppy note to that effect over the weekend, claiming that the GDP number is inconsistent with the number of jobs added this year. What the writer missed is that the rate of employment growth has slowed sharply this year: the rate of growth through the first six months of 2016 is the weakest such period since 2010, when the economy was in recovery mode. It isn’t inconsistent at all, really – employment is a lagging indicator, and so we expect it to lag growth in the early stages of the cycle and lag again near the end. The fact that both output and employment growth rates are now slowing together ought to give everyone something to think about. Gee, do you think the cycle might be getting old?

Well, of course it can’t be, because the stock market is at all-time highs, right? And the stock market knows what the economy is doing, doesn’t it. Like it did in October 2007, making its then all-time high a mere six weeks before the worst-ever post-war recession began. Or the way it did in 2000, peaking at utterly insane valuations (and an all-time high) six months before recession and one month before beginning an utterly brutal bear market. It just might be that “all-time highs” aren’t buy signals, but don’t expect to hear that from the Street. Expect instead to hear everybody repeating that employment is so much better than GDP, even though it isn’t, because, well because it sounds better.

With two-thirds of the S&P 500 reporting, the estimated year-on-year decline for the group’s quarterly earnings is at (-3.8%), implying an actual decline of somewhere around 2%. The mantra of “not as bad as it could have been” is in full swing, and indeed sales growth might have been positive for the second quarter, however marginally, thanks to the energy sector (if oil keeps tanking the way it has been the last week or so, however, that will go away again). And so of course the stock market goes up. As FactSet ruefully noted, it was the 10th quarter in the last 16 in which stock prices went up as earnings estimates declined. The stock market knows all – except when it doesn’t.

One of the best examples of the Wall Street game is the recent rebound in Apple’s (AAPL) stock price. The stock had been off 20% over the previous twelve months going into its earnings report, so a rebound on a better-than-expected number isn’t a surprise (it’s still down double digits over the period, though). What is surprising is to hear people talk reverently about how truly great the numbers were. The rebound is sweet for stockholders, no doubt, and the company did beat consensus, but the numbers are not at all great: Apple’s sales were down nearly 15% from the prior year and earnings were off over 20%. That kind of greatness won’t prop up the stock price for long.

Looking ahead to August, I’ve already made the point that we’re entering a difficult period from the point of view of the calendar, and that is still the case. Ordinarily I would expect prices to remain under pressure until the last ten trading days, when the pros head for the Hamptons or Europe and volume dries up, leaving the field clear for their optimistic juniors. All of that said, we’ve entered an unusually long flat period in the market, and while prices are still quite overbought in the intermediate term, the jobs report this Friday is likely to prove decisive, very possibly incurring a sharp move in one direction or another. One other aspect that clouds the outlook is all of the talk from strategists expecting a short-term decline before the market marches on to close the year at all-time highs again. That kind of groupthink has a way of preceding the opposite, so perhaps we will get a rally in spite of the calendar. It will all come down to the jobs report.

The Economic Beat

The report of the week was Friday’s initial estimate for second-quarter GDP, coming in at a surprisingly low (but not in this space) 1.2% in real terms. The market wasn’t completely caught off-guard by the report, as a new “leading indicators” report from the Census Bureau on Thursday comprised estimates on June inventory and trade data that had led to quick downgrades of the existing consensus guess of around 2.4%. It was still a surprise though, as the “whisper number” had only dropped to about 1.8% by the time of the release.

The report was both as bad and not as bad as it looked, and as always it is only the initial estimate. It isn’t uncommon for this report to move as much 50 basis points (or 0.5%) by the final estimate – and then there are the revisions. The first quarter estimate wandered around for some time, starting as low as 0.5%, then moving up to 1.1% in the last revision (thanks entirely to a completely unheralded change in the price deflator, as dollar output had not changed at all). In the end Q1 is back to 0.8%, though the quarterly change in nominal output didn’t change all that much.

The report wasn’t as bad as you might think because of the big boost to the price deflator (2.2% SAAR) that is used to convert “nominal” (actual dollars) to “real” (inflation-adjusted) output, the number that gets reported in the headlines. The quarterly swings in the deflator can be mystifying, not to mention at odds with other the other major inflation measures – CPI, PPI, and PCE. It’s best to average them out over time, and in fact the best measure to look at each quarter isn’t the latest quarterly cough but the trailing four-quarter numbers. 2.2% looked like a bit of a catch-up.

On the other hand, the four-quarter nominal GDP number makes the second quarter as bad as it looked, if not worse. Although the second quarter change did improve over the first, four-quarter nominal output dropped from 2.8% to 2.4%, as low as it’s been since the first quarter of 2010, a time period that still included two quarters of official recession,. As I’ve been saying, clearly the economic cycle is nearing its end, but as nobody else has been saying, it isn’t the fault of Washington: there isn’t much governments can do about the business cycle. It’s far easier for short-term measures to crash the economy than to extend it, and even debt-fueled extensions – rarely lasting more than a quarter or two – entail some sort of payback down the road.

You will see a lot of hopeful talk about how the figure was swung by inventory contraction, along with the notion that oh well, that’ll reverse quickly as our wise businesses finish working off the excess. There are a lot of things wrong with that idea, beginning with the fact that the drawdown was by no means dramatic. An inventory accumulation number similar to the first quarter wouldn’t have been enough to lift Q2 GDP above 2% anyway – it would still have been about 1.6%-1.7% (estimated). On top of that, the inventory-to-sales (I/S) ratio is still too high, and as Ford’s (F) earnings report made clear, the auto cycle (which has a big influence on inventories) is nearly over. Finally, it doesn’t seem very sensible to assume that an inventory contraction that comes near the end of a business cycle that is already longer than average is no different from one that occurred three years ago. Recessions begin with bumps in the road. and while this one wasn’t big, the I/S figure doesn’t argue for business as usual this quarter.

Manufacturing surveys in July were decidedly mixed. The latest batch showed some stabilization in Dallas, where the index swung from (-18.3) to (-1.3), and Richmond, where it went from (-7) to +10. Chicago increased from 54 to 55.8, but the market doesn’t trust the now-privately owned Chicago data very much anymore (it’s far more volatile and opaque). Kansas City, on the other hand, dropped from 2 to (-6), and both the New York and Philadelphia Fed measures – the two most influential – were slightly negative. My guess is that the national ISM (purchasing survey) number due on Monday should be slightly over 50, but it’s only a guess.

The major report next week is the July jobs report, for which consensus is around 185K. I have nothing to offer there – I confess to a gut feeling that the number will come up short, but predicting that number (especially the initial estimate) is a fool’s errand. I would only be mildly surprised by a number over 200K, and we will get some insight as we get closer on Wednesday, when payroll manager ADP reports its own guess (keep in mind that ADP doesn’t try to guess some theoretical “real” number, it tries to guess the government number).

Construction spending is also on Monday, personal income and spending Tuesday, ADP and the ISM non-manufacturing survey are on Wednesday, Thursday brings factory orders (already largely known from last week’s durable goods report) and Friday adds the international trade number for June. The last two should start off the first revisions to second-quarter GDP. About a quarter of S&P 500 companies will report earnings, nearly wrapping up the second quarter.