“Summertime, and the living is easy.” – Gershwin-Heyward, Porgy and Bess
Well, the July rally is in the books, and August is off to a typical first-of-the-month rally start. So the question is, where do we go from here?
The answer for this month at least is probably a round-trip ticket to where we’re standing right now. The major indices are overbought – most egregiously, the Nasdaq – and we’re due for a nice pullback. August usually starts off weak, the first-day rally notwithstanding, but the US calendar is light next week and that tends to favor the current trend, in this case up. There’s a dose of overseas PMI surveys before Monday’s open, including a batch from Europe. Any readings above 50 might only indicate stagnation – particularly those between 50.0 and 50.5 – but will be taken positively by the market and reprinted with misleading headlines like “strongest expansion in months,” though they mean no such thing.
It’s also the start of the summer vacation season on Wall Street and in Europe. Some have already left, but August is the heart of the season. The already-light volume stands to get lighter unless something big happens, and the Middle East is the most likely source of that for the time being. Although some of the talking heads are claiming that oil is rising on economic strength, don’t you believe it. It’s up for two reasons, the first being that the Fed didn’t talk the taper (in which case, buy the dollar and sell commodities), and the second being that traders aren’t ignoring what’s going on in the Middle East, even if the mainstream news isn’t paying much attention.
If you’ve any doubt about the sway of the Fed (market obsession, really), consider that the little two-day rally on Thursday and Friday in the face of weak GDP and disappointing jobs could never have happened without the Fed’s decision to stay full speed ahead and not talk about tapering with its modest downgrade to the economy: Growth went from “moderate” to “modest.”
The market appears to be locked into Goldilocks-land right now, and without some externality is likely to follow a simple calendar-pattern trade – a minor pullback in the first half followed by a rebound in the dozen or so days before Labor Day. It must be said, though, that August declines can be quite steep when something unexpected does come along – e.g., an Al Qaeda attack – as the lack of volume and staff makes it easier for prices to move faster in a short amount of time. It’s hard to be sure of such things but judging by what people are telling the media, complacency seems to be at truly insufferable levels right now, leaving prices ripe for a stumble.
As to the Fed decision, well now, that’s a different story. The Fed is rather boxed into a corner right now, something that many traders are aware of and most others simply don’t want to think about (or will wait until the jig is up before fleeing). The Treasury is lowering its borrowing needs this quarter, leaving the Fed to be buying up more and more of the issuance. The bank can’t let itself get into a position of cornering supply without some valid emergency. On top of that, the only apparent effect that buying Treasuries is having anymore is on the stock market, which might seem nice but – as valuations drift higher and higher from the underlying corporate reality, the return trip will become more expensive.
I don’t expect this reality to sink into the market this month. There’s no telling when such things can happen, of course, but I think that the day of reckoning is going to come later, probably in mid- to late fall. As happens with every aging bull, more and more have bought into the notion that the present state of circumstances – earnings that are barely positive, and so not actually “bad” – and a central bank buying $1 trillion of securities a year is something that can stretch out into an infinite horizon, leaving one little to do but hold out the buckets and collect the cash. Of such thinking is born abrupt and painful losses.
The Economic Beat
The jobs report was not the report of the week for a change, primarily because it didn’t bring the same joy as the Fed’s report of no change in the status quo. It wasn’t a bad report, but it wasn’t a very good one either, with many mixed signals.
First there was the miss – 162,000 versus an official consensus of about 190,000. There was a wide dispersion this time around of what the consensus really was supposed to be – I suspect a dash of spin control – but after the ADP estimate of 200,000 on Wednesday, the market was really dialed into that number as a baseline and indeed the Briefing.com consensus, which usually tries to incorporate late changes, was 195,000.
Some of the other obvious minuses include the small declines in average hourly wages and the average workweek. It’s likely that both are due to a mix shift – in the first case, most of the jobs being added are at the bottom of the wage scale, about which more below. In the second, recent household survey additions have tilted towards part-time labor – the survey estimated that about half of the new positions were part-time, and that was actually quite a bit lower than last month. It’s going to drag on the average.
The participation rate (the percentage of the population that the government counts as working or trying to work) edged down again and at 63.4% is now 0.3% below July 2012 (this is actually a lot for this figure, particularly as it should be going the other way in a recovery). The household survey’s “not in labor force” (NILF) estimated increase was 240K (people counted as neither working nor officially looking), greater than the estimated increase in the population (204K) and way, way ahead of the discouraging estimated drop in the actual labor force (-37K).
Then there are the revisions – many veteran traders hold that the direction of the revisions are a better short-term indicator than the actual first estimates, which often end up being significantly different a year later. In the current case, they were negative for May and June.
Those revisions slightly brought down the percentage increase in the Labor Department’s year-to-date payroll gains (not adjusted), from 0.93% to 0.91%. That’s below 2012 (0.95%) and 2011 (0.96%) over the same period, though the 2013 numbers are subject to likely revision. However, a spot of moderately good news was that the preliminary increase through seven months was 0.09% in 2013, compared to 0.06% in 2012. If that seems confusing, it’s because the end of the second quarter, like the end of the year, actually results in a lot of separations, such that January and July always have declines in counted workers – the numbers you see in the press are seasonally adjusted to account for this phenomenon.
The point is that it appears we’re running only slightly ahead of last year, perhaps level after revisions. The difference is small enough – and the data really isn’t that precise – that broadly speaking it really is the same, and is probably due to the auto manufacturers having reduced shutdown levels in July. One reason I say that is that the ADP survey counted 5,000 fewer manufacturing jobs, while the Bureau of Labor Statistics (BLS) number says it was 6,000 more. I suspect some estimation differences falling out of the auto sector; expect both numbers to be revised.
Another puzzling difference was that ADP counted 22.000 more construction jobs, while the BLS said it was 6,000 less. Given the time of year, one would think that ADP was closer, but construction spending was estimated to have declined so best to wait for the revisions.
The mildly good news was that the unemployment rate did fall from 7.6% to 7.4% (adjusted), due almost entirely to the drop in the participation rate and the decline in the labor force. The alternative U-6 rate also fell, from 14.6% to 14.3% (not adjusted). The household’s official count of unemployed fell by 263K, but with 240K new NILF additions, it’s not much of a victory.
Finally, nearly half the gains (about 78K) were the typical bottom rung: retail trade, leisure and hospitality, home health care. The personal income report released alongside the jobs report estimated year-on-year real disposable income growth to be a very meager +0.6% through June. The sum of it is that employment continues to creep forward, but is not robust, nor is any sector than be can outsourced apart from autos. There is robust homebuilding growth, but the activity level is not robust, with levels still near multi-decade lows.
There was some consolation that weekly claims, on an unadjusted basis, fell to their lowest level since October 2007. Before you get too excited, though, employment is a lagging indicator and claims at the end of September 2007 – three months before the recession began – were lower than September 2006. Total covered employment (i.e., workers paying unemployment insurance every quarter) rose all the way into the first quarter of 2009, fifteen months after the recession began. The current total is still lower than the first quarter of 2010.
Moving on to other areas, the recent spike in interest rates was likely the reason behind declines in both pending home sales (-0.4%) and mortgage-purchase applications (-3.7% on the week, up only 5% year-on-year). Perhaps what’s been overlooked in the excitement about rising home prices (+12.2% year-on-year and +1.0% for the month, according to Case-Shiller) is how much of the buying has been done by investors, particularly large ones like the Blackstone Group.
On the manufacturing side, the Dallas Fed survey showed the region edging down to a +4.4 rate, while both the Chicago PMI and the national ISM manufacturing surveys (forget the acronyms, the groups are otherwise the same) rose, Chicago to 52.3 (below the expected 54) and the national to 55.4. I had to smile when I heard a talking head exclaim over the “strong ISM number” Friday morning, because the increase was all in seasonal adjustments, for which we can once again thank the blip from the auto industry. The Wall Street Journal followed it up with a moronic lead about factories having their best month in two years, which goes to show how some of their staff seem to be learning on the job. They have work to do on diffusion surveys and seasonal adjustment.
I can’t tell you too much about the Chicago survey, because to my great irritation the group has just sold its report to the stock exchange monolith Deutsche Boerse, which has turned it into a paid-subscription report. The Chicago organization is a non-profit, which is one reason I’m annoyed; the other is that the exchanges, in their desperation for revenue to make up for lost volumes have increasingly resorted to trying to sell news that was once public domain. Worse, they’ve been trying sell it ahead of official release times to high-speed traders – essentially allowing them to trade on inside information, but for a fee so it’s okay – something that government agencies are starting to rightly look askance at.
The factory new orders result, which so far as the market is concerned revises the durable goods report from the prior week, came in positive at 1.5% but below expectations. New orders fell (-0.4%) excluding transportation, but the business investment category was revised upward, from +0.7% to +0.9%. However, the recent inventory build-up may mean that the series trends down again in the current quarter.
And there was the first estimate of second-quarter GDP. I wrote a lengthy piece on it for Seeking Alpha, but the essentials are that the price deflator was suspiciously low for the quarter (0.7%, well below every other annual inflation rate), resulting in a real headline rate of 1.7% that was above consensus of 1%, but would have been below consensus using the first-quarter deflator of 1.7%. Did inflation really drop that much in the quarter? I certainly didn’t notice it at the gas station or grocery store. Nominal GDP – that is, GDP before inflation adjustments – dropped from a downwardly revised 2.8% in the first quarter to 2.4% in the first. If we don’t get a rebound this quarter, the four-quarter nominal rate is going to drop below 3%, and that is unambiguously stall speed.
With construction down (-0.6%) in June, probably influenced by some rainy weather, where is the “robust” economy I keep reading about in the back pages of the Wall Street Journal (though not on the front page, interestingly enough. Maybe they’re trying to keep the Goldilocks thing going)? Perhaps it’ll show up in July retail sales: The Redbook weekly service is projecting a weather-related increase in ex-auto, ex-gas sales for the month. Or maybe not: The other weekly service, ICSC-Goldman, says the rate softened. That usually translates into an average month, but we’ll have to wait until the week after next to find out. While the consumer sentiment surveys are at high levels, they mostly measure where the stock market has been and the time since the last catastrophe.
Auto sales, it must be said, are doing well and are nearly back to 2007 levels. They have the advantage of a functioning credit market (the asset-backed market (ABS) for car loans has been well all along) and an aging U.S. car fleet. However, income is growing slowly – those car payments are crowding out other spending.
The only high-profile release next week is the ISM non-manufacturing survey on Monday. Many of the most recent regional and national surveys for manufacturing enjoyed big doses of seasonal adjustments, including New York, Philadelphia and the national ISM. It remains to be seen whether the services sector will get the same benefit. I’ll be looking closely at the June trade data on Tuesday – the estimates in the GDP report looked implausibly high – and the wholesale trade data on Friday.