“Summertime, and the livin’ is easy.” – George Gershwin, Porgy and Bess
So what you really need to understand about the stock market this summer can be conveniently summed up – for now – by the chart below.
Note the top blue line – it is the 50-day exponential moving average (EMA) for the S&P 500. It is one of the mostly widely followed trend lines.
The S&P 500 last made a significant drop below the 50-day on the eve of the Presidential election in November. At that time, it tested the 200-day average before recovering on the mania of lower taxes and higher spending as far as the eye could see. Though the so-called Trump rally has flagged at times, and is arguably over, until the 50-day trend can be decisively broken, traders are not going to sell this market without a very good reason for doing so. Those reasons do not include Washington gridlock.
Every time the S&P has tested its 50-day since November – four times in all – the index has responded soon afterwards with another great leap forward. The last two times have been especially convincing on two counts – one is that the rebounds have been sharp, two is that the rebounds have really no other reason but themselves. When a market is this chart-beholden, traders sit up and pay attention. And worship, because it is easy money (until it isn’t).
Oh yes, the market is overvalued and getting worse. Of course it is. Talking heads debate this state of affairs all day long, and you can read many articles going back and forth about valuations, including the usual ones that the problem isn’t that the market is rich by traditional metrics, but that those metrics simply don’t apply these days. Because interest are too low, or inflation is too low, or commodity prices are too low, or savers are too desperate, or whatever. This time is different.
None of that really matters. Valuation doesn’t make stocks go up and down. In the absence of something genuinely scary, the market is going to live by the 50-day for as long as it can. Should something big occur, we might get another test of the 50-day, and should something bigly, hugely bigly occur, we could even get a test of the 200-day. We are certainly due for one. But nobody is going to get in the way of this. The main strategies these days aren’t how to get out of super-valued markets, but how to rotate around the trend lines with “enhanced” strategies like sector betting or doubling down on the Nasdaq 100, possibly while shorting something no one likes these days (e.g., retail) for downside protection.
As we cruise through the guaranteed uptick of the July earnings report season, August looms and we’re due for some backing and filling soon. I’ve seen some pretty thundering Augusts in my day, and it’s often a period of weakness. But I’m not sure what can shake this chart fever (and fear). Even the resignation of the president might not be enough to do it (the markets surely prefer Mike Pence anyway). It’s probably going to have to be something worse. It won’t be lack of earnings growth, either – if earnings were up only half of expectations this past quarter, we’d still tread water. You will read stories about record highs being propelled by earnings, but that is standard issue from all but the very worst earnings seasons. Watch the chart.
The Economic Beat
The report – including my spreadsheet work – of the July 7th week was wiped out by a power outage. I really must be more careful about backups. Reconstructing, the highlight of the week was the June jobs report, which showed a headline gain of 222,000 (222K) versus expectations for about 175K. It may have been the only good economic data of the month (manufacturing surveys have been strong, but production data remains at lethargic levels), but there is nothing that the Street loves more than a positive jobs surprise. Not far behind are positive revisions to prior months, which some consider equally or even more important: May was revised 14K higher to 152K, while April got a bigger boost, from 174K to 207K (+33K).
The ADP report seemed to come up short for a change, with a report of only 158K vs. expectations for 180K. However, the ADP report only tracks private sector employment, which the Labor Department (BLS) reported as 180K, the same as the consensus guess for the ADP report. The private sector number seemed to get lost in the headline beat, which included the biggest jump in government jobs in some time.
The report wasn’t as good as it looked, though it’s still just a first estimate. The unemployment rate did tick up to 4.4% from 4.3%, but that was due mainly to a bigger labor force, as the participation rate also ticked up: the rate uptick is definitely not a problem. The issues are more one, wage growth; and two, the continued decline in the employment growth rate.
Hourly wages were up a mere 0.15% on the month, and only 2.5% year-on-year. Weekly earnings are a bit better (+2.76%), thanks to an uptick in the work week from 34.4 hours to 34.5, an uptick that in recent times has not proven itself to as sustainable. The larger issue is the trailing-twelve-month (TTM) growth rate in employment. It now stands at 1.52%, before revisions, compared to the year-ago 1.74%. That’s the lowest TTM rate for June since 2011, when employment growth was still rising. Six years later, it portends a slowing economy. The Street doesn’t worry so much about portents, though, not unless they’re accompanied by a good chart. We’ll see what the July report brings.
The ISM June manufacturing survey, as usual, was full of optimism that has yet to be borne out by production data. The survey rose to 57.8 from 54.9, with a 57.2 gain for employment. By comparison, the BLS estimated a net loss of 1K manufacturing jobs for May and June combined. The ISM sector growth-contraction score was 15-3, and the production index soared to a very high 62.4 (up from 57.1), while new orders were at 63.5 (up from 59.5). Industrial production data released the following week did not echo the survey – as usual. It’s as if the survey respondents are trying to construct their own narrative.
The Markit Economics purchasing manager survey painted a different picture, showing a slight easing from 52.7 to 52.0. It has to be said that the far more junior Markit survey has tracked other economic data much more closely this year. The ISM non-manufacturing survey followed up a couple of days later with a reading of 57.4 (previous 56.9); the Markit services number was 53.
The latest factory orders report showed essentially flat shipments over the last three months. One bright spot was a pickup in orders core capital goods, along with an upward revision to April. The TTM change is still negative year-over-year and well below levels prior to the 2015-2016 slump, but there is some improvement. Construction spending was unchanged in the first June estimate, with the more reliable second estimate for May showing a decline of (0.7%).
Coming back to the labor market, the latest Fed Labor Market Conditions Index dropped from 3.3 to 1.5, but that 3.3 figure was a big revision upward from an original 2.3. However, the market still pays very little attention to it. The labor turnover survey (JOLTS) reversed its trend of higher openings and lower hires with a rare result of fewer openings and more hires. The hire rate of 3.7% isn’t much higher than the year-ago 3.6%, but this is the kind of trend you would expect to see in a full-employment economy.
Wholesale trade is doing better – sort of. May fell 0.5% from April (SA), but was still up 6.2% (SA) from a year ago. But TTM sales (unadjusted) are steadily improving, positive for the fifth month in a row at 3.9%, 2.7% excluding petroleum (which are not price-adjusted). Inventory remains at high levels, with the inventory-to-sales ratio at 1.29 (SA). It’s been between 1.28 and 1.29 the last six months – lower than last year, but higher than historical averages.
The Fed’s Beige Book was not its usual “modest-to-moderate” self, opting this time around for “slight to moderate,” though the report did note that the majority of districts did fall into the former category. The report also noted softening retail sales, an accurate harbinger of the June retail sales report released later that week. That report showed retail sales down 0.2% (SA), for a second consecutive month of decline, though May was revised upward from (-0.3%) to (-0.1%). The June year-year (y/y) unadjusted increase was 3.24%, the softest comparison in six years. Even so, the second quarter was up 4% on an unadjusted basis, its best showing in four years.
The TTM rate overall was very little changed at 3.5%, in line with the range of the first six months, but the calendar was favorable and may look a bit worse in July. The TTM rate for sales excluding autos and gasoline, one of the so-called core rates, is faring decidedly worse. It’s up 3.3%, not terrible, certainly, but the softest rate of growth in over three years (since March 2014). The “new normal” of the current economy, with lower GDP and lower inflation, tends to skew nominal comparisons, but 3.3% nominal rates came in the midst of the last two recessions. Adjusted for inflation, we are still just above the treetops.
Inflation data isn’t making the Fed’s job any easier. Producer price inflation (PPI) was up only 0.1% in June, dropping the y/y rate from 2.4% to 2.1% (SA), and the ex-food and energy rate to 1.9%. Consumer inflation (CPI) was unchanged in June (SA), taking the y/y rate down to 1.6% overall and 1.7% excluding food and energy. That’s low, and quite a comedown from the first quarter bulge (2.8% in February, for example).
Industrial production for June rounded out the period with an increase of 0.4% overall and 0.2% in manufacturing. The main strength was in mining. The unadjusted y/y rate stands at 1.8%, down from 2% in May. Capacity utilization is a very modest 76.6%, up from 76.4% in May (previously estimated also as having been 76.6%, so wait and see).
The main focus of the July 21st week would be housing data and two more manufacturing surveys, the New York and Philadelphia Fed versions.