“I have thee not, and yet I see thee still.” – William Shakespeare, Macbeth
Ah yes, I remember it well. It was the summer of 2000, and Nasdaq stocks had been taking a beating after their historic, parabolic ascension. It was all a necessary, healthy correction (of course). Along came the June 2000 employment report, showing an actual decline instead of the expected gain of over 300K. A stunned stock market sold off sharply on the news, then ended up roaring back in an even more stunning rally and magnificent one-day reversal. The news, you see, put the Fed on hold for the foreseeable future.
Alas, it didn’t put much else on hold, and certainly not the looming end of the business cycle. The S&P 500 fell into a bear market only a few months later. The jobs market averaged gains of only about 100K the rest of the year, with 2001 ushering in further job declines, irregular at first, before falling into the eventual recession (the official start date is now March 2001, though payrolls did not peak until several months later).
Friday’s release of the September jobs report seemed to echo that day. There are important differences, to be sure – the miss on the report was much smaller this time around (only about 60K instead of 300-plus), we had a gain (+142K) instead of a decline, and the bounce-back rally was smaller as well. History never repeats itself exactly in the stock market, but it can and will rhyme. Other economic indicators, including growth in retail sales and durable goods orders, are far weaker now than they were in the summer of 2000. The employment market isn’t as tight as it was then either, as we aren’t likely to repeat that once-in-a-lifetime, bubble-built 3.8% again for many years or even decades to come.
I don’t want to belabor all the economic parallels; if you’ve been reading this column regularly, you already know them, if you haven’t, keep reading and you’ll find more of them in the coming months. That rally sure looked familiar, though.
If history is any guide, then there is still short-term opportunity in the stock market. The S&P 500 edged higher for several months after that June 2000 report, though it never quite recaptured its spring highs. We have a light news week ahead of us, suggesting a short-term momentum tailwind for stock prices at least through Thursday, when the latest FOMC minutes are released (I couldn’t possibly tell you what they will say or how the market will react). Earnings season begins this week – but very lightly, which should at least help out this week. My thought is that by the end of the reporting season, earnings will being pulling stock prices down again, but that should then be followed by the usual Thanksgiving-to-end-of-year rally (don’t plan on the exact start and stop dates, because they are different every year).
Those rallies, like the ones back in 2000, should offer opportunities for investors to get out. Enough will probably do so to keep prices from reaching their old highs, but alas, too few will escape the widespread damage to their accounts. I don’t rule out the possibility that dramatic central bank policy moves (about which I make no predictions whatsoever) could give stock prices a last boost, perhaps even past the old highs, but I wouldn’t count on one either. Above all, don’t believe that central bank policy can repeal the business cycle. It can’t. A surge in credit flows can tack on a few more months to a cycle, to be sure, and a massive surge can add even more. There is not the remotest sign of either.
The Economic Beat
It didn’t seem likely going in, but the September jobs report managed to keep its place alongside this month’s Fed meeting as one of the two major monthly events. It made headlines everywhere.
The good news in the report was that the economy continues to add jobs and the unemployment rate remained stable at 5.1%. The alternative U-6 unemployment rate dropped to 10%, the lowest it’s been since June 2008 (in April of 2000, the unemployment rate dropped to 3.8%, a multi-decade low, and eleven months later we were in recession. Episodes of the latter are born in peaks, not bottoms).
Disappointments permeated the report, beginning with the headline figure of the initial payroll estimate of 142,000 jobs (seasonally adjusted) where 200,000+ had been expected. Adding to the damage were significant downward revisions to August (now only 136K) and July that added up to a net chop of 59K. Many veteran traders hew to the old maxim that the direction of the revisions is the most important indicator in the report.
The economically-sensitive sector of goods production showed a second month of decline (adjusted), with durable goods reporting a third consecutive loss. The quality of jobs added worsened, at least by the standards of economic strength, with the categories of retail (23.7K), health care and social assistance (36.4K), and leisure and hospitality (35K) making up the bulk of gains. Add in government’s gain of 24K, and private sector growth looked feeble indeed.
That weakness was reflected in an average hourly workweek that slipped back to 34.5, where it had been lodged seemingly forever until briefly ticking up in the summer; in hourly earnings that actually declined (along with weekly earnings), and in the household survey that reported a net loss of 236K. The monthly household numbers are quite volatile, to be sure, and the small sample sizes forbid sweeping conclusions from one month’s worth of data (the year-on-year totals do typically match up well with the payroll survey), but such a sizable decline is no sign of strength, overstated though it may be. The labor force participation rate dropped to 62.4%, the lowest level in nearly 40 years, as the not-in-labor-force category added another big chunk (+579K).
In sum, the jobs report was typical of what one would see in the end stages of the business cycle. It does not definitely signal the end, but other data does indicate that the close is getting nearer. Weekly jobless claims continue to be quite low, belying the monthly jobs number, but that too is typical of the end stage – when claims begin to show a sustained rise, even to levels that are still quite low, the party is about over.
The ISM manufacturing report for September showed up with a flat reading of 50.2 (50 is neutral, the difference is trivial) against expectations of 50.5. The miss is also insignificant and indeed, the result may actually have been a positive surprise, as the monthly slate of regional reports had been a complete sweep on the negative side, with Dallas (-9.5, 0=neutral) on Monday and Chicago (48.7, 50 neutral, 53-54 expected) on Wednesday completing the negative column.
The Chicago report is the only regional survey of note that isn’t done by the region’s Federal Reserve branch, instead using a survey of purchasing managers, like the national ISM report. One shouldn’t expect the sets of numbers to always line up. The national number did align in one respect, however: the ratio of growth-to-contraction sectors fell to seven growing versus eleven in contraction. I haven’t researched the issue, but would guess that to be the first negative score in five years, if not longer. New orders were flat, while exports (46.5) and backlogs (41.5) continue to post contracting results, the latter a noticeably steep decline. In the same vein, factory orders in August were reported to have fallen 1.7%, with a downward revision to July (now +0.2%) and cap-ex spending. Manufacturing can dip in and out of weak periods during a cycle, but seeing this kind of data eight years after the last cycle ended is cause for serious concern.
Pending home sales fell 1.4% in August, potentially continuing a trend of moderation that showed up in the July existing home sales report. New home sales growth has been strong this year, albeit at low levels, so it may be early to say much more about a sector that I expect to be amongst the last to contract. The same goes for the personal income and spending report which showed gains of 0.3% for income and 0.4% for spending in August. Neither of these categories have been good leading indicators in the past, instead lagging alongside employment. Case-Shiller data showed home price appreciation to be steady at 5%.
Also on the plus side last week, consumer confidence rose to a counter-intuitive 103 last month, the highest since September 2007 – but that was also the last month before the last recession started. Construction spending showed an initial estimate (the numbers are heavily revised, with July cut to 0.4%) of a 0.7% gain for August. The ADP payroll report missed to the high side this time, with an estimate of 200K – the report is forever trying to backward-match to the government reports, so I wouldn’t make much of the difference. Auto sales leapt to a recovery high rate in September, at 18.2 million, a 10-year high. The data are skewed by the inclusion of the Labor Day sale, the highest-selling weekend of the year; expect a pullback in the October rate.
Next week brings the ISM non-manufacturing report on Monday, which should be fine. International trade reports August on Tuesday, and the data may get more attention than usually given all the blame being heaped upon the stronger dollar’s impact on exports. The main report of interest will be the release of the FOMC minutes on Thursday, but I’ll be looking at wholesale trade data on Friday, along with export-import prices (despite dollar strength, the former have been steadily negative). The soon-to-be-split-up Alcoa (AA) kicks off the third quarter earnings season on Thursday. Volatility should pick up again as earnings season gets into full swing in the ensuing weeks.