Tiny Bubbles


“Tiny bubbles, in the wine, make me happy, make me feel fine.” – Leon Pober (composer), Tiny Bubbles

Does anyone remember June 2, 2000? It was one of the more memorable trading days I’ve ever seen. On that day, the Labor Department announced that private non-farm payrolls fell by 116,000. A number north of 200,000 had been expected, and there were hopes for plus-300,000.

The market was rocked. As I recall, Dow futures were down over 200 points before the open. April had been a rocky month after a March-madness peak (does that sound familiar?) and this was a stunning slap in the face.

Then one of the biggest, most improbable reversals I’ve ever seen took place. After the opening drop, the dip-buyers did their usual rush to buy. After all, this had been a mostly winning strategy for over a year. The word spread like wildfire across trading desks that such a result would make Fed intervention a lead-pipe guarantee. Ergo, nothing to fear, because you “don’t fight the Fed.” The S&P 500 finished up over 1 1/2% on the day.

I’ve seen that episode rerun a number of times on a smaller scale over the years. A surprisingly bad number opens the market down sharply, and a Fed-to-the-rescue rumor inevitably starts. If the number was some physical stock number, like durable goods or new homes or whatever might have been hot, this too is remodeled as good news, because it will “clear inventories.”

Friday, April 5th of 2013 was not another June 2nd, but it was an imitation of sorts. The jobs number was much lower than anticipated, the market opened sharply lower, and then the narrative took back the reins. The speculation over when the Fed might start to rein in quantitative easing was suddenly transformed into a guarantee that the Fed would have to stay in the game the rest of the year, aided by some dovish Fed comments. Although the S&P finished down, some of the high-beta stuff, like the MDY (S&P mid-cap) and the IYT (transports) managed to rally back to finish on their highs, with the former nearly unchanged and the latter actually in the green.

The Treasury market was having none of it, finishing up strongly on the week. Here’s a tip – whenever the bond market and the stock market are saying two different things, you can make money fairly reliably by betting on the stock market for a week or two – momentum is momentum – but taking the bond market as the eventual winner. Depending on the time of year, stocks can gut it out for more than just a couple of weeks, but don’t be fooled. You can still take bonds and give heavy odds to boot. Note also that gold rallied sharply on Friday, reversing its recent weakness.

It’s not good to go overboard on only one month’s data. But the year-on-year job change has been lower than the 2012 equivalent every month this year, reversing the pattern of improvement from the last three years. The last time the year-on-year changes reversed in this manner was 2007.

The global economy is weakening, but it will take a bit more time for that to be clearer. On Monday the talk was about when the Fed might begin to tighten, but by Friday Treasuries had hit their high on the year. Ironically, the Wall Street Journal ran another article on Monday that wondered where the factory revival was, quoting Dan Meckstroth, an economist for a big manufacturing trade group as saying “there’s simply no statistical evidence of a broader renaissance at this point.” Clearly he isn’t reading enough Wall Street buy recommendations.

Have you notice what’s going on in China? The Hang Seng market index is down nearly 10% this year. Official Chinese export data is showing a huge discrepancy with Hong Kong data. By some odd coincidence, the government is beating up on foreign companies. Yes, it’s clear: The Chinese rebound must be on the way. Just like it is in Europe, where yet another batch of depressing PMI data was released this week, all of it in contraction territory, even Germany. But some of them beat estimates, an exciting event that ought to turn around their unemployment rates any day now.

Leveraged loans – basically, subprime lending for business instead of homes – is this year’s hot fad. The Fed expressed alarm a couple of weeks ago. On Thursday, State Street (SST) and Blackstone announced a new leveraged-loan ETF. It seems that they’re “on fire” – probably because the major investment banks have been pitching them like mad in recent months. They’ll protect against interest rate rises, you see. Remember how portfolio insurance was going to protect against equity declines? “Tiny bubbles, in my wine” goes the song. Or was it my brain?

But the fat lady hasn’t sung yet. Despite the growing nervousness over France joining the list of at-risk countries (the economy has been on the list for some time), French yields set a new all-time low at auction this week. In a world of ZIRP, you just hold your nose and hope the European Central Bank (ECB) can come through on its promises. Consensus estimates for the earnings season that starts next week are for one percent growth, a lofty goal that shouldn’t be too difficult to beat. Especially when much of the per-share gains are coming from reduced share counts.

The ECB still has an interest rate cut left to spend, and the market can switch back to rallying on hopes of no Fed contraction. The central-bank-invincible-armor story still has a few miles left on it.

Don’t forget the seasonal effect, too. Corrections are common in May and June, but crashes are almost non-existent, the “flash crash” notwithstanding. For example, had you gone into cash at the close of June 2nd, 2000, you would have missed out on the chance of an additional 3% gain over the next three months, even after that silly rally. Of course, you’d have also missed out on the 46% loss over the next 13 months, but that’s another story.

The Economic Beat

The jobs report probably wasn’t as bad as it looked from the non-farm payroll point of view – a gain of only 88,000 against expectations for about 200,000 – yet was just as bad as it looked from the household survey point of view, a loss of 206,000. The latter survey is subject to large revisions, but it’s still a long way from no change. An above-average revision of 300K, taking the number all the way back to plus 100K, isn’t out of the question, but then one would still be left with a rather unimpressive 100K.

I favor the March payrolls number to be upwardly revised – the March 2012 number was released as 120,000, seasonally adjusted; after last month’s annual benchmark revision it now shows a gain of 205,000. Wednesday’s ADP payroll report, a clue that the 300k+ whisper-number might not be happening, reported a gain of 158,000.

But before you rip the Labor Department (again), it must be said that first-quarter data is difficult to get right in real time, particularly in regards to seasonal adjustments. The reality of the budget calendar means that the reported workforce shrinks by around 2% after the end of every year (the difference between actual December and January payrolls). The mix of terminations, retirements and job changes makes for a high turnover that is difficult to report or count. The tally typically shrinks as noted in January, then rebuilds over the next few months. Some of it – hopefully a lot of it – is growth in the number of jobs, and some of it is uncounted jobs getting found again. In a good year, there is a net gain in counted jobs over the first four months, despite the sizable initial decline.

I criticized the February data last month as doing a poor job of accounting for retail trade, and it looks to me as if the problem stands only partially fixed. The decrease in March retail jobs may not be any more substantive than the increase in February; I suspect that sector net hiring should have been about zero over the two months, seasonally adjusted. Manufacturing reported a small loss of about 3K, but temp hiring remained steady at about 20K. Transportation and trade declined, and that isn’t healthy.

There are also cross-currents in the trends that are somewhat puzzling. One that stands out is that the unadjusted gain in February payrolls (1.022mm) was the highest ever for the month in my own database, which goes back to 1980. On a more representative percentage basis (to account for the higher population), it’s the biggest since 1999.

Yet 942,000 of that total, more than 90%, were “part-time jobs for non-economic reasons,” according to the household survey (meaning they would work full-time if it were offered). The household data aren’t directly comparable with the payroll figures on a monthly basis, yet clearly something is up. That was an historic increase in part-time workers, sizable even if cut in half. Workers in the “economic reasons” category fell back sharply in March, making for a puzzling pattern indeed.

The answer might unfortunately be traceable to the very large drop in the participation rate, to 63.3%. That’s the lowest rate since 1979, and doesn’t square with the seemingly robust number of additions in February (now reported as an impressive 268K in the seasonally adjusted total). When hiring is strong, the participation rate is supposed to increase, to the point that the unemployment rate may actually go up as people re-enter the labor force faster than they can be hired (or counted). Many took note of the fact that the unemployment rate would have moved up to 7.9% if the participation rate had stayed at February levels, and stock markets around the globe can be thankful that that particular result didn’t cross the tape.

The drop in the participation rate and a low jobs result is consistent with most of the recent sentiment surveys. On balance, they haven’t shown strength in hiring intentions, with the consumer confidence report showing an especially weak number and both ISM surveys indicating static-to-soft hiring outlooks.

A plausible explanation advanced at Briefing.com, amongst others, is that employers are standing pat on hiring. That has the advantage of being consistent with some of the low levels of weekly claims reported in early March. A “wait and see” attitude on the part of businesses with substantial government-related revenues makes intuitive sense – they’re waiting to see how the sequester sorts itself out. Still, that shouldn’t translate into eventual net hiring, because some sectors are going to be making cuts.

Those weekly claims did jump to a reported 385,000 last week, but that should be taken on a very provisional basis. The seasonal adjustment factor was quite large, as the Labor Department had to factor in the seasonal effects of both the end of March (actual layoffs are usually at ebb tide at the end of the first and third quarters) and Easter. It was something of a guess. I will note two warts in the report, one being that the unadjusted estimate from last week was not revised upwards, an anomaly both unusual and positive. On the other hand, the 314K raw estimate compares with 315.6K in the year-ago week, which is not positive, given that the Easter holiday should have acted to reduce claims.

We may find out a year from now that more jobs were added than thought. But for the moment, the year-on-year totals show a gradual softening. January’s year-year rate was down 25 basis points from the 2011/2012 rate. February and March both stand at 39 basis points lower. April 2013 is going to have to be awfully strong to pull the four-month initial period into positive territory for actual jobs, as happens in a good year.

There are ideological arguments being hurled about in the aftermath of the weak numbers, from claims that the prospect of increased health-care costs in January 2014 are responsible, to the notion that capitalist cabals are seeking to further enslave their hourly-wage lackeys.

However, the dominant motive for adding inputs to production is demand. Costs matter, but sales and profits matter more. If demand is anemic, it follows that spending on inputs will be as well. The recovery is now four years old, and we may have reached the plateau of the cycle. Canada did have a little June 2nd, when it reported the biggest job loss since 2009 (over 50K) in the face of an expected gain.

Imports of goods fell in February on a year-over-year basis, and weakening import demand is one of the strongest clues in recent times to a turn in the cycle. There was good news in the report too – imports of crude oil have also been falling, roughly five percent in barrel terms. Combined with a price drop of around ten percent from last year, it adds up to a big decline in money sent overseas (I’m presuming that the drop is mostly from added US production and not from decreases in demand, though clearly high gasoline prices did cut into product demand). It will also help the Q1 GDP print (imports are subtracted).

The ISM surveys of manufacturing and non-manufacturing industries shared the distinction of both printing the lowest March results since 2009. Both were below consensus, with the more influential manufacturing number providing the bigger miss. That came in at 51.3 versus a consensus of around unchanged (February was 54.2), while the non-manufacturing number reported 54.4, missing estimates for 56 and down from that level the previous month.

The reports were by no means disasters. The manufacturing report counted 14 industries as growing versus 3 in contraction, a solid score, while the non-manufacturing tally was 15-3. However the manufacturing new orders result was sharply lower (51.4 vs. 57.8), as was the prices gauge for non-manufacturing (55.9 vs. 61.7). The manufacturing employment question had a small net gain, but the much larger non-manufacturing survey showed a substantial drop (53.3 vs. 57.2). There does seem to be some softening.

Construction spending added 1.2% in February after a 2.1% loss in January. The year-on-year pattern has been fairly level for the last twelve months, with gains in the last two months down a bit from levels in 2012, when activity was boosted by warmer weather. Single-family home construction continues to do well, but I suspect a slowdown in commercial construction is coming soon. Factory orders for February advanced 3%, beating some estimates, but the revisions it contained to the previous week’s release of new orders for durable goods weren’t pleasing. In particular, the business investment category got a hefty bump downward to (-3.2%). That will cut into March shipments, a potential drag for first quarter GDP.

Motor vehicle sales continued at a decent rate (15.3mm annualized) in February, and the weekly retail reports showed strength leading into Easter. The combination of the two should lead to a decent March retail sales report. The consensus estimate currently stands at a ridiculously low 0.0% – talk about rigging the game! Look for something in the plus 0.3%-0.5% range.

That sure-fire “surprise beat” could combine with the Wednesday release of the Fed’s FOMC meeting minutes to put the market back to the important task of overreaching to 1600 on the S&P before giving up. Other reports include wholesale inventories on Tuesday, trade price data on Wednesday, the Producer Price Index on Thursday (PPI) and the Michigan sentiment index on Friday.

Alcoa (AA) will kick off the first-quarter earnings season on Monday, but the market doesn’t may as much attention as it once did, probably because the company has so often struggled in recent years. The more important reports begin on Friday, with JP Morgan (JPM) and Wells Fargo (WFC). I suspect many questions about litigation, putbacks and fines await them.

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