Divergence


“What would you do if you were stuck in one place and every day was exactly the same?” – Bill Murray, in Groundhog Day

Don’t the headlines look sweet this weekend? The Dow tops 14,000 for the first time since 2007. Both it and the S&P are nearing all-time highs, with CNBC posting a running clock showing the points remaining to get there. If nothing else convinces you that we are headed for a euphoric peak that will have a painful payback, that alone ought to do the trick. We are awash in contrarian indicators, but they aren’t likely to matter quite yet.

Yet while the market soars, we are also at risk of deepening the class divisions within society. The latest jobs report is covered in more detail below, but certain elements of Friday’s news indicate a broad structural problem.

One report Friday concerned U.S. auto sales, which were quite good in January. There was a lot of talk about confidence and security, but really nothing has changed recently on the employment front. The unemployment rate for people over 25 years old is hovering steadily at 6.5%. Some bonuses were accelerated into December, perhaps making people feel more flush (and the auto workers are getting nice bonuses), and there was no fiscal cliff (yet).

The main factors, though, are the age of the US car fleet (the mean is over 10 years) and financing availability. Credit is very difficult in the mortgage market, but the asset-backed market has been flying and terms for both buying and leasing are very attractive. It isn’t hard to get a new car if you’re working.

The unemployment rate amongst the young and/or uneducated, though, remains startlingly high – 23.4% for those under 20, 8.1% for those without college. For this group, income is a challenge, let alone cars. What’s more, two traditional routes into higher-wage jobs, construction and manufacturing, are not going to help as much this time around. In the wake of World War II, you could get a good factory job with an eight-grade education. The modern factory floor is much higher-tech than it was even twenty years ago, let alone the 1950s.

While the rising stock market does lend a certain air of optimism, as evidenced in the increase in the University of Michigan sentiment survey, it’s fleeting and not that broad, as evidenced in the Conference Board’s confidence survey. The current rise in the market has perhaps two weeks and one or two percent left in it, and I’ve no doubt that most of the year’s gains will all come crashing back to earth by late April or early May (if not before). In the meantime, most of the gains are accruing toward the wealthy.

The tape makes the news, and so you are going to read this weekend about the accelerating global economy. The World Bank, the IMF, the European Central Bank, and the Federal Reserve have all recently cut their forecasts for 2013. The irony is that the principal weakness all of them have in making forecasts is that they tend to extrapolate too much from the most recent data. But if that is the case, then where does the talk about the global economy accelerating come from?

From the stock market and its doyens, that’s where. If you read somewhere that investors “enthusiastically welcomed the job report,” don’t believe it. The Dow futures were already up 70 points before the jobs report was even released, and 40 points within a couple of hours after Thursday’s close. It isn’t because the report (whose number was below consensus estimates) leaked out in advance. It was because Friday was the first day of the month, the current momentum is up, and it’s the first quarter. That translates into program buying. The fact that the jobs report was about a “C-plus” functioned as a permission slip to jump on board. There was a Fibonacci target to reach (1510 on the S&P 500), and of course the metaphysical 14,000 on the Dow, which represents – wait for it – a round number.

Had either the jobs report, or the GDP report that not only missed estimates by a mile but reported a fourth-quarter decline, been released in October, prices would have gotten crushed. But because they came out in January, trading algorithms don’t give them as much weight, and buy anyway. They do this because the recent historical evidence shows that prices rise in the first quarter, seasonal adjustment factors exaggerate to the upside, and so this is the time to make money. It gets flipped around by the end of April or early May.

Consider that Australia’s PMI fell to a miserable 40.2 and new orders dropped below the 40-level. Its reading is approaching 2009 recession levels. Or that France’s PMI hit a four-year low. Or that Germany was negative for the eleventh month in a row (but hey, not as negative). China’s official PMI crept back towards the neutral level, but if you didn’t like that you could look at the private HSBC report instead, which followed with a small increase (try telling Australia that China is getting better). The seasonally adjusted eurozone PMI was negative yet again (47.9), but rose to its highest level in eleven months. In other words, it “recovered” last spring too. It’s not hard to infer that we’re seeing nothing more than another artifact of the first quarter and its erroneous seasonal adjustments.

One does not hear talk of a European recovery from companies doing business there. They have warned of another difficult year. So did MasterCard (MA), which lowered its outlook for first-half spending (including the US). The EU faces more difficulties ahead, not the least of which is a soaring currency. The reason for the rise in the euro is two-fold: the market believes that tail risk has been removed by the ECB (though the promise has not been tested), and the ECB may be the only central bank left on the planet not trying to devalue its currency. It’s a holdover mentality from the pre-EU days – European bankers, in particular the Germans, think currency strength is a competitive badge of honor, like the national soccer team. Their manufacturers are not celebrating.

But the global economy is accelerating, because the stock market says it is. And the stock market will keep rising, more or less, until reality or the calendar weighs it down. Reality could start to come in mid- to late February, when spending data should start showing the strain (of course, there are always seasonal adjustments that could obscure matters a bit longer), and the reality of the sequester begins to raise its head. Prices may be dramatically over-extended by then anyway.

Absent that, the correction will arrive somewhere between the beginning and end of first quarter earnings season, when the trading algorithms start to reweight the weak economic news. If we get a mid-quarter pullback, it might not be so bad. If not, the market could fall by fifteen to twenty percent between the beginning of April and end of June (remember that the debt ceiling comes back up in May). Then we’ll hear a different song. The odd part is that the beat never really changed.

The Economic Beat

The salient drawback in the latest employment number was not that the non-farm jobs total of 157,000 was 10,000 shy of consensus, but that the unemployment rate increased, from 7.8% to 7.9%. Under some circumstances, this could reflect positively on the economy, as the established model is that as the perception grows that jobs are available, people come off the sidelines faster than they can be hired.

That was only partly the case in January. The participation rate, or the percentage of people counted in the workforce, was unchanged, according to the household survey. But it did report growth in the labor force, at 335,000 for December and January combined. It’s sluggish by historical standards, but above the 2012 average of about 108,000 per month. The real problem was that the household survey showed very little job growth over the same period, a combined two-month addition of only about 50,000 compared to the establishment’s survey total of 353,000 (all the foregoing numbers are seasonally adjusted data).

The salient positive was the annual benchmark revision and its net addition of jobs. A month ago, the increase for 2012 in non-farm payrolls was about 1.88 million; now it stands at 2.25 million (unadjusted). The rate of job increase for 2012 is now tentatively pegged at 1.7%, making it equal to 2005, which suffered a slight revision downward. It’s not as good as 2006 and its 1.78%, but close to it, and nothing like the 2%+ increases of the 1990s. Still, it’s progress.

Less positive signs are that manufacturing growth appears to still be sluggish (+4,000), and leading indicators like temporary help (-8,100) and transportation & warehousing (-14,200) fell. The bulk of the gains continue to come from low-wage, low-skill occupations such as home health care, retail trade, and leisure and hospitality. Average weekly hours remain unchanged, and average hourly earnings were up less than two-tenths of a percent.

It’s probably going to continue that way. The unemployment rate for the over-25 year-old set remained at 6.5%, and for those with college degrees, it fell to 3.7%. Unemployment is concentrated amongst the young and uneducated. Anecdotally, one hears that drug use of one form or another is also elevated in the under-25 set, which is perhaps nothing new but does rise in times of economic difficulty. There is a risk of creating a larger, permanent underclass.

Construction employment has accelerated recently, but even at the current rate, it will take six years to reach the level of construction employment from 2006. That was a bubble year, to be sure, and there is good reason not to be back to that level yet. But if it takes until 2020 to reach that level, fourteen years after the peak, it would be presumably be a much smaller percentage of the population and workforce. That’s a lot of people in low-paying jobs until then, and a long wait.

The substantial upward revisions to November (a gain of 86K more) and December (44K) are definitely a positive, but I would be careful about interpreting them as inflections in the trend. Much of their strength comes from the benchmark revisions and adjusted seasonal factors, and is not borne out by either sentiment surveys or weekly jobless claims.

Claims did jump last week, as I predicted they would in last week’s column and more concretely on the eve of the report in Seeking Alpha. The catch-phrase was “quirky seasonal adjustment factors.” No argument there, but I feel obligated to point out that there has been a definite drop in the rate of year-on-year improvement. I was uncertain amidst the Sandy effects, but after three consecutive months, it seems clear now that the year-on-year decline in monthly claims has moved from about 10% to 5% per month. It should be expected that the rate of improvement (i.e., decline in claims) levels out over time as the economy grows, but this is a fairly large drop at what is still a high level of unemployment. It suggests a flattening of the glide path.

One indication of that came in the ISM manufacturing report. The survey reported that the employment indicator jumped, jumped mind you, from 51.9 to 54.0. You could hear the bulls roaring as far as Long Island Sound. The only problem is that employers showed no extra inclination at all to hire more, the result being simply a quirk of diffusion surveys. The percentage of employers planning to increase hiring actually declined from 19% to 17%. The reason for the improvement was that a big chunk of respondents who had said “lower” last month moved into the “same” column. That isn’t a bad thing, but it is by no means acceleration.

The ISM survey itself was alright, but a mixed bag. It improved to 53.1 from 50.2, and the score of sectors reporting growth versus contraction was a good 13-5, which is about what one would hope for in January. But 53.1 isn’t wonderful for this time of year, being the lowest January reading since 2009, and the comments from the respondents weren’t exactly glowing. My guess is that the level of increased activity was subdued. However, the rising price of oil (up eight weeks in a row) led to an improved number for the Dallas Fed survey, to 5.5 with a big increase in new orders (12.2 from minus 1.0).

That is going to cause a problem outside of Dallas. West Texas oil has been climbing steadily toward $100/barrel, fueled mostly by speculation (year-on-year supply is up and usage has been mostly down) and the price of gasoline is at a nominal all-time high for February 1st, coming right on top of the payroll tax restoration. It’s going to hurt spending, which has already shown weakness in weekly retail reports. But the January retail sales report is still two weeks away (and is seasonally adjusted), so we will probably stay “risk-on” until then.

The Chicago PMI was also a positive upside surprise at 55.6, but it benefited from substantial seasonal adjustment factors. New orders, for example, had nearly identical responses as December, when they were reported to be 50.4, but sprinkle a little adjustment magic and they “soar” to 58.2. With the exception of November, the new orders responses have been quite similar for the last six months. Like its national cousin, the tone of responder comments was restrained.

December construction spending rose a reported 0.9%, doubtless aided by the warmest December weather in over 50 years. I wouldn’t look for the trend to continue in January. It should boost the fourth-quarter GDP result, though, a number that would have caused a huge sell-off had it come at the wrong time of year. But this is the first quarter, when trading algorithms are willing to overlook quite a lot. I wrote up an analysis in detail on Seeking Alpha, but if you want a crib sheet, I expect that the number will be revised upward. However, the economy probably really was close to flat for the quarter. The weakening export and import data, if they hold up, are quite revealing as indications of global economic lethargy.

The calendar slows considerably next week, which favors a continuation of the rally. Monday brings factory orders, with the durable goods component already largely known. The ISM non-manufacturing index arrives Tuesday, and I expect that seasonal adjustment will help it “beat estimates” too, though whether it can beat February 2012 is doubtful. Same-store sales data for January come out Thursday, and they could be a challenge, though the sample size is small and suffers from survivor bias.

The European Central Bank meets Thursday, and that could overrule any effect from same-store sales. Right now, stocks are bought on the occasion of every central bank meeting regardless of what they do or say (though that could change if the banks really said anything new). It’s a busy calendar for earnings, but they just haven’t mattered to the broad market. Oh yes, and pay no attention to that man behind the curtain.

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