If you've been reading our analysis, you know we've been of two minds for about a month now. On the one hand, we've been warning that the rally has gotten long in the tooth and that the economy, while doing moderately okay ("moderate" - that's the word the central bank keeps using, isn't it?), isn't as good as the hype. Yet we've steadily acknowledged that the markets could keep climbing higher, as momentum is a powerful thing; more below.
Tuesday, though, kind of looked like the bell-ringer day. Admittedly, 1425 does beckon on the S&P 500 chart as some sort of testosterone-fueled summit. Yet the silliness of the last week clearly points to a market top.
The sharp turnaround in the middle of last week was set off on Tuesday by a Wall Street Journal story that speculated on what the Fed might do if it decided to embark upon further quantitative easing. We thought the headline was odd because Fed governor Richard Fisher had rebuked the markets only a couple of days earlier for being too focused, if not addicted, to thoughts about more Fed easing.
While Mr. Fisher is certainly one of the leading hawks at the Fed and not representative of the more dovish majority, it certainly appeared to be bizarre timing for a governor to say "stop looking for more money" one day and the central bank to be promising more within hours.
But it hadn't, as it turned out, and the author of the article found himself doing a second video interview soon afterwards in which he stressed that it was only a hypothetical rumination. He mildly complained about people not reading the whole story, and perhaps justifiably so, but it's hardly a new development that traders pull the trigger on the headline first and read the story later. For its own reasons (hello, Mr. Murdoch?), though, the publication first ran the headline as a fait accompli: "Fed Officials are considering a new type of bond-buying," and only later (about midnight that day) added the "if."
The second stage of last week's rally came on the astonishing development that once again Greece failed to default, bringing to mind South Park's comedic device of feigning mock horror in every episode over the death of Kenny. The unfortunate part for the markets is that one, Greece's problem - and by extension, the eurozone's - hasn't really been solved, only postponed again, and second, with no impending default to avoid, what is going to drive the market higher? Heaven forbid we should start to pay attention on the deteriorating economic data coming out the eurozone and China.
Well, the markets did find a new default worry to overcome, one it didn't even know it had. After first rallying on the astonishing news that German investor sentiment improved - coming in the face of a near-15% rally in the DAX already this year, that must have seemed like a real turning point - then again, on retail sales coming in as expected, markets went ballistic in the last hour of trading when it was leaked that the major US banks aren't going to flunk the stress tests that no-one was even thinking about. Apple (AAPL), which in case you didn't know is this year's ETF for extra beta, tacked on an additional one percent on the news. Apparently all of JP Morgan's (JPM) shareholders are going to spend the bank's dividend increase on buying iPads.
The reporters write their stories about investors being cheered by the strengthening economic data when what is really happening is trading money playing the momentum rally. Last week much attention was given to the catch-up rally in small caps. A sign that the rally is for real, said pundits. Me, I wondered who crossed a million-share block of IWMs (Russell 2000 ETF) Friday morning, to go along with a bushel of six-figure blocks that day. I don't know about you, but I'm thinking it wasn't the Magellan fund. Think someone might have been leaning on the tape? It might have been the same boys and girls that punched up a 1.3 million QQQs (Nasdaq) ticket late Tuesday, minutes after the JP Morgan dividend story crossed the wires. That's real money.
The challenge for investors remains the same: the markets are overbought and prices supported by agnostic money that has no commitment to anything but the momentum trend. Against that, that money will stick around to dally with the trend so long as no story appears capable of breaking it, and with April looming (the market's favorite month), it is just possible that we could bridge the intervening period without a correction.
But after Tuesday's action, it's hard not to think that it was the top, at least for a few weeks. Traders like to say, worry about today's move today and let next week worry about next week, but the flip side of any bender is that the bigger it gets, the more painful the payback headache is afterwards. The jobs report isn't as good as it appeared, and neither is the economy here or around the globe.
In fact, through the first full week of March, there isn't much indication that this quarter is any different from the first quarter of 2011. We have studied the February jobs report at some length, probably more than we should for our own health, and there were some nice aspects: temp employment increased, the December and January revisions were positive, the January number was the best in some time and at 227,000, the February report was mildly ahead of expectations (though not the whisper number of 230k).
However, there was something to worry about too. As much as we would like to say otherwise, it looks to us like the base case is that employment growth is peaking, and may already have peaked. Never mind that manufacturing job growth fell sharply, or that more than half of the new jobs were low quality: temps, health care, leisure and hospitality. After all, it's only one month.
No, we looked at the raw and adjusted monthly jobs data over the last ten years. The growth in January and February 2012, on a seasonally adjusted basis, are the best for those two months since 2006. That sounds great, except that 2006 is when job growth peaked in the last cycle. Looking at the unadjusted data, the monthly changes for January and February are nearly exactly the same as they were in 2006.
More worrisome is the year-on-year pattern. We are coming off one of the warmest winters ever, yet the February 2012 number is almost exactly the same as February 2011. Where are all the extra jobs? One can reasonably argue that, given the recent direction of revisions, the number will improve with time, but it's going to take a substantial revision to catch up February 2012's private payroll change (+233k) to that of February 2011 (+257k).
Turning back to the adjusted data, the percentage changes in payrolls for February 2011 and February 2012 were identical at +0.17% (the unadjusted numbers virtually so), while the January numbers were very close. If the expansion is finally gaining traction, shouldn't the numbers have been a lot better with one of the warmest winters on record, and virtually no workdays lost? Looking at the last post-recession period of 2003-2007, one can see that job growth peaked in 2005 and 2006, or years three and four. Comparing 2003-2006 to 2008-2012, the comparable numbers are lower in the latter period, and the unemployment rate is in fact higher. 2006 had a better pickup over 2005 than 2012 has over 2011. This recovery is weaker.
Employment is a lagging indicator, a fact that is often repeated ad nauseum when job growth is negative (the implication being that the worst is already over) or weak (the best is yet to come), but conveniently forgotten when job growth is showing relative strength. A better leading indicator is weekly claims data, and we also looked at monthly claims totals for the last ten Februarys.
February 2012 showed an improvement in actual (that is, not seasonally adjusted) weekly claims of approximately 150,000 over February 2011 (we used the first four calendar reporting weeks of February in our analysis; using the last four instead had little effect). This compares with the approximately 327k year/year improvement in February 2011 and the 655k improvement in February 2010. Looking at the third and fourth recovery years of 2005-2006 again, the improvement was about 194k in February 2005 and only 78k in 2006. The comparison turned negative in February of 2007: the year-on-year increase of 133k in claims was an accurate harbinger of the deteriorating economy.
Two disturbing inferences from this data are that one, the 150k improvement in 2012 looks very much like the 78k difference in 2006, once weather is taken into account. Second, the recovery in the middle of decade was being fueled by the housing bubble and its debt-fueled consumption; nothing of the sort is even on the horizon right now in terms of propping up demand.
If you're thinking autos, think again, because the recovery in auto production is nearly complete. An increase to a 16 mm/yr selling rate appears reasonable to us this year, yet we are nearly there already. The lion's share of the related employment gains are most likely behind us. If you're looking for bubbles, then there is a bubble in corporate bonds that should be supportive to corporate balance sheets, but little of that is going to make its way into stimulating end demand. What isn't tied to debt rollover is largely earmarked for share buybacks (enhancing executive compensation) and productivity investments.
This isn't at all a call for impending recession, only supportive of our view that the economic progress has been modest this year, especially in comparison with many of the headlines. On Thursday, the Wall Street Journal ran a full-page story on slowing growth in emerging markets, but deep within the main section; on Saturday, the page-one headline was "Jobs Recovery Gains Momentum," despite this February being no different from last. The market is setting itself up for disappointment.
We would also add that February job growth isn't representative of the year to come. Eventually the data gets re-benchmarked so that the monthly data rates better fit the underlying annual trend in real job growth, but until then, the maturing arc of the recovery and lagging nature of jobs growth means that the February growth rate runs ahead of the underlying annual trend in real time. In both 2010 and 2011, February hiring rates ran well ahead of the actual rate twelve months later (this is aggravated by February being one of the strongest hiring months of the year in real terms). Absent a major new development, 2012 job growth is not going to reach a new level of escape velocity, rather it is going to look a lot like 2011. It may even soften later in the year.
One of the reports that cheered markets and pushed the envelope on fatuousness in reporting was the consumer credit report for January. It's necessary to take the preliminary results with a liberal dose of salt, because the revisions tend to be large and, due to the use of quarterly data, the final number isn't really accurate for a few months. We felt some mighty wind reading some of the glowing talk about the consumer being back, but the first estimate suggests that in fact, consumer credit card usage fell. The increase in auto loans is an estimate derived from the increase in auto sales, so it shouldn't have come as a surprise. The disturbing part is that more than the entire increase came from growth in student loans, which are increasingly thought to be in a bubble state.
As for retail sales, we ran the February data there too, and here's what you need to know about the "strengthening consumer:" the year-on-year increase in actual retail sales excluding gas and autos - that is, not adjusted for anything - from February 2010 to February 2011 was 5.61%. The year-on-year increase from February 2011 to February 2012 excluding autos and gas, was - after adjusting for the extra leap-year day - 5.62%. Some difference. In fact, the year-on-year rate fell for total sales, from 9.2% to 6.5%.
Yes, we know, every morning the trade has been that if the world hasn't ended, Apple is good for another five bucks at the open. And around 10:30 AM, some big guns will come in and start throwing around some money. But friends, these good folks aren't buying anything. They're renting. Keep that in mind.
Additional disclosure: Certain managed accounts may have put positions in Apple and IWM.