Debating the Data


“There are lies, damn lies, and statistics.” – attributed to Mark Twain

Memo to Ben Bernanke: Does the latest 7.8% unemployment rate mean that QE is over? The September jobs report set off a firestorm amidst the chattering classes on Wall Street and in Washington, with accusations flying that the White House had cooked the books on the unemployment rate. Nothing new about that – every election October, the opposition party says the government is no good when the data is bad, and up to no good when it isn’t. The intriguing part is who says it. We all know there are people who wear tin foil hats at night to keep out unearthly transmissions – that’s old. But when you see your co-worker or grandmother wearing one, that’s news.

The stock markets did manage to scratch out a reasonable gain last week, in spite of Friday’s one percent plunge from top to bottom before recovering a bit in the last half-hour. But you can see that they’re tiring, the four-day rally that ended Thursday notwithstanding. There were four gap-up opening days, and if you went short on each one at 10 AM you’d have made a pretty penny.

The US economic news managed to beat some estimates here and there, and the ISM purchasing manager surveys in particular were better than expected (see below). But earnings season is fast upon us, starting with Alcoa (AA) on Tuesday and JP Morgan (JPM) and Wells Fargo (WFC) on Friday. The banks might do okay, but estimates have been lowered to the point that earnings growth is expected to be negative in comparison with the third quarter of 2011.

Overall, that’s probably an exaggeration, but it is true that many companies will have trouble showing growth in earnings and/or revenue. Even so, we expect that the usual hit rate of 60-70% of companies beating estimates will prevail.

Of more concern is management guidance. It’s one thing to play the usual Wall Street game of celebrating earnings that beat estimates designed for that purpose, but it’s another when management is out lowering guidance. The pre-season run rate of companies taking down guidance early was about four to one, which is reportedly the lowest such rate since 2009.

Yes, the economy is slowing. We invite you to look at a video clip of Mohammed El-Erian on CNBC, nicely making the case that unless the economic fundamentals start to show serious improvement shortly, then valuations are going to come down to the level of economic fundamentals. It’s a carefully professional way of saying central bank action is one thing, but prices are coming down soon if we don’t get better data in a hurry. It’s also amusing to see commentator Michelle Caruso-Cabrera – who is well to the right politically – desperately trying to get El-Erian to say if anyone reputable at Pimco thought that the jobs report was cooked (he wouldn’t go along, saying that it was mostly trader talk). Take a hat, Michelle.

In the same vein, noted optimist (we use that term tongue-in-cheek) Marc Faber also gave a gloomy assessment of the current data, saying that he expects stocks (and most assets) to fall 20% within six to nine months. Faber is the author of the well-known “Gloom, Boom and Doom” report, so he is not the sunniest of observers, though he did go long on equities in June after the market had corrected.

The more worrisome part about what Faber said is that he thinks that real GDP growth in China is between one and four percent maximum, which is much lower than what the market thinks. Faber has been sitting in Hong Kong since the 1970s, where he has been a close observer of the Asian scene. His outlook may be a matter of opinion, but I cannot recall him making any big mistakes with data and he has not often been bearish about Asia.

All that said, the markets may be able to squeeze a little higher first. We could see a big rally on Monday or Tuesday, depending on what happens with the EU financial minister meeting. If they can cobble together the usual hopeful communiqué about Greece and Spain, the black boxes will send stocks soaring again. The S&P has made several runs at the 1470 level and been rebuffed each time, but the magic elixir of ducking Armageddon has yet to fail the markets. A bailout-inspired push through that level will have everyone programming in at least a run to 1500 before calling it quits.

A delicious, if unintentional irony came Thursday evening when the Spanish finance minister (Luis de Guindos) told a London School of Economics audience that Spain in fact doesn’t need a bailout “at all,” with a straight face no less. The audience couldn’t help breaking out into giggles, but de Guindos seems to think – and presumably his prime minister feels the same way – that the mere threat of EU bond-buying will be enough for Spain to sell all the bonds it wants.

Of course, the EU will not buy bonds if Spain doesn’t play ball with its austerity program, which in turn is crushing the economy, which in turn will make Spain miss its targets again. The same goes for Greece. One has to wonder when the data will catch up with all of them.

The Economic Beat

So, how good of a jobs report was it really? It was modestly good, and really not all that different from the rest of the year. The best illustration we can give you is this: the year-over-year change in actual non-farm jobs – that is, not seasonally adjusted – in September of 2011 was 1.38%. The year-over-year change in September of 2012 was 1.39%. Supposing further that September gets a revision upwards of say, 30,000. Then it would go up to 1.4%. July’s year-on-year change was 1.39%.

We have made the point before – the jobs growth in 2012 has been quite stable. It shows up in both the claims data, the jobs data and the income data (yes, yes, we know, the government is making it all up. And your new tin hat is quite fetching).

The surprise wasn’t in the payroll data of course – at a reported 114,000, it was short of the 140,000 consensus. We wouldn’t be surprised to see it revised back to dead average. It was the unemployment rate and its drop to 7.8% (from 8.1%) that got all the attention, and will no doubt be the subject of much opposition party fuming this weekend (if the polls this weekend hold and Obama does get re-elected while the Republicans retain the House, no doubt the first order of business for the latter will be to set up a committee to investigate the Labor Department).

The unemployment rate is calculated via a different job counting method than the jobs number. The former is called the household number, the latter the establishment number. Basically it’s canvassing households by phone versus checking business forms, with the not surprising result being that there are much bigger swings in the household report. It’s also reported some large drops this year that haven’t seemed to pan out. Over time, the two converge, but on a monthly basis they are frequently apart.

In any case, the big jump in the household report came from part-time work – two-thirds of it, to be precise. When this is lined up with other data, such as the drop in manufacturing and goods-producing jobs in the establishment survey, as well as the decrease in temp hiring, it looks to us like much of the effect is seasonal. September was a good weather month in much of the country, with the added rebound effect of coming on the back of Hurricane Isaac in August. Add in the back-to-school effect, and it explains why this September’s year-on-year change is so close to last September. On a not seasonally-adjusted basis, the unemployment rate actually fell to 7.6%.

On the plus side: besides the big drop in the unemployment rate – which was across all categories – average weekly hours inched up a tenth, and the aggregate payroll index jumped up by seven-tenths. Transportation hiring picked up, and the retail sector added workers for the third month in a row. The direction of revisions for the last two months was positive.

A little more worrisome: the U-6 rate, which measures the rate of people not working full-time who would like to be so doing, was unchanged at 14.7% (seasonally adjusted). Retail is adding fewer workers than it did a year ago. Goods production fell for the second month in a row, and temp hiring was flat to down for the third month in a row. 114,000 jobs is fairly weak, even if it goes get revised up to 140.000; we need about 200,000 to keep up with population growth. The bulk of the job growth was low-income: besides two-thirds of the jobs being part-time in the household survey, about half of the increase in the establishment survey was in the low-paying sectors of leisure-hospitality and health and social care.

One last thing – after revisions, the number of jobs added actually fell (seasonally adjusted) for the second month in a row. There isn’t any breakout going on.

The ISM numbers provided some good news, but again it comes with caveats. After three months of slight decline, the manufacturing reading of 51.5 (50.0 is neutral) is a very mild rebound, particularly in light of the sharp drop in industrial production in August. Some of it is probably price-related. It’s good that we’re not spiraling downwards, but it’s no barn-burner. Better news was that the ratio of growing-to-contracting industries improved from 8-8 to 11-6.

The non-manufacturing report, which is a much larger, though cyclically less sensitive part of the economy, rose to a decent 55.1, with a thirteen-three score of growth-contraction. Prices were up somewhat, but employment readings were weaker. Once again, we feel inclined to point to the weather as playing a role, much as it did in the first quarter. Time will tell.

It didn’t look to us as if there was anything exciting going on in retail sales. Same-store sales were best at discounters, while it looked as if the middle of the market was all about share gains and losses. It ought to be said that there are very few chains reporting monthly numbers anymore. The two weekly reporting services, ISCI-Goldman and Redbook, were at odds again on the monthly increase. Estimates for Christmas have come down. Auto sales looked to be okay, though, so September might have been an average retail month.

August factory orders didn’t reveal anything new since the durable goods report. Excluding transportation, new orders remained down 1.6% and business investment up 1.1%. One thing that is not encouraging is that the level of business investment new orders is down year-on-year. It may well stay that way until after the fiscal cliff issue has been resolved. The Fed minutes didn’t reveal much new either – the market was surely hoping for somewhat lustier sentiments about the new accommodation policy, but participants seemed to have been rather careful and cautious about the whole thing.

Consumer credit rose, in particular with student loans. This is going to come back to haunt us all one day, I fear. Construction spending fell, pulled down by non-residential spending, as residential spending increased.

Europe continues to deteriorate, as does China.

Looking ahead to next week, it’s a light calendar in the US with the bond market closed on Monday. The main action should come on Monday from the meeting of EU financial ministers, who will be trying to deal yet again with Spain and Greece. The Beige Book is out on Wednesday, and the inflation and international trade reports come Thursday and Friday. Europe will be releasing industrial production data throughout the week, with the EU-wide number coming Friday morning. We’ll also see the Chinese trade data before that day’s open, and more weakness would weigh on markets.

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