From page 5 of the BLS report...

The total nonfarm employment level for March 2006 was revised upward
by 752,000 (754,000 on a seasonally adjusted basis). The previously published level for December 2006 was revised upward by 981,000 (933,000 on a seasonally adjusted basis).

That is a staggeringly large miss from their month-to-month reports vs yearly revision. The word BODACIOUS comes to mind.

Makes you wonder why the Fed & others even look at the monthly data - when almost an order of magnitude revision is likely to follow a year later.

Is the BLS feeding them better info that 'we' get?

Dryfly - the large revision is based on the discrepancy between survey data and state tax reports. The last time they deviated significantly was in the runup to the 2001 recession. This can be an indicator of lagging employment in large firms with the slack being taken up in self/small business/contracting employment. However, the discrepancy for 2006 was huge.

Off topic, but relevant: Coast Bank is up for sale. For those less familiar with the banking world, most banks unable to meet their capital reserve requirements are sold these days:
"Thomas believes that it was likely the federal agency that encouraged the bank's Bradenton-based parent, Coast Financial Holdings Inc., to hire investment banking firm Sandler O'Neill & Partners to explore options."

This is one I am watching carefully, because this bank was, I believe, started with a sale in mind. If it does not sell quickly, it will be indicative of banking nerves. I'd hate to be on the due diligence for this one!!!

CR, every month you've shown your first graph before, the red line was right in the exact middle of the grey wedge. This time it's above it. Have you dropped the extrapolation lines?

One thing that concerns me a bit about this report is that household survey only showed 31K new jobs.

The nonfarm payrolls report has much more clout, but the household survey has been more accurate the past few years. These latest revisions to the past nonfarm payrolls bring them closer to what the household survey has been saying all along.

I'd hate to be on the due diligence for this one!!!

I, on the other hand, would volunteer with a great deal of enthusiasm. As long as the "due diligence" in question involves the examination of the past examinations of this outfit. I would be exceptionally interested in discovering just how the greenest of greenhorn FDIC examiners missed the fact that this bank was, in essence, "disguising" a huge commercial spec construction loan as a series of "residential construction loans" in order to get past concentration limits and loan committee approval, not to mention elementary underwriting of the builder. I mean, that scam has grey hairs on it. A residential construction lending program that does not track and board-report on builder concentration, while allowing "No Doc" loans to the "borrower of record"? That didn't smell funny?

It just blows me away that we all sit here worrying about exotic derivatives and Star Trek-quality VAR models, while Coast pulls off an oldie from the low-tech S&L days right in front of the FDIC. Then, suddenly, the FDIC "remembers" the lesson of the S&L crisis, which was to put those undercapitalized little time bombs on the market. You're certainly right that this is all going to tell us something about "nerve."

Tanta,

Coast may be the thin edge of the wedge in Florida. So the boys from Lincoln decided to buy construction for 2nd homesloans via brokers in Cape Coral. Now they have 90+ delinquents at >10% of disbursed funds (and presumably rising). The gory details:

LINCOLN, Neb.--(BUSINESS WIRE)--TierOne Corporation (NASDAQ:TONE - News; "Company"), the holding company for TierOne Bank ("Bank"), today released additional information regarding the status of a group of its residential construction loans in southwest Florida.

The Bank's southwest Florida residential construction loans are located primarily in the Cape Coral area of Lee County. Each borrower possessed a strong credit score which met Fannie Mae or other secondary market underwriting guidelines and had also obtained a contractual commitment for permanent financing with third-party lenders upon completion of the residence.

Working with a local Florida-based mortgage brokerage firm, the Bank acquired numerous nine- to 18-month residential construction loans for individual homebuyers wishing to build a second or retirement home in the Cape Coral area. No Cape Coral loans have been acquired since December 31, 2005. None of these residential construction loans were made through the Bank's loan production offices which specialize primarily in real estate and construction lending to building contractors, not individual homebuyers.

A substantially higher than expected surge in residential construction building permit applications occurred in the Cape Coral area in early 2006 and resulted in unusual delays affecting the commencement of construction. In many cases, these delays extended beyond the original term of the residential construction loan. This backlog of residential construction permits awaiting issuance has substantially improved and the City of Cape Coral's permit issuance process has now returned to its normal four- to six-week time frame. As a result, several residential construction builders are currently fully engaged in actively constructing residential homes for the Bank's borrowers.

At December 31, 2006, residential construction loan commitments in the Cape Coral area amounted to $144.9 million with disbursed funds totaling $84.2 million. The Bank's Cape Coral loans have continued to decline as borrowers pay off loans. Loans 90-days or more delinquent totaled $9.4 million at December 31, 2006. Based on current information, the Bank believes it has sufficient general reserves in its allowance for loan losses.

Expired

CR, I would be inclined to go with your #3 but, of course, that's speculation on my part.

Re: Limited losses in residential construction jobs

I know you’re all going to want to call me a heretic for this, but maybe housing is stabilizing. It’s sure starting to look like things have leveled off, at least for now.

Also, ECRI now says a recession is improbable for 2007. I can’t say I’m seeing much to disagree with them.

Dryfly,

First, understand what I mean when I say stabilizing. I do not mean real estate has reached a bottom and is ready to resume it's assent to the heavens. Real estate is still at a level that is unsustainable in the long-term, but if the economy remains relatively healthy over the year, I think the level real estate is at may be sustainable in the short-term.

Take a look at CR’s REI as % of GDP chart:
http://bp3.blogger.com/_pMscxxELHEg/RcEGWge-4GI/AAAAAAAAABE/Nu4VOp6NCPs/s1600-h/RIGDPQ42006.jpg

All bottoms have been accompanied by a recession. But if we don’t have a recession this year, what’s to say we see the bottom of real estate this year or even next year?

Take another look at the chart and focus in on the 1981 area. I wouldn’t be surprised if we see something like that happen again. RE goes on hold for a bit, maybe even ticks up a bit, but it’s just treading water. As long as the economy is relatively healthy (say, 2%+ growth), I think the market will survive.

Once the recession comes, 2008 or 2009 perhaps, the other shoe drops.

vader I've lived in a number of towns with large populations of undocumented workers... you have no idea how low they can go before they throw in the towel. We haven't seen that kind of determination since Hooverville.

And most do have resources - large extended families. One job can be a toe hold for 2-3 families for an indefinite period of time.

If you doubt me check out a latino grocery & see how much bags of rice & pintos cost. Throw in some tortillas, some dried anchos, guajillos and or pasillas for mole and it isn't a lot different that back in Jalisco or Oaxaca.

Compare that with the average American worker with a large home, SUV, wife & kids who expect more than rice & beans and a two families in a two bedroom... and see who hangs in there longer.

My bet is with the illegals unless INS gets serious.

You periodically bring up this chart that tracks nonfarm payroll job growth during the Bush presidency. The data suggests average to above average job growth, but the question is, what "types" of jobs are being created? For example, are the new jobs more low paying than the historical average, what industries have supported the most job growth, and which the least, etc. I suspect the job growth in the last six years has created a higher percentage of low-paying jobs (after inflation) with even less stability and benefits. If this is correct, I see no point in trumpeting (not that you're meaning to, but it could be interpreted as such) job growth under Bush if in fact what has happened is an insidious expansion of the lower middle, or "working" class (the group that stradles the 4th & 5th quintile), and record gaps between the very rich and the very poor. I'm hoping you can post some data that reflects on, and puts into perspective, the true underbelly of the entire employment infrastructure, particularly during the Bush years (actually, this recent phenomenon goes back three decades; but I believe it's simply been exacerbated under Bush).

Calculatedrisk: "...This is just the beginning of the loss of several hundred thousand residential construction jobs over the next year or so..."

Doesn't anybody but (anal-retentive) me actually look at the data before they post, just as a cursory reality-check so they don't sound...less informed than they should be? There are only about 975,000 or so TOTAL salaried residential construction jobs. So what's the thrust here, that virtually every residential construction worker is going to be out of work?

In past recessions the only way construction job-loss reached the multiple hundreds of thousands was when ALL construction, residential and non-residential, was included. (And the number of non-residential construction jobs is several times the number of residential, BTW.)

I'm all for contrarianism, listening to your own drummer, ignoring the "established wisdom", etc. But only if the facts support that minority view. Which they don't, in this case.

Sebastia

Aaron,

You make some good points, and thanks for the data links. So, many of the new jobs created are volatile or transient in nature, and in the case of healthcare, many are lower paying (e.g., geriatric aides). I suspect there has also been a significant transformation out of the manufacturing component and into the typically lower paying service sector (e.g., retail). I'd love to see some data on that.

Name, I think you mis-directed your comment.

"..Sebastian, CR is talking about residential plus residential specialty trades, BLS series CES2023800101 and CES2023610001. Combined employment as of end Jan 2007 was 3,330,500 (preliminary)..."

Thank-you for pointing that out, although it would have been nice if he'd said what he meant.

I do have another question, though. Upon what is he basing his opinion that there will ultimately be a loss of several hundred thousand jobs in this sector? That's a recession-level number, with no evidence of recession on the horizon. (Evidence with proven historical causal relationships, I mean.) TIA for any response.

Sebastian

"soon soon" should read "should soon"

Sebastian,

As to your first question, prior to 2001 the two BLS series were consolidated. From 2001 on, the Bureau of Labor Statistics separated the old series into the two current series. The old series is comparable to the sum of the two new ones. The long-term historical correlations between permits, starts, completions, and employment have been studied exhaustively.

Historically, housing starts are considered a leading indicator of recession. Employment, on the other hand, is a lagging indicator. You won't see unemployment go up until after a recession has begun.

You are correct in saying there is no evidence of recession, but a number of leading indicators, housing being one, are suggesting that a recession is possible, if not likely.

Sebastian, I would recommend that you use seasonally-adjusted data, but that's OK. Personally, I use unadjusted data only for cycle-to-cycle comparisons. Having said that, there is no such thing as an infallible indicator, but housing is one of the best. Housing did not lead the last recession because that one was investment-led. Housing signalled in 1966, but there was no recession because of the big fiscal stimulus (Viet Nam). It picked up the other six. You have to learn how to use the indicators.

As far as Wright is concerned, I assume you're looking at his Model B, and you are correct. However, B is perilously close to signalling. Last time I looked, his Model A was already signalling. The Estrella-Trubin model is also very close. The 10-year treasury/bond-equivalent 3-mos term spread has been signalling for some months now and it is the gold standard. That term spread is also highly correlated with future GDP growth. The NAHB-Wells Fargo index, also signalling, is correlated with future consumption growth and highly correlated with starts. Also, look at the flow of funds, particularly the leveraging-up in the household sector. Manufacturing is also in the tank and no one is investing in capital equipment. No investment, no growth.

What's being signalled? Low Growth. Forecasting a recession is a sucker's bet because recessions are dated by the NBER. What we're looking at now is a growth slowdown that could turn into a recession. CR made the call for a 2007 recession in January, as I did, but I'd never do it on the record, although I did forecast a slowdown over a year ago.

Personally, I'm looking for a currently evolving credit event in the household sector to tip the economy. It's CR's and Tanta's "specialty". Study CR's archives, they'll answer a lot of your questions.