More on NFP: More Recognition of Disbelief

Wednesday, May 09, 2007 | 06:49 AM

The WSJ's Marketbeat reported yesterday that the latest analysis to join our skepticism on NFP is none other than Goldman Sachs:

"Second-guessing government statistics is gaining more credence as time goes by. Greg Ip over at Washington Wire has this post, detailing how economists at Goldman Sachs believe the Bureau of Labor Statistics might be overestimating the number of jobs created in new firms, thus making employment growth look stronger than it is. It’s a point bloggers such as Barry Ritholtz have made as well."

Let's take an even closer look at the NFP details then we did on Friday.

The initial read of the April Employment Report was rather punk -- Payroll growth at a 2-year low, Unemployment down, but sarting to tick higher -- but the devil is always in the details. This report gave Lucifer plenty to play with.

We already mentioned the BLS Net Business Birth/Death Model wholly created a record breaking 317,000 jobs --  the largest amount of monthly jobs ever created from the model -- out of thin air. There simply is no justification for it. But we are not alone in that criticism:

John Williams Shadow Stats has an explanation for the unreasonable jump in B/D jobs. “Based on prior-period revisions, skewed seasonal factors and an enhanced bias factor, the April data appear to have been massaged so as to avoid showing a monthly payroll contraction. Reversing the pattern of most recent months, the Bureau of Labor Statistics (BLS) reported major downward (instead of upward) revisions to  previously published data, and seasonal factors became skewed again. Where the monthly year-to-year  changes in both the seasonally-adjusted and unadjusted series should be the same, applying the base unadjusted growth rates to the adjusted series would have reduced the monthly jobs gain from 88,000 to just 32,000.” 

Goldman Sach's Jan Hatzius writesNonfarm Payrolls: Weaker than Reported? The Labor Department’s ”birth/death model”—which imputes net job creation by new firms in the monthly nonfarm payroll figures—may currently be causing a significant overstatement of employment growth. (1) Conceptually, the danger of overstatement is greatest when the economy is slowing, as it has in recent quarters. (2) On a seasonally adjusted basis, employment birth/death adjustments have been unusually large recently, perhaps because the model “concluded” from the large upward 2006 benchmark revision that new firms were being created at a faster rate. (3) The quarterly Census of Employment and Wages (QCEW) suggests that true employment growth had already started to trail the official estimates as of Q3 2006. (4) The household survey of employment has weakened considerably in recent months.

• Q1 GDP growth was 1.3% -- a fraction of 2006 GDP -- compensation declined 1.5% in Q1; If Housing is in such a horrid slump, and Manfucturing has been slowing, why on Earth does BLS think small businesses created the most jobs ever for a single month in April? How did an unfathomable 49,000 construction jobs get created (via B/D), when the Net Construction jobs actually measured by BLS  -- not merely hypothesized -- fell 11,000?

Merrill’s David Rosenberg: “A point to note: fulltime jobs, the key generator of personal income growth, plunged by 687,000 in April. That was the largest slide since the economy was knee-deep in recession  in August 2001. Such a decline has only occurred three times in the history of the Household Survey going back to 1968.”  Mr. Rosenberg notes that “the employment-to-population ratio sagged to 63% from 63.3%....The last time the employment/population ratio fell that much in one month was in October 2002, when the Fed was consumed with deflation fear and was on the precipice of cutting the funds rate two more times.”

• A friend emails me this:  "For years certain permabulls screamed that the Establishment Survey (CES) understates job growth and the Household Survey (CPS) is a more accurate depiction of the economy.  For April the Household Survey shows a job LOSS of 468k! For 2007 the Household Survey has a net LOSS of 140k jobs. The Household employment adjusted to Payroll methodology (excludes the self-employed and counts each job a multiple-job holder has) fell by 70K."   

• BLS continues to understate the unemployment rate, as it has been doing for most of the past 6 years,  by reducing the ‘pool of available workers’.  BLS reduced the labor force by 392,000 in April.  This kept the Unemployment Rate from jumping higher than 4.5%. The ‘participation rate’ declined to 66% from 66.3%, which is an unusually large monthly decline. That's 0.3% times 143 million or so workers -- instead of rasisng the unemployment rate, we lower the labor participation rate by 429,000 workers.

• BLS continues to overstate actual employment -- From BLS: “Household survey. The sample is selected to reflect the entire civilian noninstitutional population. Based on responses to a series of questions on work and job search activities, each person 16 years and over in a sample household is classified as employed, unemployed, or not in the labor force. People are classified as employed if they did any work at all as paid employees during the reference week; worked in their own business, profession, or on their own farm; or worked without pay at least 15 hours in a family business or farm.  People are also counted as employed if they were temporarily absent from their jobs because of illness, bad weather, vacation, labor-management disputes, or personal reasons.”

In other words, if you “Worked without pay,” the BLS Household survey considers you gainfully employed.   Beyond absurd . . .

Casey Weldon notes, “Over the last few years, a month-month decline in Self-Employment has become a very RARE occurrence.  The only other month-month decline posted in the last YEAR was in Dec-06.  The number of Self-Employed in the US contracted by (-) 24,000 in April.”

Wednesday, May 09, 2007 | 06:49 AM | Permalink | Comments (30) | TrackBack (1) add to | digg digg this! | technorati add to technorati | email email this post



TrackBack URL for this entry:

Listed below are links to weblogs that reference More on NFP: More Recognition of Disbelief :

» WSJ: Overstated Job Market Strength? from The Big Picture
We have long stated that the official BLS data was a rather artificially optimistic assessment of the actual employment picture. Our thesis, supported by many confirming 3rd party data, points to a bifurcated recovery since the 2001 recession. Indeed, ... [Read More]

Tracked on May 24, 2007 7:33:03 AM


Government statistics are bunk. When they became unreliable and why is the story.

Posted by: Tom C., Stamford,Ct. | May 9, 2007 7:52:29 AM

The comments to this entry are closed.

Recent Posts

December 2008
Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      


Complete Archives List



Category Cloud

On the Nightstand

On the Nightstand

 Subscribe in a reader

Get The Big Picture!
Enter your email address:

Read our privacy policy

Essays & Effluvia

The Apprenticed Investor

Apprenticed Investor

About Me

About Me
email me

Favorite Posts

Tools and Feeds

AddThis Social Bookmark Button

Add to Google Reader or Homepage

Subscribe to The Big Picture

Powered by FeedBurner

Add to Technorati Favorites


My Wishlist

Worth Perusing

Worth Perusing

mp3s Spinning

MP3s Spinning

My Photo



Odds & Ends

Site by Moxie Design Studios™