NFP WTF ?
As mentioned yesterday, thanks to a dental crown install, I missed much of the fun (at least I had a steady supply of nitrous while in the chair). I saw the data, posted on it, and ran out the door.
Let's get into the details of what was by most traditional measures, a mediocre NFP report:
• Private sector employment growth was stagnant.
• Health care, and bars & restaurants were responsible for more than half the rise.
• Temp jobs, usually a good leading indicator of future job creation, fell 20,000.• Government was responsible for nearly all of the upward revisions to the July and August;
• Manufacturing continues to lose lots of jobs quickly. (Negative 60,000 over the past two months).
• Retail hiring is trending downwards;
Having been out of the loop all day, I am somewhat perplexed over the subsequent reports about this NFP report. I cannot say I understand why people are outright gushing over this otherwise middling number:
The Jobs Machine (WSJ)
Job Growth Looks Rosier, Easing Recession Fears (NYT)
Jobs: A September Shocker (BusinessWeek)
Talk about the soft prejudice of low expectations: These headlines are over a mere 110,000 jobs. There was obviously relief that this wasn't even a worse number. But understand the details: This is in no stretch of the imagination a solid month of job creation. This was weak NFP growth, failing even to keep up with population growth.
To review: Any report under approximately ~150k month (subject to revisions) is weak. It means that job creation is failing to keep up with population growth. Thanks to the shrinking Labor pool, this has not had much of an impact on the Unemployment Rate, which ticked up a mere 0.1% yesterday.
Let's put these data points into some context average monthly job creation:
- In 2006 was 226,000 new jobs created per month
- In 2007, that number fell to 122,000;
- In Q3 2007, that number fell to 74,000.
Even more damning has been the Birth/Death adjustments. In the 1990s, these were relatively insignificant. In 2001-02, it was less than 10% of total NFP each year. After a 2001 change to the format and weighting (effective in 2003), the B/D contribution to NFP went up dramatically. From 2003-06, Birth Death adjustments contributed between 35-41% of the total non-farm payroll job creation.
Then there is 2007. If we take this year's B/D adjustment versus NFP, we get a B/D contribution that is in excess of 76%. (Trailing 12 months B/D versus 2007 NFP data annualized).
Strong? Shocker? Jobs machine?
Hardly . . .
UPDATE: October 7, 2007 7:32am
I (belatedly) see the NYT's Floyd Norris reached the same conclusion in yesterday's Times:
"JOB growth in the United States has slowed significantly over the last year, a fact that renews questions about the ability of the overall economy to withstand a slowdown in housing . . .
The job figures each month are based on a survey of employers, but the numbers released are not exactly what the survey shows. Instead, the Labor Department adds or subtracts jobs, based on estimates of how many people found work with newly created companies, less those who lost jobs at companies that went out of business. The figures come from what the department calls its birth-death model.
The figures are eventually corrected based on data on the number of people covered by unemployment insurance, which is more accurate but available on a delayed basis.
At turning points in the economy, the birth-death model figures can be considerably off. If the economy is strengthening, as it was in recent years, the result can be an underestimation of job growth. Indeed, a year ago the government said it needed to add 810,000 jobs for the year ended in March 2006. But over the following year, the estimates were too generous.
Over the last 12 months, most of the reported job growth has come from the birth-death model, not from the actual survey of employers. If the model overestimated employment by new companies, then the figures for that period are likely to be revised down a year from now. . . ."
Monthly Data Aside, Trend Is Weak
FLOYD NORRIS
NYT, October 6, 2007
http://www.nytimes.com/2007/10/06/business/06data.html
The video below is an example of a somewhat gushing review. (I am obviously not seeing what everyone else is)
Sources:
charts courtesy of CHRIS PUPLAVA
Historical Net Birth/Death Adjustments
BLS, Department of Labor Statistics
http://www.bls.gov/ces/cesbdhst.htm
CES Net Birth/Death Model
BLS, Department of Labor Statistics
http://www.bls.gov/web/cesbd.htm
The Jobs Machine
WSJ, October 6, 2007; Page A20
http://online.wsj.com/article/SB119162827290050813.html
Job Growth Looks Rosier, Easing Recession Fears
MICHAEL M. GRYNBAUM
NYT, October 6, 2007
http://www.nytimes.com/2007/10/06/business/06econ.html
Jobs: A September Shocker
Michael Englund
BusinessWeek, October 6, 2007
http://www.businessweek.com/investor/content/oct2007/pi2007105_921337.htm?
U.S. Economy Down, Not Out
Overall Health Is Seen In Face of Deceleration;
Job Gains Lift Optimism
KELLY EVANS
WSJ, October 6, 2007; Page A3
http://online.wsj.com/article/SB119158634981050085.html
Saturday, October 06, 2007 | 08:22 AM | Permalink
| Comments (43)
| TrackBack (0)
add to de.li.cious | 
digg this! | 
add to technorati | 
email this post
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c52a953ef00e54ef1a47f8833
Listed below are links to weblogs that reference NFP WTF ?:
Comments
I am glad I sold all my stocks at the end of July. Look out below!
Posted by: The Dirty Mac | Oct 6, 2007 8:27:06 AM
Who said anything about stocks? I am discussing Job creation.
I have noticed that whenever the market rallies hard, the uber-bullish contingent is in full throated voice over anything and everything. Herb Greenberg calls it his email hostility-o-meter.
Its a fascinating psychological element that needs to be quantified. I need to figure out how to extract the bullish chest pounding as a contrary indicator for ST trading reasons . . .
Thanks, Dirty Mac, for your input.
Posted by: Barry Ritholtz | Oct 6, 2007 8:40:00 AM
Barry,
I wrote about this yesterday as well as everyone started the party:
http://www.urbandigs.com/2007/10/jobs_report_fed_thoughts.html
However, my numbers tell me:
2004 = 172,000 jobs per month created
2005 = 212,000 jobs per month created
2006 = 189,000 jobs per month created
*2007 = 122,000 jobs per month created
*includes monthly jobs data up to September
2006, being the difference from your # and mine. I used this for stats:
http://data.bls.gov/PDQ/servlet/SurveyOutputServlet?data_tool=latest_numbers&series_id=CES0000000001&output_view=net_1mth'
Anyway, this is further proof the US economy is a mature economy, and we are seeing decelerating jobs growth. The emotional element at play was that this was a mediocre number, not awful not great. With revisions, who the hells knows whats going besides the fact that job growth is decelerating, the one thing we can decipher.
As for the fed, any additional rate cuts besides one more 1/4 cut, will be the fed's way of telling us the problems under the surface are NOT going away and economic weakness is obviously coming and is priority, which they clearly spelled out with the last move. I think we have at most 1/4 cut left, and I hope we dont even get that!
Since when is 4.75% fed funds rate restrictive?
Posted by: UrbanDigs | Oct 6, 2007 9:07:10 AM
Up is down. Remember when we were told that the housing market recieved "good news" because new home sales were up 2.8% in July 2007. All the media was going on about how this was a really positive thing for the housing market. Of course that was an increase within the margin of error and was being compared to the previous month not YAG and the previous month had been a disaster. I don't think anyone is still yammering about the good news about the housing market like they were a month and a half ago. Perhaps in a few months we will also look back at this "good news" and have the same sort of hindsight?
In other "good news" at least americans are using up their plastic.
If i believed in Santa and was in the market for things made in China, I'd be going shopping this xmas as there should be some great year end deals.
I also think the decoupling idea is sound for those US based companies that are truly international in scope. Can someone get us info on how many US companies are truly international and how many are US based? The number of companies depending mostly on the health of the US consumer must be pretty high. If they all get hit hard can the decoupled giants save us all? Or will they not give a damn as they sit in their Dubai headquarters.
Posted by: 12th percentile | Oct 6, 2007 9:22:03 AM
Fascinating - so how much nitrous got loose in the trading rooms and talking heads greenrooms ? Looks like a lot. Very nice dissection and breakdown. Your key question and comment above is dead on though - how to judge this in terms of trading opportunities ? Any thoughts ? Back to your four factors (fundamentals, technicals, psych and sentiment ?)
YoY the number is down to 1.19% growth which continues and accelerates a downtrend that begin in Jan06. But as it gets worse and the slowing gets worse it's no longer a slowing is it ? FWIW when you take your 150K figure as a "head-above-water" bogie and net it out against average new job creation we've not only not recovered from the downturn but made no new net progress since Apr04. In fact running back 'til Jan00 when Dr. Plangloss was the house we're six million jobs in the hole. Boy does that explain a lot.
So are we on Goldilocks 2.0 or what ?
Oh BTW - the charts backing up the data and a little discussion are at http://tinyurl.com/327bdt
Posted by: dblwyo | Oct 6, 2007 9:33:35 AM
I thought these numbers (employment) were ALWAYS considered a lagging indicator???
Besides being spun junk, don't companies max out on layoffs AFTER the consumer pulls back? How about an analysis of consumer confidence, the Fed, and house values dropping? Hey Barry, I even have a title for that Blog entry:
THE CONFIDENCE GAME!
cheers to all.
Posted by: lurker | Oct 6, 2007 9:49:20 AM
Goldilocks 2.0? Larry is that you?
I dont think any of the #'s yet include the ramifications of this credit squeeze! Its not like a company comes out with an announcement and then the next day its executed. While its good that companies are coming out with this news to add transparency to a marketplace that hates uncertainty, we still dont know the ultimate results of all this.
Its like we are in the eye of the hurricane. Will the 2nd half pass or not? Looking back, the fed's move was to cushion the blow when it does hit, it was preventative!
This jobs report was not that great, and the markets are treating bad news as good, a sign of a bullish market. Now, what shakes that trend is what I dont know. If you see these corporate earnings/layoffs/restructuring last beyond the next quarter, there will be some problems.
But keep in mind that upcoming data will show the result of what we went through!
Posted by: UrbanDigs | Oct 6, 2007 9:56:41 AM
It seems to me there are two basic approaches to analyzation: the scientific method versus the political method.
The scientific method begins with the raw data and developes a hyphothesis that incorporates all the facts.
The political method begins with an agenda and then emphasizes corresponding facts while ignoring conflicting facts.
When Wall Street was begging for rate cuts, we had downward revisions and a -4K current number; with Wall Street fearing more bad numbers, we had upward revisions and a better-than-expected current number.
I have real concerns that the statistical agencies have been politicized to the degree that agenda supercedes accuracy.
Posted by: Winston Munn | Oct 6, 2007 10:14:44 AM
Urbandigs..
i really dont believe in conspiracy theories....but it does look very convenient that jobs fell drastically when the FED had to cut rate....and then they got revised by almost 800%.
i wonder how hard it is to cook numbers at the direction of administration......just like books are cooked to show good balance sheet
Posted by: techy | Oct 6, 2007 10:20:46 AM
Best breakdown of employment numbers that I have seen so far:
http://www.financialsense.com/Market/wrapup.htm
Posted by: Justin | Oct 6, 2007 10:28:42 AM
I am not defending, because I think this is the worst job market on my lifetime, Ajay Kapur, founder of First Horse Capital, a Hong Kong hedge fund, in the FT yesterday said "The risk of a recession is from slowing consumption, but is very modest.
The bulk of consumption is driven by the rich - the plutonomists - in an economy as unequal as the US.
By our estimation, the top 20 per cent of the population account for more than half of US consumption, while the bottom 20 per cent account for around 5 per cent.
The negative wealth-effect of the house market slowdown will hit the middle and lower income consumer, but is likely to have a lesser impact on the top 20 per cent.
Overall, we are bullish. Growth looks okay, inflation is under control. Sentiment is especially low. Liquidity is picking up, and valuations are very attractive."
Hong Kong is not Wall Street and it is a very interesting take on the situation from someone outside this country. But it does explain why the market would be up not down.
I can't find anything on it but after today's wacky WSJ editorial I wanted to look back and see them talking about the burger-flipper economy jobs that Clinton created, which is the opposite of true, just like now.
Posted by: me | Oct 6, 2007 10:29:08 AM
I have posted a chart illustrating that when the y/y % change in Employment rate drops below zero, there is an acceleration to the downside.
A 110k jobs/month, by the way, translates to an unemployment rate of 5.3% in a year.
Posted by: will rahal | Oct 6, 2007 10:40:02 AM
Alan Abelson in Barrons has similar doubts about the report, and quotes others as saying that there is a "pre-recession look to it."
Also, his thoughts on the financial firms' write-downs of mortgage related losses (and further possible losses, and decrease in profits) is a good read.
Posted by: PeterR | Oct 6, 2007 10:49:07 AM
U.S. hiring weak again in September
Private-sector employment grew by 58,000, ADP says
Posted by: Philippe | Oct 6, 2007 10:49:18 AM
very interesting Will! Great chart.
Techy - As far as conspiracy theories, ehh, theres no point buying into that camp. I feel it clouds my observations, revisions happen. While I agree this past one for AUG was a bit insane, I doubt there is anything behind it other than a flawed model for the data.
Posted by: UrbanDigs | Oct 6, 2007 11:18:12 AM
The gushing is useful because it is keeping me from panicking. I can ignore the water at my ankles because the Titanic cannot sink.
Posted by: Urban Sombrero | Oct 6, 2007 11:24:19 AM
Barry - The usual good job!
But I'm left to wonder. Given the huge dollup of jobs courtesy of the BLS "net birth/death model, the dispensing of the M-3 money supply figure, hedonics, chain links, the questionable weight given to medical expenses - would accurate statistics show the country to be in a situation of stagflation or worse?
Posted by: John Badalian | Oct 6, 2007 12:12:51 PM
I think the market sentiment was for yesterday to be a good day. As in, I suspect any number over 10K jobs was going to get us yesterdays market moves.
It is similar to the banking news coming out. Billions in write offs are perceived to be strength in Financial's.
The Fed cuts have worked on market sentiment better probably than was even expected.
Too bad it is all about hiding from reality. That can unfortunately last only so long.
Posted by: Ralph | Oct 6, 2007 12:26:53 PM
Barry, I do hope you will also publish an indicator established from the uber bears full-throatedness found consistently in the comments section of this blog.
As for the pundits, I don't find the gentleman in the video you posted to be even semi-gushing. He simply states the numbers avoided more downside. As far as pundits go, I might even describe him as objective.
As for the jobs report, unless one is positioned for a recession or for financial armageddon, I don't see how it can be viewed other than positive, relative to the previous 3 month's reports.
Yes it was weak. The numbers have been weak. The trend has been weak. The key here is that weak does not automatically equal recession. Weak can equal soft landing. Weak can mean slower growth.
If one is hoping for a recession, or wishes to exact some sort of moral punishment on investment banks and other users of credit derivatives, then I guess this job report is negative, as it means that the likelihood of a recession or crash has lessened.
But I want to come back to the jobs report, in order to establish why the report is positive. Instead of jobs, lets substitute dollars. If we know we need 150K dollars a month to survive, then obviously, when instead of a check, we get -$4,000 withdrawn from our account last month, things are looking pretty disasterous. However, for this month, not only do we get a check for 110K for September, we also get a bonus of 89K for August (hey, sorry about that -4K). And suddenly, while we are not making our budget of 150K/month, we are also not headed for bankruptcy.
And that is the point here. If one wants a recession, financial armageddon, stock market crash, whatever, then sure, it was a negative jobs report.
However, if one is looking to avoid a recession, has hopes for the indexes finishing in positive territory for the year, and is looking for the economy to survive the weakness and start to grow again in the future, then this was a positive report.
Posted by: Woodshedder | Oct 6, 2007 12:30:07 PM
Nice input Barry -- thanks.
I also liked the comment about the scientific vs. political approach to problem solving. As a science type, let me say, that if you (barry) are right, that B/D adjustment has been a net positive (of any size) over a more than 5 year time frame, then (scientifically :) ) it is obviously politicized.
It is a complete (and I do mean complete) lie if a statistic that is supposed to even out seasonal, and other temporary variations is producing net positive signals year in and year out. That's not a political statistical too if ever there was one.
I have had my doubts about B/D Adj, but like you felt that it didn't make sense to indulge in conspiracy theories. But it does seem to me that if the stat is really producing net gain (and a lot of net gain) then it cannot be what it purports to be, and thus, unfortunately, we have clear, scientific proof, that there is a con going on.
Posted by: VoiceFromTheWilderness | Oct 6, 2007 12:44:11 PM
Does anyone understand or have a explanation that would explain the B/D model's ever increasing numbers? I don't know enough about how they do this to come up with an explanation that doesn't involve a naive explanation like lots of little companies starting up. Everything I look at points at consolidation, but that's just in my little industry. And, that being said, I'm at a startup myself that just hired a few folks...
Sujal
Posted by: sujal | Oct 6, 2007 12:50:58 PM
http://www.stats.bls.gov/web/ceshighlights.pdf
Some nice graphs at the above BLS site. They do well to illustrate the weak employment situation.
Posted by: Woodshedder | Oct 6, 2007 12:55:12 PM
A crown? WTF?
The biggest number of the week, and you're in a dentist chair?
Besides, what self-respecting dentist is working on Friday? Things must be different on the east coast.
Posted by: Grodge | Oct 6, 2007 1:14:28 PM
Good luck avoiding a recession, you must think there will never again be one. Or perhaps you just want to wait a little while?
Posted by: Aaron | Oct 6, 2007 1:23:32 PM
To inject anecdotal hominems into the stats debate, as street-level response to this Fed "gusher" (and all their "management" teen spirit, which for lack of a better word is still "pump-and-dump" in the classic sense), "hold and go long" on these observations:
1) A lot more people living on the street and freeway offramps now, when they should be moving into assisted-living in advance of winter, the safety net obviously shredded;
2) A lot more crazies howling up and down the street, when they should be moving into treatment and retraining programs in advance of hyped "anticipated credit-con rebound";
3) In private clubs and bars, a lot more nervous chatter and sudden inexplicable almost turrets-syndrome outbursts, as one might expect facing personal financial recession into pre-retirement meltdown;
4) Near absolute zero interest in politics;
Translation: The key is to study employment by peer group cadres, not in the aggregate.
20-30 year old employment is nowhere near expectations. Low salaries, meaningless service level jobs subject at any moment to tele-outsourcing, against rising cost of living which is far, far above Fed "CPI";
30-40 year old employment at salaries only a fraction of prior decades for same level of training and experience. Retiring principals are turning the screws on associates against flat-line mid-level job opportunities;
40-50 year old employment seems OK, there is more than enough retirement for them to step into the boomer's shoes, and a recognized shortage of people with quality job skills;
-but-
50-60 year old employment is being crushed, musical chairs, RIF's just pre-pension, a standard lament, become a howl, in the face of real cost growth and credit shrinkage.
The 20-30's are still living at home with 40-50's and late-starting 50-60's, which is nothing new, but becoming national pandemic.
And assisted-living is the fastest growing industry for new jobs and construction, in essence, a fine seine net set to capture the last effusion of wealth from the go-go 80's, before the tragedy of the dot-con debacle rears it's ugly head, Mom and Pop unable to retire, and Jr still stuck living at home.
Add rapid US$ decline, credit fees > 28%, 1999 market chop, stir well, serve hot.
From a man-in-the-street ad hominem versus "view-from-36,000-feet" hyperspective, this less-than-replacement job statistic makes perfect sense, and showcases how desperate the White House and Wall Street are to spin **up any bit of fluff**.
Even major brokerages are spamming their clients, begging them not to, "miss this great run-up opportunity, by sitting on the sidelines in cash". Sound familiar? 2000?
Can you spell W-a-M-u?
--
Here's a ironically chilling VFX comparison:
1920's
http://tinyurl.com/2yfxye
2000's
http://tinyurl.com/2cvafr
The Great War is over, US won, it's all a big party (line) now. Grab some bathtub gin, and jump in, the water's fine. The water always will be fine for you, Mr. Torrance.
http://tinyurl.com/2b5vnp
Market winters can be fantastically cruel.
Posted by: Oldman Lincoln | Oct 6, 2007 2:01:12 PM
