UnThreaded | Threaded | Whole Thread (52) | Ignore Thread Prev | Next
Author: Pituophis Big gold star, 5000 posts Old School Fool Add to my Favorite Fools Ignore this person (you won't see their posts anymore) Number: of 457445  
Subject: An Analysis of Jobs Data 1/1/01-5/31/04 Date: 6/13/2004 11:46 AM
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Recommendations: 171
AN ANALYSIS OF EMPLOYMENT DATA -- JANUARY 1, 2001 - MAY 30, 2004

Brave New World

I cannot remember a time in my lifetime when there was as much skepticism about government supplied data. Not that the data has ever been all that reliable, but increasingly it seems that the government, particularly the Department of Labor's Bureau of Labor Statistics (BLS), has abrogated their charge to inform the financial community and instead now seems to think that their role is to placate, to assuage, to keep the natives calm, or as Aldous Huxley might say: to supply the populace with their happy drug, their soma. With millions of people on Prozac and other anti-depressives these days, maybe that analogy is not so metaphorical.

The employment data are a case in point: A casual observer would have no reason to doubt that the U.S. is in the midst of an absolute employment boom. Certainly that (boom) premise is explicitly expressed in the public statements of the Secretary of Labor: "I'm pleased to see strong job growth, and that's what I am concerned about," Labor Secretary Elaine Chao said in an interview. "Every month of strong job growth is good news to me and it's good news for America."

But has there really been "strong job growth?" Let's take a hard look at the data and see if we can answer that question.


Employment versus Unemployment

An in-depth analysis of the government supplied data don't seem to support the conclusion of strong job growth. But don't take my word for it: The Employed, 16 and Over series from The Household Survey (Appendix 1), tells us that from the lowest employment month in recent history, November 2001, until the most current month for which data are available, that employment has increased from 136.2 million to 138. 8 million, an increase of 2.6 million jobs. Or, if you choose to look at the Establishment Survey, from which the "non-farm" payrolls number (Appendix 2) that is released each month is derived, and again measure from the lowest employment month in recent history, employment has increased from 129.8 million in August 2003 to 131.2 most recently, an increase of 1.3 million jobs. Now that's strong job growth...isn't it? The government would prefer that you don't think about that question too long and definitely don't look at the rest of their data - afterall, curiosity killed the cat an' all - oh well, I'll take my chances, and I'll hope you'll try to follow my analysis.

To determine if an increase of 2.6 million jobs in almost 3 years or 1.3 million jobs in 9 months is "strong" job growth we have to take a look at some of the other data beginning with the number of people that are still unemployed: From Household Survey - Unemployed, 16 and Over (Appendix 3), looking at that same time frame (November 2001 - May 2004), which puts "job growth" in the most favorable light, you will see that while 2.6 jobs have been added (the BLS and the administration like to say "created"), the number of unemployed has actually increased by 200,000.. Even in the May 2004, when the headline screamed, "248,000 JOBS ADDED IN MAY!", unemployment actually increased by 49,000 jobs.

And heaven forbid that we might take an even more critical look at the data, such as looking at job "growth" and/or unemployment growth (that's right,UNemployment growth) since, say, January 2001, because then we would see that the number of jobs added since January 2001 is........ooops, I guess that should say "jobs lost" - 1.2 million fewer workers on the non-farm payroll and a whopping 2.2 million more workers UNEMPLOYED in May 2004 than in January 2001

Is that strong job growth? You decide.


The Civilain Labor Force and the Participation Rate

And most of the workers that have come off of the unemployed rolls over the last year have done so because of a declining Civilian Labor Force (Appendix 4). - that is, many job seekers are supposedly no longer seeking work and therefore they are not counted as unemployed. Dividing the Civilian Labor Force by the Civilian Noninstitutional Population gives up the Participation Rate. Between November 2003 and May 2004, the BLS is telling us that the Civilian Labor force shrunk by 213,000 workers and that the Participation Rate has decreased from 67.2% in January 2001 to 65.9% most recently.

So I ask you again, is that strong job growth?


Population Growth

And are these data as wildly contradictory as they seem, with "strong" non-farm payroll increases coincident with rising unemployment? Yes and no. First, let me explain why, to some degree, the data are not contradictory: One reason that increasing unemployment might not be contradicted by data that shows job growth is really rather simple and can be summarized in two words: Population growth.

To analyze population growth you might want to look at two series: The Total Civilian Noninstitutional Population (Appendix 6) (which is the series that the BLS uses to compute such things as the Participation Rate) and the Total Population Including Armed Forces Overseas (Appendix 7). Similar growth numbers can be derived from either series but let's look at the series that BLS uses. Again, to be fair and consistent, let's look at the November '01 - May '04 time frame that puts job growth the best possible light: The U.S. Noninstitutional Population has grown from 216.1 million in November '01 to 223 million in May of this year. So while Elaine Chao would like us to focus on the fact that 2.6 million jobs have been added according to the Household Survey, she definitely is not going to mention the fact that the U.S. population has grown by 6.9 million people or 230, 000/month in that same time frame.

To be completely fair, I should point out that some demographic analysis is necessary to determine what portion of the increasing population would actually be participating in the work force. If we assume that the participation rate for the new arrivals is the same as that for the existing population (which, I believe, IS the BLS assumption), then 65.9% or 4.5 million of the new arrivals would be looking for jobs. So, by the rosiest estimates there are 2.6 million jobs created since November 2001 for the 4.5 million people that need jobs. That also comes to exactly 150,000 new jobs needed per month just to stay even - that is, for zero job growth. And even with all of these new workers being added, still, as noted above, while the population was growing by 230K/month, the Civilian Labor Force shrunk by 213K jobs since November '03 removing many workers from the official ranks of the unemployed. (And I should note here that while only 4.5 million of the 7 million new arrivals are looking for work, ALL of them are creating a greater strain on limited resources - but that is the subject of another post.)


A Brave New Procedure - The Business Birth/Death Model

So you see, you can have job "growth" and rising unemployment - up to a point . If jobs appear to be growing at a rate of greater than 150,000 jobs per month and still unemployment is growing AND the unemployment rate (Appendix 8) is staying even or going up - then, mission contol, we have a problem. And some of the jobs data released over the past few months simply does not compute. Look at the most recent release for instance. Supposedly, 248,000 jobs were "created" in May and yet there were more unemployed people in May than there were in April. Huh? Often, when that type of discrepancy arises, you can look for the answer in the Participation Rate. And often the Participation Rate will rise - formerly discouraged job seekers once again start looking for work - when the job market gets better. But the Participation Rate has held steady since February. So where is the the discrepancy?

As previously reported on this board, and hardly anywhere else, the BLS has implemented a new procedure, the "Business Birth/Death Model" (Appendix 9), which is responsible adding 412,000 jobs this year, 733,000 since February, and is responsible for creating 1.1 million jobs out of thin air since it's implementation in April of 2003. The BLS reported that non-farm payrolls only increased by 1.3 million total jobs in that time frame - so the Business Birth/Death Model has accounted for all but 200,000 or 85% of the reported job increases in the last 13 months. So if the numbers don't seem to add up, there is a simple explanation: THEY DON'T ADD UP.

Let's look at this another way: If this was 2002 or 1999 or 1956 and we had identical performance in the job market, we would not be celebrating "job growth" at all - we would be lamenting terrible JOB LOSSES. Does this seem right or logical to anyone that virtually ALL of the reported job growth in the last year is due to a new procedure and that never before in history would we have considered this job market performance good???


Conclusions

When I started the process of researching these data to produce this article, I really thought I was going to find some evidence of job growth. Even I am surprised at just how horrid the job picture really is. If you are looking at the job market as evidence of an improving economy, I think you should heed this WARNING: I am firmly convinced that the job data are providing virtually NO EVIDENCE of a growing economy at this point. I should also add that I am not implying malicious manipulation of the data. The data are what they are. Most people just look at the headlines and miss the real story.

So what does this mean for investors. I wish I knew. Maybe the economy does gain more and more traction and we begin to see some REAL job growth. Maybe the stock markets will correctly anticipate that growth. Most main stream economist, virtually everyone you hear on CNBC, is parroting that mantra. Even on this board, the consensus seems to be that equities are going to go up for awhile before reality sets in - even MrPlunger is talking about a growth theme and I've learned to listen when MrP speaks. And there are some other signs of "recovery", though I find it odd that we are talking about "recovery" almost 3 years after one negative growth quarter...and don't get me started on the GDP data. Even if everything that the growth story is based on is all smoke and mirrors, I'm sure you've heard the old adage that goes something like "the markets can stay irrational for longer than you can stay solvent." And of course, you've got the Fed factor which has been the engine of "growth" for the last three years, though if you ask me, that type of growth is somewhat akin to making a deposit into your brokerage account with your credit card. But the Fed has been on a money creation/debt monetization rampage, unprecedented even for this Fed, over the last couple of months, and last spring I learned an important lesson about fighting the Fed and I'm not going to make that mistake again. My investment choices could be wrong, but I'm getting up in years and can't afford to take a lot of chances so I'm pretty much fully hedged but I'm giving NO advice in that regard....just be careful out there...

=============================================================

A P P E N D I C E S

----------------------
Appendix 1. Civilian Employment: Sixteen Years & Over
Series ID: CE16OV
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Thousands
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555.
[........]
2001-01-01 137790
2001-02-01 137581
2001-03-01 137738
2001-04-01 137275
2001-05-01 137063
2001-06-01 136842
2001-07-01 137091
2001-08-01 136314
2001-09-01 136869
2001-10-01 136447
2001-11-01 136234
2001-12-01 136078
2002-01-01 135715
2002-02-01 136362
2002-03-01 136106
2002-04-01 136096
2002-05-01 136505
2002-06-01 136353
2002-07-01 136478
2002-08-01 136811
2002-09-01 137337
2002-10-01 137079
2002-11-01 136545
2002-12-01 136459
2003-01-01 137447
2003-02-01 137318
2003-03-01 137300
2003-04-01 137578
2003-05-01 137505
2003-06-01 137673
2003-07-01 137604
2003-08-01 137693
2003-09-01 137644
2003-10-01 138095
2003-11-01 138533
2003-12-01 138479
2004-01-01 138566
2004-02-01 138301
2004-03-01 138298
2004-04-01 138576
2004-05-01 138772

===================================

Appendix 2. Total Nonfarm Payrolls: All Employees
Series ID: PAYEMS
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Thousands
Date Range: 1939-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: Handbook of Methods Establishment Survey -
http://stats.bls.gov:80/opub/hom/homch2_itc.htm

Frequently Asked Questions -
http://stats.bls.gov:80/cps/cps_faq.htm
[....]
2001-01-01 132388
2001-02-01 132492
2001-03-01 132507
2001-04-01 132236
2001-05-01 132237
2001-06-01 132087
2001-07-01 131972
2001-08-01 131831
2001-09-01 131564
2001-10-01 131203
2001-11-01 130871
2001-12-01 130659
2002-01-01 130494
2002-02-01 130404
2002-03-01 130447
2002-04-01 130379
2002-05-01 130381
2002-06-01 130406
2002-07-01 130295
2002-08-01 130306
2002-09-01 130259
2002-10-01 130342
2002-11-01 130305
2002-12-01 130096
2003-01-01 130190
2003-02-01 130031
2003-03-01 129921
2003-04-01 129901
2003-05-01 129873
2003-06-01 129859
2003-07-01 129814
2003-08-01 129789
2003-09-01 129856
2003-10-01 129944
2003-11-01 130027
2003-12-01 130035
2004-01-01 130194
2004-02-01 130277
2004-03-01 130630
2004-04-01 130976
2004-05-01 131224

==========================

Appendix 3. Unemployed: 16 Years & Over
Series ID: UNEMPLOY
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Thousands
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555.
[.....]
2001-01-01 5997
2001-02-01 6072
2001-03-01 6136
2001-04-01 6274
2001-05-01 6227
2001-06-01 6481
2001-07-01 6583
2001-08-01 7057
2001-09-01 7151
2001-10-01 7723
2001-11-01 8020
2001-12-01 8291
2002-01-01 8126
2002-02-01 8184
2002-03-01 8278
2002-04-01 8578
2002-05-01 8397
2002-06-01 8384
2002-07-01 8400
2002-08-01 8335
2002-09-01 8269
2002-10-01 8363
2002-11-01 8565
2002-12-01 8698
2003-01-01 8428
2003-02-01 8581
2003-03-01 8519
2003-04-01 8799
2003-05-01 8957
2003-06-01 9245
2003-07-01 9048
2003-08-01 8929
2003-09-01 8966
2003-10-01 8797
2003-11-01 8653
2003-12-01 8398
2004-01-01 8297
2004-02-01 8170
2004-03-01 8352
2004-04-01 8164
2004-05-01 8203

=======================

Appendix 4. Civilian Labor Force
Series ID: CLF16OV
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Thousands
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555..
[...]
2001-01-01 143787
2001-02-01 143652
2001-03-01 143873
2001-04-01 143549
2001-05-01 143290
2001-06-01 143323
2001-07-01 143674
2001-08-01 143372
2001-09-01 144020
2001-10-01 144171
2001-11-01 144254
2001-12-01 144369
2002-01-01 143842
2002-02-01 144546
2002-03-01 144384
2002-04-01 144675
2002-05-01 144902
2002-06-01 144738
2002-07-01 144879
2002-08-01 145146
2002-09-01 145606
2002-10-01 145442
2002-11-01 145109
2002-12-01 145157
2003-01-01 145875
2003-02-01 145898
2003-03-01 145818
2003-04-01 146377
2003-05-01 146462
2003-06-01 146917
2003-07-01 146652
2003-08-01 146622
2003-09-01 146610
2003-10-01 146892
2003-11-01 147187
2003-12-01 146878
2004-01-01 146863
2004-02-01 146471
2004-03-01 146650
2004-04-01 146741
2004-05-01 146974

===============================
Appendix 5 Civilian Participation Rate
Series ID: CIVPART
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Percent
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555.
[....]
2001-01-01 67.2
2001-02-01 67.1
2001-03-01 67.1
2001-04-01 66.9
2001-05-01 66.7
2001-06-01 66.7
2001-07-01 66.8
2001-08-01 66.6
2001-09-01 66.8
2001-10-01 66.8
2001-11-01 66.7
2001-12-01 66.7
2002-01-01 66.4
2002-02-01 66.7
2002-03-01 66.6
2002-04-01 66.7
2002-05-01 66.7
2002-06-01 66.6
2002-07-01 66.6
2002-08-01 66.6
2002-09-01 66.8
2002-10-01 66.6
2002-11-01 66.4
2002-12-01 66.4
2003-01-01 66.3
2003-02-01 66.3
2003-03-01 66.2
2003-04-01 66.4
2003-05-01 66.3
2003-06-01 66.5
2003-07-01 66.3
2003-08-01 66.2
2003-09-01 66.1
2003-10-01 66.2
2003-11-01 66.2
2003-12-01 66.0
2004-01-01 66.1
2004-02-01 65.9
2004-03-01 65.9
2004-04-01 65.9
2004-05-01 65.9
===============================
Appendix 6. Civilian Noninstitutional Population
Series ID: CNP16OV
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Not Seasonally Adjusted
Frequency: Monthly
Units: Thousands
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555..
[......[
2001-01-01 213888
2001-02-01 214110
2001-03-01 214305
2001-04-01 214525
2001-05-01 214732
2001-06-01 214950
2001-07-01 215180
2001-08-01 215420
2001-09-01 215665
2001-10-01 215903
2001-11-01 216117
2001-12-01 216315
2002-01-01 216506
2002-02-01 216663
2002-03-01 216823
2002-04-01 217006
2002-05-01 217198
2002-06-01 217407
2002-07-01 217630
2002-08-01 217866
2002-09-01 218107
2002-10-01 218340
2002-11-01 218548
2002-12-01 218741
2003-01-01 219897
2003-02-01 220114
2003-03-01 220317
2003-04-01 220540
2003-05-01 220768
2003-06-01 221014
2003-07-01 221252
2003-08-01 221507
2003-09-01 221779
2003-10-01 222039
2003-11-01 222279
2003-12-01 222509
2004-01-01 222161
2004-02-01 222357
2004-03-01 222550
2004-04-01 222757
2004-05-01 222967

==========================
Appendix 7. Total Population: All Ages including Armed Forces Overseas
Series ID: POP
Source: U.S. Department of Commerce: Census Bureau
Release: Not Applicable
Seasonal Adjustment: Not Applicable
Frequency: Monthly
Units: Thousands
Date Range: 1952-01-01 to 2004-02-01
Last Updated: 2004-05-27 9:35 AM CT
Notes:The intercensal estimates for 1990-2000 for the United States
population are produced by converting the 1990-2000 postcensal
estimates prepared previously for the U. S. to account for differences
between the postcensal estimates in 2000 and census counts (error of
closure). The postcensal estimates for 1990 to 2000 were produced by
updating the resident population enumerated in the 1990 census by
estimates of the components of population change between April 1, 1990
and April 1, 2000-- births to U.S. resident women, deaths to U.S.
residents, net international migration (incl legal & residual foreign
born), and net movement of the U.S. armed forces and civilian citizens
to the United States. Intercensal population estimates for 1990 to
2000 are derived from the postcensal estimates by distributing the
error of closure over the decade by month. The method used for the
1990s for distributing the error of closure is the same that was used
for the 1980s. This method produces an intercensal estimate as a
function of time and the postcensal estimates,using the following
formula: the population at time t is equal to the postcensal estimate
at time t multiplied by a function. The function is the April 1, 2000
census count divided by the April 1, 2000 postcensal estimate raised
to the power of t divided by 3653..
[....]
2001-01-01 283933
2001-02-01 284164
2001-03-01 284368
2001-04-01 284593
2001-05-01 284823
2001-06-01 285074
2001-07-01 285321
2001-08-01 285588
2001-09-01 285863
2001-10-01 286131
2001-11-01 286377
2001-12-01 286609
2002-01-01 286838
2002-02-01 287048
2002-03-01 287258
2002-04-01 287487
2002-05-01 287712
2002-06-01 287962
2002-07-01 288205
2002-08-01 288468
2002-09-01 288742
2002-10-01 289001
2002-11-01 289248
2002-12-01 289479
2003-01-01 289703
2003-02-01 289910
2003-03-01 290116
2003-04-01 290340
2003-05-01 290562
2003-06-01 290809
2003-07-01 291049
2003-08-01 291310
2003-09-01 291582
2003-10-01 291837
2003-11-01 292080
2003-12-01 292309
2004-01-01 292527
2004-02-01 292734

=============================

Appendix 8. Civilian Unemployment Rate
Series ID: UNRATE
Source: U.S. Department of Labor: Bureau of Labor Statistics
Release: The Employment Situation
Seasonal Adjustment: Seasonally Adjusted
Frequency: Monthly
Units: Percent
Date Range: 1948-01-01 to 2004-05-01
Last Updated: 2004-06-04 8:36 AM CT
Notes: The Bureau of Labor Statistics (BLS) announced several
revisions to the Household Survey on Friday Feb.7th 2003,
with the release of the January 2003 Data. They introduced
the Census 2000 population controls (which affect data
back to 2000 and cause a break in the data in January 2000),
a new seasonal adjustment procedure, and new seasonal
factors back to January 1998. For further information
contact the Current Employment Statistics (CES) homepage at
www.bls.gov/ces or by calling 202-691-6555.
[.....]
2001-01-01 4.2
2001-02-01 4.2
2001-03-01 4.3
2001-04-01 4.4
2001-05-01 4.3
2001-06-01 4.5
2001-07-01 4.6
2001-08-01 4.9
2001-09-01 5.0
2001-10-01 5.4
2001-11-01 5.6
2001-12-01 5.7
2002-01-01 5.6
2002-02-01 5.7
2002-03-01 5.7
2002-04-01 5.9
2002-05-01 5.8
2002-06-01 5.8
2002-07-01 5.8
2002-08-01 5.7
2002-09-01 5.7
2002-10-01 5.7
2002-11-01 5.9
2002-12-01 6.0
2003-01-01 5.8
2003-02-01 5.9
2003-03-01 5.8
2003-04-01 6.0
2003-05-01 6.1
2003-06-01 6.3
2003-07-01 6.2
2003-08-01 6.1
2003-09-01 6.1
2003-10-01 6.0
2003-11-01 5.9
2003-12-01 5.7
2004-01-01 5.6
2004-02-01 5.6
2004-03-01 5.7
2004-04-01 5.6
2004-05-01 5.6

================================
Appendix 9. CES Net Birth/Death Model

In 2004, the CES sample includes about 160,000 businesses and government agencies drawn from a sampling frame of Unemployment Insurance tax accounts which cover approximately 400,000 individual worksites. The active CES sample includes approximately one-third of all nonfarm payroll workers. The sample-based estimates are adjusted each month by a statistical model designed to reduce a primary source of non-sampling error, the inability of the sample to capture on a timely basis, employment growth generated by new business formations.
There is an unavoidable lag between an establishment opening for business and its appearing on the sample frame and being available for sampling. Because new firm births generate a portion of employment growth each month, non-sampling methods must be used to estimate this growth.
Earlier research indicated that while both the business birth and death portions of total employment are generally significant, the net contribution is relatively small and stable. To account for this net birth/death portion of total employment, BLS is implementing an estimation procedure with two components: the first component uses business deaths to impute employment for business births. This is incorporated into the sample-based link relative estimate procedure by simply not reflecting sample units going out of business, but imputing to them the same trend as the other firms in the sample.
The second component is an ARIMA time series model designed to estimate the residual net birth/death employment not accounted for by the imputation. The historical time series used to create and test the ARIMA model was derived from the UI universe micro level database, and reflects the actual residual net of births and deaths over the past five years. The ARIMA model component is updated and reviewed on a quarterly basis.
The net birth/death model component figures are unique to each month and exhibit a seasonal pattern that can result in negative adjustments in some months. These models do not attempt to correct for any other potential error sources in the CES estimates such as sampling error or design limitations.
The most significant potential drawback to this or any model-based approach is that time series modeling assumes a predictable continuation of historical patterns and relationships and therefore is likely to have some difficulty producing reliable estimates at economic turning points or during periods when there are sudden changes in trend. BLS will continue researching alternative model-based techniques for the net birth/death component; it is likely to remain as the most problematic part of the estimation process.

The table below shows the net birth/death model adjustment used in the published CES estimates since the establishment of the most recent benchmark level for March 2003.

2003 Net Birth/Death Adjustment (in thousands)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
... ... ... 128 192 164 -83 124 33 45 30 62

2004 Net Birth/Death Adjustment (in thousands)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
-321 115 153 270 195

=================================================================


.......................
Post New | Post Reply | Reply Later | Create Poll . Report this Post | Recommend it!
Print the post  
UnThreaded | Threaded | Whole Thread (52) | Ignore Thread Prev | Next

Announcements

Post of the Day:
Value Hounds

Ubiquiti Analysis
What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Community Home
Speak Your Mind, Start Your Blog, Rate Your Stocks

Community Team Fools - who are those TMF's?
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and "#1 Media Company to Work For" (BusinessInsider 2011)! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.
Advertisement