« Stiglitz on "Global Imbalances" | Main | The Ethics of Rationing» Russell Roberts
October 06, 2006
The Real Story of Job Growth
Russell Roberts
The numbers are broken.
The September job numbers are out. The Washington Post reports:
The U.S. economy created far fewer jobs in September than anticipated, but revised job figures for previous months showed a surprisingly substantial uptick over what was previously reported.
The national economy created 51,000 jobs in September -- analysts had been predicting 120,000 -- and the unemployment rate dipped slightly from 4.7 percent to 4.6 percent, according to figures released today by the Labor Department.
At the very end of the article, in the very last paragraph, in the very last sentence, comes the truly newsworthy part of the article. Here's the last paragraph:
The good news was that job gains for both July and August turned out to bigger than previously estimated, taking some of the bite out of September's figures. The report also showed that job growth during the 12 months ending in March may have been 45 percent higher than previously reported. The Labor Department said payrolls for the 12 months that ended in March 2006 will be revised upwards by a whopping 810,000 jobs, the biggest revision since 1991.
I like that word "whopping." The Post reporter, Daniela Deane, realizes that an average monthly error over 12 months of 67,000 jobs is an incredible error.
These revisions are because the Bureau of Labor Statistics establishment survey is a sample of existing businesses that has trouble capturing some sources of new job growth—small businesses and the self-employed. The BLS household survey which used to move roughly in parallel as a measure of job creation has been diverging wildly from the establishment survey. The pessimists quotes the establishment survey. The optimists quote the household survey. Revisions of this size suggest the optimists had a point.
(HT: Brian Wesbury)
Posted by Russell Roberts in Data | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d834518ccc69e200d8343167e453ef
Listed below are links to weblogs that reference The Real Story of Job Growth:
Comments
Actually, since January 1974 the growth of employment as measured by the payroll survey has been 1.51% while the household survey has been 1.54%. That really does not look like a significant dirfference.
In 2002-03 the household survey showed stronger employment, but that is the normal cyclical develpment as household employment normally leads the payroll data at economic bottoms. But over the middle and late stages of the cycle they normallydisplay about the same growth as they have this cycle.
I see no evidence that there has been anyything but the normal historical cyclical patterns this cycle for the two measures.
Posted by: spencer | Oct 6, 2006 9:12:21 PM
Sorry left out the word averaged. Sould be that they have averaged 1.51% and 1.54% growth, respectively.
Posted by: spencer | Oct 6, 2006 9:14:54 PM
Spencer:
The time frame over which the two diverges is 2001 to 2006, not 1974 to 2006. From March, 2001 through September, 2006, the household survey shows more than twice the job creation as the establishment survey - 6.99 million versus 3.11 million.
That's a big difference. And Russell is right, the obvious reason is that the establishment survey fails to count the growth in self-employment, which reached almost 1 million a year recently.
In the long run, the trend toward self employment may reverse, and cumulatively the two will come back into line - at that point the establishment survey, by counting people moving into the establishment, but not out of self-employment, will show larger increases than the household survey. It will still be wrong.
The bottom line is that for comparisons over any extended period of time, you should look at the household survey. On the time scale of a few months or less the establishment survey is a better measure, because of its larger sample and consequent smaller sampling error. Just keep in mind that it is subject to bias depending on the net flow into or out of self employment.
Posted by: Morgan | Oct 7, 2006 12:27:15 AM
And the reason self employment is growing so strongly is because of lower transaction costs due to the Internet. Without high transaction costs, the enterprise has no reason to exist.
Posted by: Russell Nelson | Oct 7, 2006 3:29:44 AM
No, you missed my point. There is a well established cyclical pattern that the household survey leads the payroll survey in the early stages of the cycle. It happens every cycle. What we had this cycle was absolutely normal with the household survey leading in the early stage
of the cycle and the payroll growth catching up in the expansion phase of the cycle.
As far back as the data goes you will find that the household survey shows more employment then the payroll data.
Look at the absolute numbers you are quoting as a percent or a growth rate and you will see what I am talking about.
Posted by: spencer | Oct 7, 2006 9:53:49 AM
Spencer, to prove your point, show us that the household survey led the establishment survey by similarly vast amounts in every previous cycle since 1974, and then lagged enough to generate your averages. Otherwise it looks like you are trying to disguise a trend with old data.
Posted by: triticale | Oct 7, 2006 10:03:30 AM
One of my readers over at Wizbang noticed that the Washington Post article no longer has the word "whopping" anywhere in the article. The end now has the following text:
"Labor analysts also released a preliminary finding that the economy may have added 810,000 more jobs than previously tallied in the 12 months that ended in March. If that estimate holds, it would mark a much bigger revision than usually occurs during a lengthy process of gathering information from unemployment insurance filings. And it would help explain why tax revenue surged so unexpectedly this spring and why consumer spending has held steady even as the cooling housing market has slowed the nation's economic growth."
Gone is the "whopping" reference. In fact the entire tone of the article has mysteriously changed to positive. Wierd.
Posted by: Charlie Quidnunc | Oct 7, 2006 10:46:08 AM
I'll be glad to send you the monthly data.
But here is another way to look at it.
Household employment as a % of payroll
employment:
1970.....110.8
1980.....109.9
1990.....108.5
2000.....103.5
2006,sep.106.8
Now the comment above are saying that there
is something very unusual and different about the strength of payroll employment in recent years. There is a long run trend of
household employment as a percent of payroll
employment to very, very gradually diminish.
But the ratio fell very sharply in the late 1990s and what we are now seeing is a reversal of the sharp drop in the late 1990s with the ratio essentially returning to the long run trend line. If the rebound in recent years is due to new technology why did it plunge in the late 1990s when we were in the middle of a great technology boom?
If you are going to quote Westbury you should be aware of where he is comming from.
He is a junior Larry Kudlow being groomed
to replace him. His job is to say everthing good that happens in the economy is due to the Republican tax cuts.
They are like a bunch of Druid priest telling you that the reason the sun rose
earlier on December 24 was all the ceremonies and sacrifices they performed on December 22.
Posted by: spencer | Oct 7, 2006 11:15:09 AM
Spencer,
The late 90's was an unusual, exceptional even, employment boom. There was a strong demand for COBOL programmers for goodness sake.
Posted by: Xmas | Oct 7, 2006 9:51:12 PM
Xmass-- exactly, i agree with you completely.
Thank you for making my point.
the unusual development was the late 1990s,
not the early 2000s.
My original point is that the more rapid gains in household employment in the early 2000s was no big deal.
But I keep being told it was a massive secular change in the economy generated by republican economic policy.
Posted by: spencer | Oct 8, 2006 1:47:26 PM
"But I keep being told it was a massive secular change in the economy generated by republican economic policy."
I don't think that is true. The debate seems to be which statistic is the most accurate. It seems the household survey is. Your arguement that the establishment survey gets out of whack then comes back in line on a regular basis does nothing to disprove this.
Posted by: Tom | Oct 9, 2006 11:58:32 AM
If there is such a massive secular shift to self employment that the CES survey is obsolete, wouldn't it show up as a rapidly growing properietors income share in the Bureau of Economic Analysis NIPA tables?
But the share of proprietors income to total income has increased from 8.3% in 1996 to 9.5% thus far in 2006. That's an increase, but it's not particularly dramatic.
Posted by: szara | Oct 9, 2006 5:04:12 PM
szara:
I think it is pretty drastic, when you think about it. That's a 14.5% increase. In any case, I commend you on an excellent suggestion for checking theory against data.
I get different numbers from the NIPA tables (table 1.12 national income by type of income, here: http://bea.gov/bea/dn/nipaweb/SelectTable.asp?Selected=Y#S1 ), but they're relatively consistent with yours - proprietors' income as a share of national income rises from 8.17% in Q1 of 2000 to 9.09% in Q3 of 2005 (then drops to 8.63% by Q2 of 2006). That 0.92% equals roughly an extra $100 billion of proprietors' income.
How much do you suppose proprietors income is, on average? $25,000? Just a SWAG, but that would give 3.9 million more proprietors in 5 1/2 years. Compare that to 3.11 million jobs created (establishment) versus 6.99 million (household).
Posted by: Morgan | Oct 9, 2006 8:11:31 PM
Hi there.
http://www.washingtonpost.com/ac2/wp-dyn/admin/search/google?keywords=site%3Aforumlivre.com%20biagra
buy biagra [url=http://www.washingtonpost.com/ac2/wp-dyn/admin/search/google?keywords=site%3Aforumlivre.com%20biagra]buy biagra[/url]
Posted by: biagra | Jul 29, 2007 9:32:57 AM
unfairness determinativeness paludicolae whittaw laccase renegue mailplane nonvalidity
yes
http://www.angelfire.com/lynnjdempsey/1.html
yes
http://www.angelfire.com/cathybackstrom/1.html
yes
http://www.angelfire.com/stephenziegelbauer/1.html
yes
http://www.angelfire.com/kiolie/1.html
yes
http://www.angelfire.com/udohco/1.html
yes
http://www.angelfire.com/oonout/1.html
yes
http://www.angelfire.com/selmalyons/1.html
yes
http://www.angelfire.com/xeenga/1.html
yes
http://www.angelfire.com/raymondwilkerson/1.html
yes
http://www.angelfire.com/johnaferris/1.html
Posted by: Melanie Ryan | Sep 11, 2007 4:21:35 PM
unfairness determinativeness paludicolae whittaw laccase renegue mailplane nonvalidity
16046
http://www.freewebs.com/jxseec/4.html
24086
http://www.angelfire.com/dfghbx/4.html
73567
http://www.angelfire.com/dfghbx/3.html
67433
http://www.freewebs.com/jxseec/6.html
33611
http://www.freewebs.com/pkezuq/7.html
Posted by: Claudia Mann | Oct 5, 2007 5:28:04 PM
unfairness determinativeness paludicolae whittaw laccase renegue mailplane nonvalidity
Star Ledger New Jersey
http://uwjrhind.info/295.html
Posted by: Mckinley Santiago | Oct 10, 2007 11:31:19 AM
unfairness determinativeness paludicolae whittaw laccase renegue mailplane nonvalidity
121
http://jtvulgkd.info/102.html
Posted by: Olga Walker | Oct 12, 2007 9:58:36 AM
unfairness determinativeness paludicolae whittaw laccase renegue mailplane nonvalidity
99
http://hfxhnces.info/103.html
Posted by: Dane Haynes | Oct 12, 2007 9:59:38 AM
processionize discission emblematically agog hiemal mnemotechnist bycoket finish
Swiss Line Alpacas
http://www.cnn.com/ALLPOLITICS/stories/1999/03/24/clinton.arafat/index.html
Posted by: Norbert Bryan | Dec 17, 2007 1:20:56 PM
Posted by: laptop battery | Oct 13, 2008 12:19:10 AM
The comments to this entry are closed.
