Emails (excerpted), Cheryl Abbot, regional economist, U.S. Bureau of Labor Statistics, May 9, 2013

From: Selby, Gardner (CMG-Austin)

Sent: Thursday, May 09, 2013 10:02 AM

To: Abbot, Cheryl - BLS

Subject: RE: Fresh inquiry




... I have a second claim that we hope to verify today.


It is that over the last 10 years, Texas created 33 percent of the net new jobs nationwide. Would you be able to peel out figures relevant?


Thank you .



10:37 am

We’ve had this discussion before – I believe it was during the last presidential campaign. Although many people/media stories use descriptions such as ‘this state created X percent of all national jobs,’ BLS does not. The reason is that the state CES (nonfarm employment) estimates are derived independently from the national estimates – and the states are not controlled to the national estimates. Shown below is an excerpt from the Technical Note of the latest Regional and State Employment and Unemployment release (Regional and State Employment and Unemployment Technical Note).



   Caution on aggregating state data. State estimation procedures are designed

to produce accurate data for each individual state. BLS independently develops a

national employment series; state estimates are not forced to sum to national

totals. Because each state series is subject to larger sampling and nonsampling

errors than the national series, summing them cumulates individual state level

errors and can cause significant distortions at an aggregate level. Due to these

statistical limitations, BLS does not compile a "sum-of-states" employment series,

and cautions users that such a series is subject to a relatively large and volatile

error structure.


To illustrate part of that discussion, consider this very simplified hypothetical where the United States consists of only three states. First, the state estimates and the national estimates are derived independently, so there is no guarantee that the national employment change would match the ‘sum-of-states’ employment change, but let’s assume they do for this example. State ‘A’ creates 50,000 jobs during the period, state B loses 20,000, and state C loses 30,000. If the sum-of-states matched the national estimates, the U.S. change would be zero. So how does someone characterize the contribution of State A (50,000 job gain) to the national change of zero jobs?



From: Selby, Gardner (CMG-Austin) []

Sent: Thursday, May 09, 2013 10:53 AM

To: Abbot, Cheryl - BLS

Subject: RE: Fresh inquiry


This caveat in mind, is there a way to reach such a conclusion using BLS figures?

11:01 am

Again, BLS just wouldn’t state it that way – particularly when we’re in a period in which some states are gaining and some are declining. If I was going to compare the job change for the U.S. to a state, I would simply characterize it the way your other email did, as a straight-forward comparison of the two different series: ‘…..while the U.S. lost 2.5 million net new jobs over the last five years, Texas created 530,000 net new jobs.’ Or I could use a percent comparison (these are totally made-up percentages!), ‘…..while the U.S. lost 7.5% of total employment over the last five years, Texas added 7.5% to its total job base.’



2:21 pm

...The issue is that BLS independently develops a national employment series and the state estimates are not forced to sum to the national total. ...


Texas has been a very powerful job creator; during the period, the state created 1.75 million net jobs – and that was equal to one-third of the net jobs created nationwide.