May preliminary employment report positive – ETSU economist questions accuracy of local Q1 unemployment data

May’s local unemployment numbers are out and the overall preliminary picture is good even though unemployment rates are up and the number of people reporting they have jobs is down. But you might want to look at the most often quoted labor market metric through more critical lenses.

ETSU Economist Steb Hipple is questioning the accuracy of the monthly household survey, which is used to determine local unemployment rates. Dr. Hipple is also a research associate for the Bureau of Business and Economic Research. His comments came during an interview during a WETS Business Matters broadcast last week to discuss his Q1 labor market report. He made similar observations in that report which can be found by CLICKING HERE but the radio interview gives additional insights.

There are two primary labor market reports each month. The first is commonly known as the payroll survey, and it’s just that. It’s a survey of employers about how many people are working and in what jobs.  This survey is the larger of the two reports so it’s considered more accurate. But it’s not available on the county or city levels.

The second report is commonly known as the household survey. It focuses on localities, how many people are working and how many who are looking for a job. It’s the report that drives the monthly unemployment rate. The household survey is a smaller sample so its accuracy is at its best on the national level although it is reported on the county and city (above 25,000) level.

Locally the payroll survey and the household survey reported conflicting stories during the post-recession recovery years. The payroll report was showing job growth. The household survey was showing lower employment and lower unemployment rates as more people left the labor force. That continued through Q1 even after the Bureau of Labor Statistics’ most current revision of the numbers.

Dr.  Hipple told the hosts of Business Matters late last week, “You can’t have one survey going up and another going down and claim they’re observing the same phenomena.”  The data we have watched and payed the most attention to is the household survey, but it’s “telling a story that is apparently wrong.” Dr. Hipple said he does not see a problem with the household data on the national level. It’s the local level where he has an issue.  Dr. Hipple is a routine Business Matters guest to discuss his quarterly sales tax and labor market reports.

Which report does he think is telling the most accurate story about local labor markets and why?

“Right now I’m hanging my hat on the payroll data.”  That’s because retail sales affirm the growth in the payroll jobs data. Unfortunately the payroll report is only available on the MSA level. The household data – once you get below the national – level “is not robust, he said. It’s subject to extensive revisions.”

In theory both the payroll and household surveys are needed for a complete picture of the labor market. The payroll survey provides a gauge of monthly change in jobs on the MSA level, but it does not include agricultural jobs and omits some self-employment, part-time and contract jobs.

The household survey provides a broader picture of employment including agriculture and self-employment. Its focus is on the number of people employed and those looking for a job.

When you look at the data on a monthly base you can see the two reports beginning to sync. But each current monthly report is preliminary and is revised when the following month’s report is releases. Usually the revisions are slight.

Both the monthly jobs and employment data are showing growth in the three-county Johnson City MSA for February through May. The telling check will come with the final Q2 numbers. The preliminary report will be available in July. The household survey adjusted numbers will available in the August report.

With the exception of March, both the household and payroll survey in the four-county Kingsport-Bristol MSA were showing growth.

The easiest way to visualize those movements is by year-to-year percentage change on the monthly basis then adjusts for the revised household report each month.

WETS posts podcasts of Business Matters. Those can be found by CLICKING HERE. When this article was filed the podcast of the most recent program with Dr. Hipple had not been posted.

May’s payroll report coverage can be found by CLICKING HERE.

Tennessee’s unadjusted unemployment rate last month was 5.7%. The U.S. rate was 5.3%.

Here’s what the local May non adjusted household report on the unemployment rates looks like:

TRI-CITIES – 5.8%, up 0.5%.

JOHNSON CITY MSA – 6%, up 0.6%


BRISTOL – 5.9%, up 0.6%

JOHNSON CITY – 6.1%, up 0.6%

KINGSPORT 6.1%, up 0.7%



  1. […] If you want the background of the different story aspects of the employment and jobs report check out the Business Matters podcast when it posted at  or read up on it at my posts  “Tri-Cities labor market continues positive trend”  or “May preliminary employment report positive – ETSU economist question accuracy of local Q1 emplo…. […]

  2. […] For almost two years the local jobs and employment numbers opposed each other. The number of jobs was increasing but employment was declining. That was the basis for ETSU Economists Dr. Steb Hipple’s comments after the release of his Q1 labor market analysis questioning the accuracy of the local employment data. That report can be found by CLICKING HERE. […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: