Reporter’s Notebook: Crunching the numbers

Little did I know last week that a simple story about employment numbers would cause so much consternation in the world of social media.

That’s exactly what happened last Monday when Gov. Jim Justice and Steve Roberts, president of the West Virginia Chamber of Commerce, touted an increase of employment in West Virginia between October 2018 and October 2019 of 19,000.

The numbers, released by WorkForce West Virginia and based on data from the U.S. Bureau of Labor Statistics, showed that the seasonally adjusted civilian labor force numbers at 801,000 for the first time since March 2013. Total employment was 763,000, which was the highest that number has been since January 2009 toward the tail end of the Great Recession triggered by the housing bubble popping.

During the press conference, those numbers were questioned. A different set of BLS numbers were cited, showing that total seasonally adjusted employment was 733,300, or 29,700 less than the number used by WorkForce West Virginia. Another story later that day cited economists who also questioned the state’s numbers.

So, was Gov. Justice using trumped up employment numbers and trying to mislead the media? The short answer is no. The long answer follows below.

I had questions about the Governor’s numbers as well, at least until I did some digging on the BLS website and found the numbers WorkForce West Virginia was using. I also have to give a shout-out to Aaron Sojourner, a labor economist at the Carson School of Management at the University of Minnesota, who made me aware that BLS has two kinds of employment data.

The first BLS dataset is the payroll survey, which counts based on payroll reports of employers. The second is the household survey, which is done for the BLS by the U.S. Census Bureau and is based on monthly interviews. When you see stories in the paper talking about the national and state unemployment rate (West Virginia’s is 4.8 for October), it’s based on the household survey.

The Indiana Business Research Center at Indiana University’s Kelley School of Business says “the household and payroll data complement each other in that each provides significant types of information the other cannot. Population characteristics, for example, are only obtained from the household survey, whereas detailed industrial classifications are much more reliably derived from the payroll survey.”

Indiana, as well as West Virginia and many other states, uses the household survey for employment numbers. The payroll survey might be preferred by your favorite economists, but the household survey does a better job of including self-employed people — entrepreneurs — in their numbers. The payroll survey does not include the self-employed or business owners. It also doesn’t include workers on unpaid leave or even workers on strike.

If you use the 733,300 October employment number from the payroll survey and subtract from the October 2018 number, the state only saw a 3,200-increase in employment in the past 12 months. That’s a far cry from the 19,000-increase for the same period of time using the household survey, but that doesn’t mean the 19,000 is or is not correct. It means it’s a range.

The question isn’t whether the Governor’s Office used incorrect numbers. They didn’t. The question is whether the numbers the Governor touted last Monday are too optimistic.

I think they are, which is why I asked Justice at the press conference whether the state was doing a touchdown dance too early (based on the social media reaction to my story you would have thought I didn’t challenge the Governor at all. Feel free look it up on YouTube).

We just had a major coal producer file for Chapter 11. The Ohio Valley has already taken employment hits with the closing of the Ohio Valley Medical Center. The natural gas industry in the state is slowing down, with pipeline construction workers gone until the court issues regarding the Mountain Valley and Atlantic Coast pipelines are worked out.

Just look at the year-to-date personal income tax collections since the start of the fiscal year in July. They’re 2 percent below estimates and were down 10 percent below estimates for the month of October. Is our rocket ship ride on an upward trajectory, or are we at the top of the first incline of the rollercoaster about to come racing down? Naturally Gov. Justice and Roberts with the state chamber are optimistic.

We’ll know for sure going into the new year.


Gov. Justice, speaking at that same Nov. 25 press conference, complained again about not being able to name the drug recovery and job training program Jobs and Hope after himself.

“You can’t put my name on it. It’s the craziest junk I’ve ever seen,” Justice said.

The program, first unveiled at last January during Justice’s State of the State address, was originally named JIM’S Dream. If you recall, “JIM’S” stood for “Jobs In Making U Succeed.” If you also recall, there is no “U” in “JIM’S,” so they turned the apostrophe into a letter “U.”

Due to the Trinkets Law that went into effect in 2016, there are limits to what government officials can put their name on. So, the Governor’s Office renamed the program Jobs and Hope West Virginia.

While the Trinkets Law limits what an official can put their name on, it didn’t stop Justice from putting a name of a government program on his political signage. I’ve covered a number of Justice campaign town halls and I was going back through photos I’ve taken and noticed that on some of the posters used promoting Justice’s accomplishments, a phrase is placed above his logo.

“Jobs and Hope.”

I don’t know if this goes against any state ethics or campaign law, but it surely doesn’t look great.

Contact Steven Allen Adams at sadams@newsandsentinel.com


Today's breaking news and more in your inbox

I'm interested in (please check all that apply)