First, let's establish the facts. Since the oil and gas industry rushed into the state in 2007, there's been a 20 percent spike in overall employment. The industry directly employs between 40,000 and 50,000 people, and many more indirectly.
Ask anyone counting shale jobs, and that's what they'll tell you.
If you're in North Dakota.
In Pennsylvania, shale job numbers are harder to pin down. The estimates can vary by a factor of 10. The debates are heated.
"We keep hearing all the different claims," said Rep. Steven McCarter, D-Montgomery. "The governor routinely says 200,000-plus [jobs]. You have another economist saying it's only one-seventh of that. Then the governor comes back and says many of these are ancillary jobs ... the answer is: We don't have good information."
In June, Mr. McCarter introduced a resolution directing the Legislative Budget and Finance Committee of the state House to study the number of jobs being generated by the industry -- paying specific attention to jobs in ancillary industries, workers from out of state, and what wages and benefits are offered.
The bill has gotten no traction, but Mr. McCarter -- who convinced 18 co-sponsors to sign on -- plans to keep pushing the issue and reintroduce the legislation if it dies in committee.
"We all use sound bites, and we use data any way we want to use them," Mr. McCarter said.
But, he said, it's important to have unassailable statistics when debating whether to tax the industry or to give it incentives.
North Dakota has it easy.
As Dean Bangsund, professor of agribusiness and applied economics at North Dakota State University put it, "We don't have any muddling of the causation here."
In other words, the shale boom there started with a relatively empty slate.
"There was nothing major happening here," Mr. Bangsund said. "Along comes the Bakken [Shale] and we've got numbers shooting up tremendously. I don't think there's any doubt, from an employment perspective, that the development of shale [is] the biggest driver of change in our economy."
It helps to have a nice, flat baseline to track the impact of a new and growing industry. It also helps to have a state population that's one quarter the size of the Pittsburgh metropolitan area. Against that backdrop, the addition of tens of thousands of jobs really stands out.
In Pennsylvania, whether you go with the conservative estimate -- counting only the jobs that touch a well pad or a pipeline -- or if you bunch in all jobs in 30 other related industries, oil and gas employment still comprises a sliver of the state's workforce -- between 0.4 percent and 3 percent of the employed population.
"There have been jobs created by shale," said Stephen Herzenberg, executive director at Harrisburg-based Keystone Research Center which advocates for a conservative estimate of shale jobs. "In the period of time over which this industry has grown, those jobs have been particularly needed.
"But it's still true that the scale of the jobs is modest."
The Pennsylvania Department of Labor and Industry calculates Marcellus Shale employment by counting jobs in six core industries and jobs in 30 other industries that the state designates as ancillary. These include areas such as truck transportation, power plants, sewage treatment plants, and engineering.
That counting mechanism underestimates some jobs, since it doesn't account for lawyers negotiating oil and gas deals or diner waiters hired to staff a suddenly in-demand shack in Washington County.
But it inflates other jobs by placing the entire trucking industry into the count. The numbers don't say what percentage of truckers on the road are hauling frack water and which ones are delivering products to Walmart.
North Dakota has similar ambiguities, according to Michael Ziesch, manager of that state's labor market information center.
North Dakota needs to get a better handle on ancillary jobs because, according to Mr. Ziesch, the distribution of oil production taxes to municipalities depends on an accurate industry job count. That's why the North Dakota Legislature gave the department $120,000 to study the issue.
A team of two researchers at the labor department is culling through the records of every company that they suspect might be involved in the Bakken. If they can't tell from the documentation whether a firm works in the shale, they'll reach out directly.
"People on the ground in western North Dakota need to know [this] because they're trying to plan for an economy several times larger than what's they've ever seen before. The problem we're having is nobody knows where this thing is gonna end up," Mr. Ziesch said.
The Keystone Research Center is pushing for something similar in the Marcellus Shale region. It wants to see an interstate government commission tasked with coming up with a robust model of shale's economic impact in this region.
Mr. McCarter says he's after the same thing. "Pennsylvania needs an economic boom," he said. "If this industry's not helping us, we need to tout other things."
He's also interested in determining how many workers are brought in from other states to serve the shale industry, an issue that some have used to downplay the impact of shale development on Pennsylvania's economy.
In Pennsylvania, companies often rush to announce they're tapering the influx of out-of-state workers and replacing Texas license plates with Keystone ones, because they know that bringing in workers from elsewhere is unpopular.
In North Dakota and specifically in the Bakken Shale counties, there's a different dynamic at play, Mr. Bangsund said.
Job openings outnumber job seekers. Unemployment in Bakken counties is about 1 percent. And the state is looking to attract transplants. That's one reason it's looking to beef up its housing stock -- to accommodate not only the incoming workers but their families.
"If we want to grow our workforce, we need to bring people into the state," he said.
Anya Litvak: firstname.lastname@example.org or 412-263-1455.