Tuesday, November 12, 2013

Pa.'s claim of 200,000 Marcellus jobs doesn't add up

Gov. Tom Corbett
The Associated Press and StateImpact PA are pushing back a bit on Gov. Tom Corbett’s claim that the natural gas industry supports 200,000 jobs in Pennsylvania. Unfortunately, thanks to the ineluctable "he said/she said" conventions of news reporting, neither story really conveys unequivocally just how questionable the claim is. So let me offer my take.

First of all, no one disputes the direct jobs numbers. As the AP reports, “core” oil and gas extraction and pipeline jobs grew by 17,414 to 28,155 in four years. Those are Marcellus jobs, no doubt about it – it’s not as though anyone else was digging wells at the time. So we're good so far. 

But 17,414 new jobs (or 28,155 total jobs) are a far cry from 200,000. Where do the rest come from?

As AP and StateImpact report, they originally came from the Department of Labor and Industry’s monthly “Marcellus Shale Fast Facts” report. The Fast Facts include a large table showing the number of jobs in “Marcellus Shale related industries.” Here's the September edition: 

It's a bit hard to see, but note the 203,814 jobs at the bottom there. Those are the "200,000 jobs" we're talking about. 

At first, people assumed these were actual Marcellus-related jobs. Then someone realized the agency was just reporting statewide category totals, without making any effort to determine what fraction of them actually had a connection to the gas industry. Think about that: It's like measuring the effect of building a stadium in Pittsburgh by including all the construction workers in Philadelphia, Altoona and Scranton. 

The DLI's disclaimer on this point is interestingly cagey: 

While the vast majority of Marcellus Shale related employment can be found in these industries, not all establishments in these industries are involved in Marcellus Shale.

Since "not all" can mean anything down to "almost none," the DLI has granted itself a fudge factor roughly as wide as its data. 

None of the news stories I've seen convey how truly off-base this table is. Take a look at it, and think about the real world. The guy laying down a sewer main in York? He’s a Marcellus Shale worker, under “water, sewer and related structure construction.” The woman driving a bakery truck to Lancaster’s Central Market? “General freight trucking, local.” Every engineer in Pennsylvania counts. Every road worker.  Every "non-residential site preparation contractor." It's absurd. 

Perhaps in response to such criticism, the Corbett administration has backed off citing the Fast Fact numbers in the past few months. Officials now say the 200,000 number really refers to the industry's impact on overall economic activity, the so-called multiplier effect. From the AP story:    

For instance, [administration energy executive Patrick] Henderson said, any tally should count the extra employees hired at a convenience store and gas station to accommodate the heavy patronage by drilling hands. In any case, the exact number is not as important as is the idea that a significant number of people base their livelihoods on the industry, and policymakers need to be mindful of that, Henderson said.

Well, it's interesting that an exact number used repeatedly in speeches and ads and industry publicity turns out to be "not important" when it's questioned. That criticism aside, clearly at minimum the Corbett administration wants us to think the 200,000 figure is the right order of magnitude. Is it? Here’s another chart from the Department of Labor and Industry:

As you can see, 200,000 is roughly the number of jobs Pennsylvania has gained back since the recession bottomed out in early 2010. In other words, if the Corbett administration’s assertion is correct, than to a first approximation, all of Pennsylvania’s post-recession jobs gains are Marcellus jobs. Moreover, we managed to add 200,000 jobs without reducing our unemployment rate below 7.7 percent (as of August, the most recent figure). Does either of those claims seem plausible to you?

Think about that latter claim a second. Pennsylvania has about 500,000 unemployed people. If the Marcellus is the only thing standing between us and another 200,000 unemployed, that means unemployment would be 40 percent higher without the gas industry, putting it at 10.78 percent. 

That's two percentage points higher than the 8.8 percent unemployment Pennsylvania had at the height of the Great Recession. It's 1.28 percentage points higher than in beleaguered Nevada, which thanks to its historic housing bust has the worst unemployment in the U.S. In other words, by attributing 200,000 jobs to the Marcellus industry, the Corbett administration is essentially accusing itself of disastrously mismanaging the rest of Pennsylvania’s economy!
Fortunately for us all, it's not true: 

Tim Kelsey, a Penn State professor of agricultural economics and a co-director of the university's Center for Economic and Community Development, said an academic economist's very rough calculation of the job impact would be closer to multiplying by two the increase in core jobs - 17,414 - to get a figure of about 34,000 or 35,000. And that includes ripple effects, Kelsey said.

So why say otherwise?

"You get to a certain number where people realize it's a big enough deal that you have to pay attention to the public policy that affects it, and that's the goal," Henderson said. "Without question, the number of Pennsylvanians working in oil and gas is a big number."

Translation: “If people believe it’s this big a jobs creator, they won’t oppose the industry-friendly policies we favor.” If fudging the numbers by a factor of six is what it takes, then so be it.


  1. Recruiters as well as recruiters will certainly acquire a ton regarding purposes for every single Jobs administrative, so one of these are not able to pay for to take the time to study every single CONTINUE complete.