Monday, October 03, 2011

Desperately Seeking Spike

In Pennsylvania, the 5 counties with the largest number of permitted Marcellus Shale pads are Bradford, Tioga, Lycoming, Washington and Susquehanna respectively. 

Washington County is by far the largest county among the group. It is also part of a larger metro area.  So set Washington aside just for a moment and think about the other 4 which are the core of Marcellus Shale development in Pennsylvania to date.. especially in the NE and north central parts of the state.

I just added up the employment counts for those 4 counties over the last 5 years and in aggregate this is what you get for the time series:

I must have made a mistake.


Blogger Jim Russell said...

Is there an employment boom in any county associated with shale gas or oil, anywhere in the US? I ask because a quick review of that top-10 county list from the BLS doesn't indicate anything of the kind:

Elkhart, Ind. +6.2
Ottawa, Mich. +4.7
Washington, Pa. +4.3
Prince William, Va. +4.3
Benton, Wash. +4.3
Butler, Pa. +4.2
Loudoun, Va. +4.2
Williamson, Tenn. +4.1
Washington, Ore. +4.0
Collier, Fla. +3.8

Monday, October 03, 2011 5:20:00 PM  
Blogger Bram Reichbaum said...

Hmm well ... a great recession hit in mid-to-late 2008 ... should we be wondering, "How bad would the graph look were in NOT for Marcellus?"

Tioga County was one of that aforementioned Congresspeson's alleged of "0% unemployment" counties, though. Imagine how that must be skewing the composite sample.

Monday, October 03, 2011 5:35:00 PM  
Blogger C. Briem said...

ah, the unknown of the counterfactual. How much of the 100K+ new Marcellus jobs are hidden in the counterfactual depression that would have hit those counties?

Monday, October 03, 2011 6:30:00 PM  
Anonymous BrianTH said...

Couldn't we look at otherwise comparable (I guess that would be non-metro-area) non-Marcellus counties in PA to get an idea of what the counterfactual would look like? In other words, we need a control, and isn't that what we should be using?

You could also do something like plot job change versus permit count for your set of comparable counties, and see if there was a correlation.

Monday, October 03, 2011 9:34:00 PM  
Anonymous MN said...

You're all missing the fact that the people doing the drilling and fracking and water-hauling and sand-hauling, etc., don't actually live in the counties they are drilling in (with perhaps the exception of Washington County). And they often stay in hotels hours away from the sites. I'm not defending the claim that the high growth rate in Washington is all from Marcellus, I'm just trying to say it's more complicated than lots of permits leads to employment growth.

The truth is always somewhere in the middle.

Monday, October 03, 2011 11:06:00 PM  
Blogger C. Briem said...

Man Camps!

Oh, I don't think the point is being missed... being denied by many, but not missed. As I have said here before, it's like much of Pennsylvania is a big North Sea oil rig.

Thought it is really remarkable there has not been any net job growth at all in those 4 core Marcellus counties. Bram's counterfactual caveat not withstanding.. if 1% of the 140K jobs some like to throw out there were created in those 4 counties it would be a noticable jump.

Monday, October 03, 2011 11:16:00 PM  
Anonymous MN said...

I have an idea... let's all take a road trip! No, wait, there aren't any hotels up there. Lots of trees though. Would be beautiful this time of year.

Monday, October 03, 2011 11:35:00 PM  
Anonymous Anonymous said...

I've been having a cordial email conversation with the reporter who printed the "zero unemployment" quote. In the course of that, I checked Wikipedia and found this definition of "full employment":

"The United States is, as a statutory matter, committed to full employment (defined as 3% unemployment for persons 20 and older, 4% for person aged 16 and over), the government is empowered to effect this goal, and a job is a right. [T]he US has, as of 2011 never achieved this level of employment, nor has such a reservoir of public employment been created."

Hate to say it, but it appears that, for whatever reason, places like Washington and Butler are close to full employment, despite the down economy. And, being the generous sort, maybe the congressman meant full employment, not zero unemployment.

Tuesday, October 04, 2011 9:31:00 AM  
Blogger C. Briem said...

The problem is that those counties in his district (he referred specifically to HIS counties) are not anywhere near close to 4% unemployment even. They are all quite high by relative or historical standards. Tioga County 8.2% for example. McKean is in his district: 9.4%. Carbon County: 11.2%.

Nothing is near full employment by the most liberal of definitions.

The lowest unemployment rate in his district I think is Centre County and that I am just speculating is more because of Penn State than any shale development. So even in a relative sense within his district it makes no sense.

and then it isnt all that big a deal is it? as I pointed out earlier.. the entire US hit 4% unemployment in the late 1990's.. the entire US!

but that Wikipedia entry is curious anyway. There is no agreement on what the definition is of full employment anyway.

Tuesday, October 04, 2011 9:42:00 AM  
Blogger C. Briem said...

and... Washington County = 7.9% unemployment. Butler County = 7.2%. I am not sure anyone has considered those 'full employment' since the depths of the depression.

Tuesday, October 04, 2011 9:59:00 AM  
Anonymous MN said...

Not just man camps! I don't even know of any "man camps" in SWPA. I'm talking about men - and some women - who live here, raise families here, pay taxes here, buy houses here, whose employers are here (in SWPA) but they travel in crews up to places like Tioga County. So employment in Tioga County wouldn't increase - at least not by much. Maybe some retail and leisure/hospitality jobs.

Tuesday, October 04, 2011 10:19:00 AM  
Anonymous The Wiz said...

This has to be one of the hardest industries to quantify. They use transient crews that move from well to well, county to county. The crews are so specialized that some will spend less than a week at a well pad. They are a mix of state and out-of-state talent. Much of the work is subcontracted, much to out-of-state companies that also have a mix of talent. And a large number of people have come to Pa to cash in on the rush, many forming their own businesses.

Add with the unique payment system of large bonuses to land owners followed by larger royalties once a well is in production, combined with the complex tax issues involved in all this, and it is very difficult to pin down the total economic impact of the industry.

If nothing else, it will keep the economists busy for some time trying to figure it all out.

Tuesday, October 04, 2011 11:18:00 AM  
Anonymous The Wiz said...

It would appear from the graph that total employment in late 2010/early 2011 was higher than the peak in 2008 before the economy crashed. Can't be too many areas of the country where total employment is now higher than before the crash.

Tuesday, October 04, 2011 11:24:00 AM  
Anonymous BrianTH said...

1% of 140K is 1400. It seems entirely plausible to me that the job count in those four counties combined is considerably more than 1400 over what it would have been (and rather than just guessing, why again aren't we looking at non-Marcellus, but otherwise comparable, counties?).

But maybe Chris meant 10%--14,000 would be a little harder to miss, I suppose.

Tuesday, October 04, 2011 12:27:00 PM  
Anonymous Anonymous said...

Sorry Chris, I misinterpreted Jim Russell's stats on employment in Butler and Wash. Co. I take it those numbers reflect percent inc. in employment, not percent of unemployment.

Tuesday, October 04, 2011 1:15:00 PM  

Post a Comment

Links to this post:

Create a Link

<< Home