In his
column arguing that Texas is a public policy success story, Douthat quotes a
blog post by Matthias Shapiro: "[T]ake the energy sector completely out of the equation and Texas is
still growing faster than any other state."
Shapiro's methodology is to look at the sectoral breakdown of employment numbers and find that the energy sector and related sectors account for "about 25% of the job increases in the last year."
Sigh. So imagine a small town in the middle of farm country. There are a few non-agricultural jobs - teachers, police, a generalist lawyer (wills, real estate, family law), maybe a family doctor, maybe a baker, a barber, grocer, whatever. Now a factory comes to town, employing 800 people. There is a boom in housing construction for the (roughly) 800 new residents. Non-farm employment soars - more barbers, more bakers, more teachers, more police, more everything. There could even be some professions in town that weren't there before. The total increase in employment is 2,000 jobs, but 1,200 of these are in sectors other than manufacturing.
Now there are rumors that the factory is leaving. The mayor is asked, "Aren't you worried about the employment situation if the factory leaves?"
"Only a little," the mayor replies. "When you look at factory employment in this town, it accounts for less than half of the new jobs. Even if the factory leaves, we will still have 1,200 jobs that we didn't have before. We'll be fine."
A prominent columnist concurs: "Nor are the jobs confined to the manufacturing sector."
The columnist quotes a blogger, who notes, "Take the manufacturing sector completely out of the equation, and the town is still growing faster than any other town in the state." The blogger uses the same methodology as Shapiro: he simply looks at the percentage of jobs in the town that are categorized as "manufacturing" or related sectors, and notes that the majority of jobs aren't even in the manufacturing sector.
So hopefully the error here is obvious. Each job in an export sector tends to support several jobs in non-export sectors such as services and local goods production. If the factory shut down, not only would 800 jobs be lost, but there wouldn't be a customer base to support most of the other 1,200 jobs that have been created in the town. Shapiro's methodology is laughably crude, because it doesn't take this into account at all.
For what it's worth, my understanding is that the aerospace industry in southern California was once estimated to have a multiplier in the 3 to 4 range - each job in that industry generated 3 to 4 jobs indirectly (I will try to find a more precise estimate and update this post). So as a back-of-the-envelope matter, Shapiro's numbers seem consistent with roughly
all of the Texas employment "boom" stemming from the energy sector. And this wouldn't be surprising: the price of oil has fluctuated a lot, but oil remains very expensive by historical standards. I imagine the Norwegians are doing just as well as the Texans, if not better, despite the notable absence of Rick Perry or anything resembling his policy agenda.
(And indeed, Norway's unemployment rate was 3.6% in 2010, according to the CIA World Factbook. And 76% of its labor force is employed in "services," so we're not talking about an oil-based economy here, right, Douthat/Shapiro?)