STRANGE, isn't it, that the unemployment rate in America is so high, given that you can't go ten minutes without someone collaring you to tell you how many jobs they're fixing to create in the energy sector. Barack Obama, slightly on the back foot after having failed to create as many "green jobs" as his campaign predicted last time around, nevertheless touts clean technology as a key component of his proposed American Jobs Act. Rick Perry claims that he will create 1.2m new jobs in the energy sector. Mitt Romney has actually put the figure a bit higher: nearly 1.5m. The left argues that they're actually not ambitious enough; the Center for American Progress, for example, hits Mr Romney for scoffing at the idea of "green jobs": 2.7m of them are "right in front of ya, Mitt." An e-mail drops in my inbox from a solar company, saying that five years ago you could fit the entire industry in a ballroom, and now solar employs 100,000 people in the United States. A billboard for oil and gas glibly promises several million new jobs, if America would just support the industry. People who support the controversial Keystone XL pipeline, which would ferry oil from the tar sands of Canada to the refineries of Texas, point out that building the pipeline would create 20,000 jobs, just like that.

At the risk of being obvious: energy policy is not a jobs programme. Here are three reasons why politicians shouldn't try to create jobs through energy policy: it's ambiguous, it's inefficient, and, most importantly, it's undesirable. 

On the first point, energy is, of course, an industry that employs millions of Americans. Some parts of the energy industry are growing, and in some cases the jobs thereby created are good ones. The Brookings Institution, for example, reports that "clean economy establishments" grew at a 3.4% annual rate between 2003 and 2010—yielding generating a "clean economy" with 2.7m workers, as mentioned by the Center for American Progress—and that segments such as wind and solar "added jobs at a torrid pace, albeit from small bases". As we've discussed before, however, the current size and health of the "clean energy industry" really depends on what you count as a green job.

Projections about future job numbers depend on slews of counterfactuals about technology, policy, domestic demand, and on global markets. At the Council on Foreign Relations, for example, Michael Levi makes a courageous effort to break down exactly where Mr Perry's 1.2m figure comes from. Even if one wrestles with the counterfactuals and comes up with a number, the analysis is still vulnerable to attack from additional counterfactuals. With regard to the Keystone XL pipeline, for example, advocates cite a June 2010 report from The Perryman Group (PDF) which reckons that building the pipeline would create 20,000 jobs right away, and many more over the medium- to long-term. A new report from Cornell's Global Labour Institute, however, counters that construction will only create a couple thousand jobs, most of them temporary, and that the net employment impact may actually be negative, if the pipeline ices investment in renewable energy. These methodological questions and counterfactual complexities arise in most analyses. It doesn't mean they're not worth doing. It does mean that proferred numbers be taken with a grain of salt—as a lagniappe, perhaps, rather than a justification. We should be even more circumspect when we're talking about energy jobs, given that our energy portfolio should be informed by more important considerations.

Secondly, and relatedly, using your energy policy to create jobs is likely to be inefficient, if not counterproductive. As a general rule, and although it varies by segment, energy is capital-intensive rather than labour-intensive. With regard to oil and gas, for example, the Dallas Fed finds that between 1997 and 2010, a 10% increase in energy prices spurs a big increase in rig count (6.2% for oil, 4.9% for gas), but there's only a modest effect on employmentin oil (0.36%) and none for gas. This shouldn't be surprising; the great value of oil is in the commodity itself, not in the knowledge or service or craftmanship of the barrel. To be sure, the localised employment impacts of the energy sector may be profound; around 2009, southeastern Louisiana had some of the lowest employment rates in the country, and even in the immediate aftermath of the Deepwater Horizon explosion, locals were fretting about the job-killing impact of a moratorium on offshore drilling. And certainly some segments of the industry are realising job gains that exceed the national average. But if the top priority is job creation, then energy policy isn't the best way to get there. The Bureau of Labour Statistics has an apolitical list of the occupations that are projected to grow most quickly between 2008 and 2018. Some of them require training. That would be a better place to start. 

The most important reason not to arrange a national energy policy with a view towards maximising job creation is that such an effort will come at the expense of other priorities, which should take precedence. That is, we want energy that is plentiful, cheap, sustainable (if not renewable), clean (relatively), safe (again, relatively), predictable, and broadly or equitably accessible. The exact balance of those concerns will be informed by contextual preferences and capabilities, but in all cases it is a balancing act and almost invariably a tricky and controversial one. So the idea of pinning our hopes for job creation on the energy sector, and goosing policy accordingly, strikes me as fundamentally misguided. It may be a factor in favour of public-sector support for various initiatives: a stimulus measure that pays a hundred people to retrofit a bunch of houses, keeping them employed for a year and yielding immediate energy efficiency gains. But on a national scale, energy policy should be informed by overarching and enduring concerns.