The Elasticity of Supply of Air Traffic Controllers

If you'd asked me before this air-traffic-controller-sleeping bit came about, I'd have told you that I believe the supply of air traffic controllers to be rather inelastic; I presume it takes some amount of specialized training and at current wage levels there aren't a lot of unemployed air traffic controllers. (Perhaps I'm wrong.) But as I read the reports of a new FAA regulation requiring two air traffic controllers at night, I'm having some elasticity doubts.

In an interview this morning the head (or perhaps just a representative) of the Department of Transportation said that they've got the air traffic controllers to fill this new regulation and the money to do so. Now, granted, when a government agency says they've "got the money" to undertake an expansion, that has a little different flavor than when Wal-Mart or FedEx says they've got the money. But in light of the tight budget situation now (and, in fact, budget cuts projected for the Department of Transportation were mentioned in the interview), it seemed like this wasn't an expansion-in-expenditure scenario.

So what are the possibilities here?

1. My assumptions above are wrong; there is slack in the air controller labor market and the new regulation will reduce it.

2. The government is putting on a front that this won't cost money when in fact it will cost more money. Relatively speaking, I would guess this money number is small, but still this is a solve-it-with-money scenario, not a utilize-untapped-resources situation. If there's not a large amount of unemployed air traffic controllers, then this is also a raise-wages-to-incentivize-entry-into-the-industry situation. But what's the latency in assuming a job in this industry? 6 months of training? 3? 12? More? I don't know, but I feel it's got to be at least a number of months. Again, not saying it should be-- but I'm guessing that it is.

3. They are only requiring 27 towers to abide by these new rules. Compared to the number of overall air traffic control centers (of varying sizes), this is small. However:

a) These sites are geographically separate; we're not taking a pool of laborers and requiring extra hours at a variety of locations. Along the lines of issue #1, are there spare air traffic controllers to be had, or are more hours required of the existing working pool?

b) Not unrelated to this is the fact that air traffic controllers are thought to be over-worked right now as it is (not judging the merits of that claim, but that's still relevant), so where exactly these extra working hours comes from is still unclear in my mind.

Something--somewhere-- is off, either my perceptions of the state of the market or in the stories being told. Knowing me I probably gravitate to #2 but I think that oversimplifies the problem.

Justin, will you be leaving for your new air traffic controller position soon?