I’ve read through that headline-grabbing AP report on the impact of stimulus-funded road and bridge spending on employment about five times, and I’m still not entirely clear what the AP’s reporters discovered. But I think it’s this: That there was no correlation between the movement of county unemployment rates and the presence or absence of stimulus-related construction projects in each county.
Unemployment rates can be a tricky measure—they’ll often rise in counties where the economy is getting better, because an improving economy lures formerly discouraged locals back into the labor force (which is made up of those with jobs and those actively looking for them) and sucks in job seekers from elsewhere. Local unemployment measures can also be subject to a pretty significant margins of error. So I’m not sure the AP study really proves anything at all.
Still, it makes sense that road projects wouldn’t be massive job creators. They may still be in India, but here in the U.S., where most of the work is done by a few skilled operators of a few big machines, they’re just never going to be a big part of the solution to our employment problems. The same would seem to be true for infrastructure spending in general. The main value of such projects is in the infrastructure they create (or repair), not in the jobs they create as part of the construction process. And the thing is, our infrastructure needs a ton of work. Trying to justify all that work as a jobs-creation measure seems like a backward way to go about it. But improving-our-infrastructure-because-it-needs-improving hasn’t been much of a political winner for the past few decades.