Here’s a good point from Mark Thoma and Pavlina Tchernevna about fiscal stimulus. If what you want to do is increase the budget deficit by some amount to decrease unemployment, the most efficient way to do that on a dollar-per-dollar basis is to spend the funds employing the unemployed people on public projects of some kind. But though this is a highly efficient way of getting people into jobs it’s not at all the most efficient overall allocation of the money. But that in turn means that if you spend a lot of time and energy trying to design your stimulus to be efficient in the sense of getting real value for your money, that you almost certainly won’t be getting the most employment bang for your buck.
A stimulus designer, in other words, faces a choice at the margin between maximizing GDP and maximizing employment. And when constructing ARRA both congress and the administration tended to choose GDP. Over time of course GDP growth will lead to employment growth. But as we know, there are substantial lags in this process and perhaps a case to be made that it makes more sense to do things the other way ’round. Certainly one concern I have with regard to the tangible infrastructure projects that ARRA is funding is that the administration’s zeal to avoid anything that looks like “waste” is itself creating a wasteful level of delay and waste-monitoring, when it would make more sense to just plunge ahead and get projects out the door.