So will that infrastructure spending work?

As Marc notes below, the stimulus bill passed the House with zero Republican support. I see the token corporate tax breaks really paid off.

Much of the pre-vote skepticism centered on the spending and, in particular, the infrastructure bits of the bill. Lots of Republicans opposed the infrastructure portions on the theory that building projects take a long time to start a long time to complete, and thus aren't effective short-term stimulus. And lots of Democrats asked that infrastructure portions of the bill be made larger, on the theory that many building projects are actually  "shovel ready" and can thus kill two birds (short-term stimulus and long-term growth) with the same bridge.

So was it worth fighting over? Should we have slashed it or stuffed more of it in? I dunno. But one paper that I find helpful on the subject is this Brookings/Hamilton report on infrastructure, co-authored by Manasi Deshpande and Douglas Elmendorf (now of the CBO, and yesterday's report on the stimulus). I like the paper for a couple of reasons: First, it was written in July 2008 -- after a general debate over fiscal policy started, but before this specific debate over Obama's stimulus bill. And, second, it's a paper that comes from the erstwhile think tank of not just Elmendorf but Peter Orszag (of the OMB) and Larry Summers. And it comes with a roundtable discussion on infrastructure between Elmendorf, Summers and others.

I found it helpful in thinking through some of the infrastructure/stimulus skepticism, so I've posted some thoughts from the paper and roundtable after the jump:

1. It's hard to know how much we should be investing in infrastructure. In part this is because measuring the demand for infrastructure is easily distorted: a lot of us use infrastructure more-or-less for free (e.g., driving) and almost never pay proportionally to its cost. This leads to inefficient overuse. And in the case of something like broadband, it's hard to how much we should be spending because we haven't exactly been investing in broadband for centuries.

2. Relative to the rest of the world, the America's infrastructure investment is about average. Relative to American history, it's declined recently. (Investment as a percentage of GDP has remained about constant -- 2.5 percent. But net investment -- that is, investment adjusted for infrastructure depreciation -- has fallen from 2 percent of GDP in 70s and 80s to 1 percent of GDP in the 90s.) This might mean that we aren't investing enough.

3. The federal government does relatively little infrastructure spending -- the vast majority of the money goes through the states. And the states tend not to be super-efficient about spending it. Surveys of the states actually show that officials list political favors and vote seeking as bigger reasons for pursuing particular projects than they do, say, the public interest.

4. But it's not totally clear that federal infrastructure money can be spent with great speed. Typical federal highway spending only pays out about 25 percent of the allocated funds in the first year. You can, however, find examples of money being spent quickly, given the right spur. Several hundred million dollars in federal infrastructure commitments were made in the wake of the I35-W bridge collapse in Minneapolis. Nine months later, 86 percent of that money was spent.

5. Why should we have hope that there will be rapid spending from the current bill? One reason is that lots of ongoing projects have been slowed or stalled, and they might be easy to pick up again with additional federal dollars.

6. Why have these projects slowed or stalled? In part this is due to state and local budget constraints, some created by the financial crisis itself. And in part it's due to the price of inputs for infrastructure, which have/had increased 70% since 2004. This means that if you allocated funds for a five-year infrastructure project in 2004, you probably probably wouldn't be able to complete it on schedule.

7. Another consideration is the structure of the American labor force. The past five years have been a period of decline for manufacturing, especially for men with little education. But much of that slack had been picked up by construction jobs created by the housing bubble. These workers are now faced with a double whammy: no manufacturing jobs, and a bubble that is no longer a cushion. This suggests that stimulating employment in construction would be relatively easy.