Joe Weisenthal of Business Insider writes the following:

Over 2 years since the Recovery Act was signed, the economy is still really ugly. Today’s jobs report, while slightly better than expected, is still really awful.

But that doesn’t mean the stimulus failed.

The CBO estimated that it created about 3.3 million jobs. Or maybe it “created or saved” that many jobs, but it clearly had a positive impact on employment, which was the primary goal. Also, GDP resumed growing not long after the stimulus was enacted.

But clearly the stimulus didn’t create a lasting or robust recovery, and this is where the Keynesians get it wrong.

As we’ve discussed in this space, however, the CBO’s estimate of the ARRA’s impact on the labor market is of limited utility. 

Drawing on Feyrer and Sacerdote, we have reason to believe the following:

(1) Support programs for low-income households were expansionary;

(2) Grants to states for education did not create jobs, though we can assume that the grants crowded out some amount of borrowing on the part of state governments.

And the work of Sahm, Shapiro, and Slemrod suggests that 

(3) the reduction in withholding that was an important component of the 2009 stimulus law boosted spending by far less than one-time payments in 2008. 

My view on fiscal stimulus is in a sense not even a view on fiscal stimulus. When it is unusually cheap to do things you ought to do anyway, it makes sense to do those things. For some of our interlocutors, this implies that we should borrow lots of money to finance public spending, because it remains relatively cheap to borrow. I would suggest that we should borrow to finance public spending that serves a lasting public purpose, and I imagine I have a narrower sense of what that entails. For example, I’d want to make conditional rather unconditional grants to states, with an eye towards reducing the cost of providing fringe benefits, etc.

The paradigmatic example of the kind of spending that would make sense is infrastructure spending. The ARRA was drafted under the assumption that we weren’t about to experience a multi-year Great Contraction. Had that been widely understood, one assumes that a multi-year program of public infrastructure investment would have taken higher priority. As Alice Rivlin argued at the time, short-term measures, like grants to state governments and UI extensions, etc., were logically separable from infrastructure spending, which should have emphasized long-run viability, the creation of sustainable funding mechanisms, etc., over “shovel-readiness.”

I’m hoping that Josh Barro will write about the idea of transportation infrastructure as an asset. As Josh explained to me, transportation infrastructure is often understood by taxpayers as the kind of thing that we pay off and that’s it — it’s wrong to charge tolls, etc. But in fact transportation infrastructure is a depreciating asset that requires infusions of new investment and it also has the potential to serve as an income-producing asset for taxpayers.

I can’t say how many members of Congress would have embraced a package something like the following:

(1) Let’s make it much easier for state and local governments to impose congestion charges, to better deploy existing resources and perhaps to generate surpluses that could be used to subsidize other government functions;

(2) Let’s spend a sizable sum on large-scale transportation projects that can meaningfully be described as being in the national interest, and perhaps make grants to states for a variety of other projects on a competitive basis;

(3) Let’s insist on fixed-price contracting and suspend various labor regulations that would otherwise raise the cost of infrastructure projects, because, remember, the whole point of this exercise is taking advantage of the unusually low cost of inputs. 

Andrew Samwick, to his great credit, has been making something like this argument for the better part of the last four years, and my sense is that he has been largely vindicated. Now, I take strong exception to the slapdash arguments for infrastructure made by people like Larry Summers, who I wouldn’t want serving on whatever independent commission decides on where the spending ought to go. But the basic idea is a good one. 

I’d also argue that in January of 2009, a fair number of Republican lawmakers would have been amenable to something like this approach, though I imagine the number would be smaller at present. 

I’ll also add that I’m very suspicious of the idea of a national infrastructure bank, which sounds like an opportunity for off-balance-sheet borrowing. Let’s be as transparent as possible.