The Agenda

NRO’s domestic-policy blog, by Reihan Salam.

Jared Bernsten vs. Ed Glaeser


Ed Glaeser wrote a post on whether Recovery Act grants were effectively targeted to reduce unemployment. This was a follow-up to a really excellent post on how federalism is changing during the downturn, as states grow increasingly dependent on federal aid. 

I don’t think that states should cut spending now, but I’m not happy with the idea that the federal government is stepping in and eliminating the states’ need to handle their own obligations. In principle, states can create sufficient rainy-day funds and allow for sensible borrowing during economic downturns. Moreover, handling state budget shortfalls with federal bequests creates all manner of oddities. Every one of the five least dense states in the country has been awarded $948 per capita or more in recovery funds (Alaska got $1,600 per capita), while New York has been awarded $567 per capita, and Florida $452 per capita. This situation is particularly absurd since Florida’s unemployment rate is 11.8 percent and the unemployment rate in the sparsely populated Dakotas is below 5 percent.

Unfortunately, more transfers from Washington to the states may be the only politically feasible way to limit cuts in the badly needed functions of state and local governments, like schools and the police, but this seems like a bad long-run solution. Hopefully, as the crisis eases, the country can move toward a system where states keep their own financial houses in order and we set up fiscal institutions that save them from needing to slash spending during a downturn.

This strikes me as an excellent idea — provide an institutional cushion designed to ease the need for pro-cyclical cuts, but limit discretion so that we can constrain rent-seeking. But to return to Glaeser’s “Wasted Stimulus” post, he was making a fairly basic point:

There are two reasons that density predicts which states got the stimulus funds.

A technical reason is that transportation spending plays a large role in the stimulus funds and transportation spending is highly skewed toward thenation’s least dense areas.

The deeper reason is that the Senate is designed to empower states with smaller populations. One-tenth of the Senate represents the five states getting the highest Recovery Act grants per capita — Alaska, the Dakotas, Wyoming and Montana — which collectively have 3.7 million people, or 1.2 percent of the American population.

This, incidentally, is a point often made by urban liberals, and it’s a fair one. Bernstein’s reply struck me as unresponsive. He notes that Glaeser didn’t factor in spending on Medicaid and unemployment insurance, to which Glaeser said the following:

Mr. Bernstein correctly points out that there are many programs, like unemployment insurance and food stamps, that are targeted toward the unemployed. My question was whether the public works spending, especially transportation spending, was targeted toward high unemployment areas.

What Glaeser didn’t add is that it’s not clear that additional federal funds for unemployment insurance and food stamps actually reduce unemployment, which was the question Glaeser set out to answer re: public works spending. The straightforward case is that these measures put cash in the hands of cash-poor individuals, and thus the funds were likely to be spent rather saved, thus helping to maintain demand. On the other hand, there is the impact on labor supply, per Casey Mulligan. One can, of course, make a strong case that Congress was right to make additional federal funds available regardless of the impact on the duration of unemployment. 


Sign up for free NRO e-mails today:

Subscribe to National Review