Lauren Cohen, Joshua Coval, and Christopher Malloy have a fascinating paper (PDF) investigating the question of what happens to a local economy when political events elsewhere in the nation put one of its politicians in charge of a congressional committee. The typical politician's eye-view of things is that this should boost the local economy. The new chairman is able to divert money from elsewhere in the country to his district, which should boost economic activity. An alternative point of view is that you get a kind of extreme form of crowding out, in which the availability at the margin of public funds reduces local business investment. Cohen, Coval, and Mallow find somewhat surprisingly that the crowding out effect is dominant which should be very interesting to politicians who may want to rethink their ideas about optimal re-election strategy.
But the authors appear to believe their findings have important implications for the debate over the use of discretionary fiscal policy during recessions, which I think is totally mistaken. They write:
We then investigate the private sector consequences of these seniority-linked government spending shocks by studying the behavior of the public corporations headquartered in the congressman’s state. From the Keynesian perspective, the predicted response to spending shocks is unambiguous. Because the spending is “free” from the recipient state’s perspective—it will largely be paid for by residents of other states as opposed to increases in the state’s taxes or interest rate—the funds are expected to help firms increase investment, employment, and output. The neoclassical point of view is quite different. We present a simple model in which individuals face a labor-leisure trade-off in the presence of government spending. Increased resources from the government that are not expected to be funded by taxes or borrowing induce individuals to increase their consumption and leisure. The resulting decline in the marginal productivity of capital compels companies to scale back investment and output.
Our empirical results support the predictions of the neoclassical model. Focusing on the investment (capital expenditure), employment, R&D, and payout decisions of these firms, we find strong and widespread evidence of corporate retrenchment in response to government spending shocks. In the year that follows a congressman’s ascendency, the average firm in his state cuts back capital expenditures by roughly 15%. These firms also significantly reduce R&D expenditures and increase payouts to their investors. The magnitude of this private sector response is nontrivial: in the median state (which receives roughly $452 million per year in increased earmarks, federal transfers, and government contracts as a result of a seniority shock), capex and R&D reductions total $48 million and $44 million per year, respectively, while payout increases total $27 million per year. These changes in firm behavior persist throughout the chairmanship and begin to reverse after the congressman relinquishes the chairmanship. We also find some evidence that firms scale back their employment, and experience a decline in sales growth in response to the government spending shock.
This to me seems like an example of taking a very interesting empirical result that doesn't happen to be super-salient for high-profile partisan political debates and then trying to inject it into a higher-salience more partisan controversy. But the fiscal stimulus debate is about what to do if a 0 percent short-term nominal interest rate fails to produce full employment. Can increasing the bugdet deficit boost a national economy facing that situation? Cohen, Coval, and Mallow doubly fail to address the issue. First, the time periods they're investigating don't match the description of a period in which stimulus advocates say higher deficits would boost real output. Second, they're not studying the phenomenon of higher deficits.
As a theoretical matter, what the paper seems to do is illustrate the central importance of marginal return on capital. The way I would put that in the context of a Keynesian story about recessions is this. Sometimes "shocks" happen that reduce the marginal return on capital and depress investment. But the quantity of investment is a function of both the marginal return on capital and the interest rate. So the standard policy response is to reduce interest rates sufficiently to generate a quantity of investment consistent with full employment. However, the interest rates in question may be below zero. In that case you might try two different things. One is that the monetary authority (with the tacit cooperation of other policymakers) could attempt to elevate inflation expectations, thus lowering the real interest rate. The second is that fiscal policymakers (with the tacic cooperation of the monetary authority) can increase budget deficits in an effort to raise the marginal productivity of capital. How does that work? Well imagine there are a bunch of vacant storefronts, empty apartments, half-full trucks, and idle airplanes. The marginal return on capital is low. Now we have a big old tax cut. Consumers have more money in their pockets, and use the funds to occupy some of the apartments and airplanes and buy goods that fill the trucks and the storefronts. Now the marginal return on capital is increasing. Alternatively, the government could buy a bunch of stuff that needs to be shipped around on trucks and airplanes. Again, the marginal return on capital is increasing. Technically speaking you get more "bang for the buck" with government purchases, but the tradeoff is that it's more difficult to organize government purchases than lower taxes or increased transfer payments.
At any rate, the fiscal policy in a recession debate will no doubt rage on. But this paper is much more directly relevant to the debate that goes on inside the office of every member of congress. How much time should we spend porking for the district and how much on advancing a public interest policy agenda. The implication here is that the tradeoff is not what you think and may not even exist! If the authors can persuade people of that, it would have dramatic implication for American governance in situations that arise much more frequently than a recession at the zero bound.