Brad DeLong has an impassioned plea for further fiscal stimulus under the headline “Fiscal Policy: The Obama Administration Is Not Making Much Sense These Days”. His basic argument is simple: if the $787 billion package was designed using assumptions which turn out to have been overoptimistic, then surely now that unemployment is heading into double digits a second major stimulus is warranted.
But the problem is that spending trillions of dollars is actually extremely difficult, and it’s even harder if you try to front-load it. Government, by its nature, moves slowly, and I get the impression that the “easy” spending — and then some — was all included in the initial stimulus bill. The “shovel-ready projects” have already been funded, and any extra stimulus might well take years to kick in.
In an efficient market, a credible government promise to invest hundreds of billions of dollars in mass transit and nuclear power and smart electrical grids and so on and so forth would have an immediate stimulative effect: people would start spending now, in anticipation of all those government dollars which are going to arrive in a few years’ time. But we’re in a liquidity crunch, and we’re not in an efficient market, and unfortunately government spending only seems to cause any stimulus as and when the checks are written, if then. (Insofar as they go to companies who are running down their inventories, there’s no stimulus at all.)
I don’t think there’s any doubt that there are diminishing marginal returns to fiscal stimulus plans — which brings me to Paul Krugman’s take on the subject, where he asks how an examination of the marginal costs and benefits of deficit spending could possibly come to the conclusion that stimulus of $800 billion was exactly what was needed.
Let me try to hazard an answer to that. Start with the guiding assumption, as stated by Larry Summers when the stimulus bill was going through Congress, that the risks of spending too much paled in comparison with the risks of spending too little. And because the effects of government spending on GDP and unemployment are hard to predict with any accuracy, there was a strong case that a monster $800 billion stimulus bill was in many ways the prudent course of action.
Since then, however, the economy has done much worse than anybody thought it would. Which is one way of saying that the stimulus has not done as well as people thought it would. This is a useful datapoint — and one way of looking at it is to conclude that the stimulus was so big that the last few hundred billion dollars have had virtually no positive effect at all. And that any extra stimulus would similarly achieve very little.
If there is to be a second round of stimulus, then, I think it should certainly go to areas largely untouched by the first round. I like arts spending; DeLong wants an “aid-to-states-that-maintain-effort package”. But unless and until we can demonstrate that the marginal benefit of extra stimulus spending hasn’t already diminished to something negligible, it does seem fiscally reckless to throw good money after wasted funds.
Update: Commenters rightly say that there’s a limit to how much can be done with arts subsidies. Again, they’re the kind of thing where a little goes a long way, but a lot doesn’t go a lot further.
And Brad DeLong responds, saying that of the $787 billion earmarked in the stimulus bill, only $14.5 billion has been spent so far, and that total will barely exceed $50 billion by October. Which seems to me a good reason why we don’t need to increase the total — we’re having a hard enough time spending the money we’ve already got.