At a Carnegie-Rochester conference I attended over 30 years ago, when I was an assistant professor at the University of Rochester B-school, Arnold Harberger criticized a pile of economists' papers on Latin American development. One of his criticisms still stands out in my memory: the papers, he said, were "priceless." That is, they left out price theory or contradicted it. Much of macroeconomics is that way, especially the crude Keynesian macro that Paul Krugman is pushing.
An article that points that out beautifully is Robert Murphy's "Does 'Depression Economics' Change the Rules?" Krugman's answer is yes. Murphy's answer is no. In making the "no" case, Murphy brings in standard price theory. He points out that Krugman and fellow Keynesian Mark Thoma assume that when the government spends more on building some project, it will use resources that are otherwise idle. That assumption if false. Murphy writes:
Within the broad category of "labor" we find a similar situation, once we actually contemplate doing this project for real. If the city of Houston wants to build a new bridge, is it really the case that every last person even remotely involved with the project, will come from the ranks of the unemployed who are within commuting distance of the Houston bridge site? Surely the project will draw on engineers, construction foremen, and other skilled workers, who were still gainfully employed even amidst the recession, and who therefore will not be able to work on as many private-sector projects as they otherwise would have.
What is particularly ironic in this discussion of idle resources is that it is the pro-stimulus Keynesians who ought to be very fastidious in their recommendations for government spending projects. After all, if the whole point is to draw down resources that have been thrown out of work, then care should be taken to tailor the stimulus package for the resources in question. Is it really the case, for example, that bridges and roads require labor and other inputs in the same proportions as housing construction and finance? Does the construction of a new sewer system require the services of investment bankers and roof layers in such combinations that local government spending can perfectly offset the bursting of the housing bubble?
A little price theory, realizing, for example, that not all workers are perfect substitutes, would have gone a long way here.
Murphy also addresses why the resources are idle in the first place, pointing out the overexpansion in certain sectors of the economy. Although Murphy blames Greenspan, Jeff Hummel and I have said why we think that's unjustified. But you can unbundle Murphy's package. None of Murphy's points depend on whether the idle resources are due to Greenspan's monetary policy. What we know is that three sectors that need to shrink are housing, autos, and financial services. And guess what sectors the government is subsidizing.