Suppose that there were a major fire, and that in order to put out the fire you would need, say, a trillion gallons of water.  Can you imagine a city council that would say, “oh no, we can only afford 734 billion gallons of water, so let’s leave out about a quarter of the neighborhoods.  It’s the right thing to do because we won’t go into debt, and future residents will be better off for having had a quarter of the city burn down.”?

Or, for a better analogy, suppose that your ship is sinking, through a hole that that is 10 feet in diameter.  How about saving on repair costs but inserting a plug that covers only 75 percent of the leak?  Sound like a good plan?  Not so much.

The reason that we need fiscal stiumus is that monetary policy is impotent to provide sufficient stimulus (not generally true, but true now, and essentially no one disagrees with this view).  With an unemployment rate of 7.6 percent, the economy is well below its potential level of output — we are about five percent below potential GDP, and the situation is getting worse by the hour.  The current cost of putting unemployed resources to work in this setting is very low, because the alternative is that those resources will not be used at all. Deficit-financed spending, public and private, can create current income and will reduce unemployment and the risk of future unemployment.  Some of the income generated by the stimulus, and some of the stimulus itself, will go into investment, and hence lead to increases in future income.  The income gains are valuable in themselves, and will offset a good deal of the taxes required to service the debt.  This analysis would be completely different if the economy were somewhere near full employment.  In that case the new spending, both private and public, would substitute for other activity, and the increase in the deficit would reduce investment and growth.  (To go back to the sinking ship analogy, patching a leak where there is no leak is simply a waste of resources.)

Everything that I have said above is oversimplified, of course, but the public discussion of the size and shape of the stimulus package seems to be missing the point.  The point isn’t to have the cheapest stimulus package possible; the point is to align the size and timing of the package with the size of the problem.  The most immediate and effective form of stimulus is to support state governments, because their revenues are falling and they will be forced, by their own constitutions, to reduce spending and lay off workers.  So the immediate stimulus effect of a dollar of support for state spending is a dollar, growing to about two dollars once the effects percolate through the economy.  (Note that what is really going on is the avoidance of a dollar’s spending reduction, growing to two dollars, at the worst imaginable time.)  In this context, Congress gets all sanctimonious about waste in government.  Halleluljah!

Paul Krugman’s recent columns and blogs on this subject have been excellent, by the way.  I commend them to the world of libraries.

And one more thing.  If we happen to make a mistake and overstimulate the economy, monetary policy will be perfectly effective in reining things in.  One way of characterizing the goal of fiscal policy in the current crisis is to restore the economy to a place where monetary policy can work.  The task is urgent.