The Laffer curve of the Democrats

From Paul Krugman's website, a commenter asks an incisive question:

Is it just me, or is this argument for government spending starting to sound just a bit like the Democratic version of the Laffer Curve? i.e. more borrowing means more tax revenue, instead of lower tax means more tax revenue !

The tendency to attribute outright magical powers to government spending has gotten slightly out of control.  It's appropriate to ask the same question that should have been asked of Republicans in 1980:  if all this is so marvelous, why don't we just do it indefinitely--slash tax rates to zero, borrow and spend forever?

The answer is that there are declining returns to all of this.  At some point, the Laffer Curve maximizes, and any further cuts cost the government money. Similarly, the trillionth stimulus dollar probably isn't nearly as effective as the first.

I'm becoming extremely concerned about the stimulus, for the following reasons:

1)  Where is the strong evidence that the kind of truly massive stimulus people like Krugman are pushing for will do anything but provide a very temporary respite before the economy slumps back, more indebted and no better off than before?  The chief complaint about the two historical examples we have, the Great Depression and 1990s Japan, is that such stimulus was not sufficiently tried.

2)  What about the permanent income hypothesis?  If we make the stimulus spending temporary, I presume we have the same problem we do with tax cuts--rational consumers will save most of the extra income.  If we make it permanent--that's a different, but bigger, problem than we have now.

3)  We are a nation of net dissavers, which contributed greatly to the bubble.  Can we really prolong this?