Will Tax Increases Kill The Recovery?

I wanted to write about Martin Feldstein's Wall Street Journal op-ed in a timely fashion -- which is to say I wanted to write about it yesterday, when it came out. Largely for reasons of procrastination, that didn't happen. Still, the argument he makes ("Tax Increases Could Kill the Recovery") is one that comes up a lot and I want to say one brief -- if slightly wonkish and tedious -- thing about it.

When I read Feldstein's headline I thought, "But none of the tax increases take effect until 2011!" And while Feldstein anticipates that argument, I'm not sure his response makes a lot of sense:

Even if the proposed tax increases are not scheduled to take effect until 2011, households will recognize the permanent reduction in their future incomes and will reduce current spending accordingly.

This is an argument based on Ricardian Equivalence -- the theory that it doesn't matter whether the government uses debt or taxation to finance its spending, since, if the government uses debt, the perfectly rational robot-people will lower their present spending in anticipation of higher future taxes. And while I'm not really in a postition to criticize Feldstein on Ricardian Equivalence -- he wrote one of the most important papers on the subject -- I do find it odd that he thinks it's made the leap from fun-sounding theory to ineluctable divine truth. That households "will recognize" the budget constraint and "will reduce" their present spending accordingly suggests a mechanism as predictable as night following day. Does anyone actually believe that?

I suspect that Feldstein cannot believe this strong version of Ricardian Equivalence -- in which the gains of present government spending are always and everywhere vitiated by the decline in present household spending -- because Feldstein supported a deficit-financed fiscal stimulus. And indeed, when I went back and checked Feldstein's op-ed on the subject of fisal stimulus I found that he has this to say about Ricardian Equivalence:

Under normal circumstances, I would oppose this rise in the budget deficit and the higher level of government spending. When an economy is closer to full employment, government borrowing to finance budget deficits can crowd out private investment that would raise productivity and the standard of living. Budget deficits automatically increase government debt, requiring higher future taxes to pay the interest on that debt. The resulting higher tax rates distort economic incentives and thus weaken future economic performance.

[...] Nevertheless, I support the use of fiscal stimulus in the US, because the current recession is much deeper than and different from previous downturns. Even with successful countercyclical policy, this recession is likely to last longer and be more damaging than any since the depression of the 1930's.

In other words, there are some situations in which a deficit-financed increase in government demand does not lead to an immediate and one-for-one decline in private demand.

I prefer the Marty Feldstein of three months ago. Which one does Feldstein prefer?