My recent post on tax cuts and budget deficits concluded that evidence does not support the contention that wealth effects result from tax cuts, or that deficits have detrimental effects on interest rates as a result of wealth effects. However, when income taxes are considered, the economy will respond to differences in the timing of taxes.
Because taxes on income and capital affect the after tax marginal product of labor, and after tax return on investment, taxes vs. deficits during different periods may affect the allocation of work, production, and investment over time. A consequence of this may be that if the government cuts taxes today to stimulate the economy then in the next period when the deficit is settled, or taxes are raised, there may be a reduction in output. These opposing reactions still approximate the Ricardian Equivalence result.
It has been proposed that Ricardian Equivalence may fail if the permanent income hypothesis fails to hold across time. ( if the permanent income hypothesis holds, then a temporary increase in income- from a tax cut- would have a minimal impact on spending) If households believe that their future tax liabilities will be high in the future ( to settle a deficit) only if their incomes are high, then they may have less incentive to save a tax cut. In this case a wealth effect is created and they increase consumption. These results would then lead to the problems so often associated with tax cuts and deficits.
David Romer. Advanced Macroeconomics, 2nd Edition. McGraw Hill. 2001
Barsky, Mankiw, and Zeldes. 1986. ‘Ricardian Consumers with Keynesian Propensities.’ American Economic Review 76 (Sept): 676-691