One of the most prominent theories in macroeconomics is the Ricardian Equivalence proposition that if taxes are non-distortionary then the mix of tax and debt financing of government purchases is irrelevant in the sense that there is no impact on the equilibrium sequence of key real variables. The proposition is easily understood in the context of the "Ramsey model" in which infinitely-lived representative agents solve dynamic optimization problems and have rational expectations about the future course of the economy. The extension by Barro (1974) to an overlapping generations model with finitely-lived agents, who make bequests to their children, showed that Ricardian Equivalence holds more generally than one might think. At the same time it is widely understood that Ricardian Equivalence does not generally hold if agents are not dynamic optimizers, if households are liquidity constrained, if taxes are distortionary, or if government spending is not exogenous to financing.
However, an apparently key assumption that has not been examined in detail is the role of rational expectations (RE). The RE hypothesis is subject to the criticism that it makes very strong assumptions about the information and knowledge agents are assumed to have. A substantial recent literature has emphasized the importance of learning dynamics arising from boundedly rational deviations from RE due to imperfect knowledge of the economy.
This viewpoint raises the question that if expectations are not fully rational because, for example, they are made using adaptive (or statistical) learning rules, will Ricardian Equivalence still hold? Will it at least approximately hold if expectations are approximately rational?
Researchers: George Evans and Kaushik Mitra
Related CDMA Working Papers:
1008 George W. Evans, Seppo Honkapohja, and Kaushik Mitra 'Does Ricardian Equivalence Hold When Expectations are not Rational?'.
|
