Working Paper: CEPR ID: DP7792
Authors: George W. Evans; Seppo Honkapohja; Kaushik Mitra
Abstract: This paper shows that the Ricardian Equivalence proposition can continue to hold when expectations are not rational and are instead formed using adaptive learning rules. In temporary equilibrium, with given expectations, Ricardian Equivalence holds under the standard conditions for its validity under rational expectations. Furthermore, Ricardian Equivalence holds for paths of temporary equilibria under learning provided suitable additional conditions on learning dynamics are satisfied. New cases of failure of the Ricardian proposition emerge under learning. Most importantly, agents' expectations must not depend on government's financial variables under deficit financing.
Keywords: Expectations; Ramsey model; Ricardian equivalence; Taxation
JEL Codes: D84; E21; E43; E62
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| agents are dynamic optimizers and their expectations are formed through adaptive learning (D84) | Ricardian equivalence holds (E62) |
| agents' expectations depend on government financial variables under deficit financing (E62) | Ricardian equivalence can fail (H39) |
| the government budget is balanced out of equilibrium (H60) | Ricardian equivalence can fail (H39) |
| agents' expectations do not react to current government financing variables (E19) | Ricardian equivalence holds in temporary equilibrium (D50) |
| certain learning dynamics (C69) | new cases of failure of the Ricardian proposition emerge (F11) |
| Agents' expectations are influenced by the history of debt or taxes (D84) | Ricardian equivalence can fail (H39) |
| The expected paths of consumption and capital remain consistent with Ricardian equivalence under adaptive learning (D84) | Ricardian equivalence holds (E62) |