Using UK data the REPIH is rejected due to the predictive content of consumer confidence, and not labour income or any other macroeconomic variable. We provide evidence suggesting that this cannot be explained by liquidity constraints and account for this finding in terms of precautionary saving. Extending the CCAPM model to allow a time varying conditional variance we find a high level of confidence is associated with greater optimism about the level of consumption but also a higher forecast variance. Allowing for time aggregation, the overidentifying restrictions implied by this model are accepted. We estimate a small but statistically significant intertemporal elasticity of substitution.
No takes yet. Share an insight, caveat, or question.
Acemoğlu et al. (1994) studied this question.