Re-examining the consumption-wealth relationship: The role of model uncertainty

This paper discusses the consumption-wealth relationship. We use data on consumption, assets, and labor income and a vector error correction framework. This framework defines a set of models that differ in the number of co-integrating vectors, the form of deterministic components and tag length. Further models can be defined through parametric restrictions and, in particular, interest centers on a weak exogeneity restriction that says that the co-integrating residuals do not affect consumption and income directly. Key results in previous work relate to the roles of permanent and transitory shocks in driving wealth and how consumption responds to these shocks. We investigate the robustness of these results to model uncertainty and argue for the use of Bayesian model averaging. We find that there is a large degree of model uncertainty. Whether this uncertainty has important empirical implications depends on the researcher's attitude toward the theory used to motivate a co-integrating relationship between consumption, assets and income. If we work with models consistent with this theory and impose the weak exogeneity restriction, we find precisely estimated results that show that permanent shocks have only a small role in driving assets and that the predominant transitory shocks have little effect on consumption. These findings are consistent with the previous literature. However, if we work with a broader set of models and let the data speak, we find that the exact magnitude of the role of permanent shocks is hard to estimate precisely. Thus, although some support exists for the view that their role is small, we cannot rule out the possibility that they have a substantive role to play