This paper discusses the consumption-wealth relationship. Following the recent influential work of Lettau and Ludvigson [e.g. Lettau and Ludvigson (2001), (2004)], we use data on consumption, assets and labor income and a vector error correction framework. Key findings of their work are that consumption does respond to permanent changes in wealth in the expected manner, but that most changes in wealth are transitory and have no effect on consumption. We investigate the robustness of these results to model uncertainty and argue for the use of Bayesian model averaging. We find that there is model uncertainty with regards to the number of cointegrating vectors, the form of deterministic components, lag length and whether the cointegrating residuals affect consumption and income directly. Whether this uncertainty has important empirical implications depends on the researcher's attitude towards the economic theory used by Lettau and Ludvigson. If we work with their model, our findings are very similar to theirs. However, if we work with a broader set of models and let the data speak, we obtain somewhat different results. In the latter case, 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.