How Signifiicant are Housing and Financial Wealth Effects in the US Context? An Identication-Robust Re-Estimation.

This paper presents an identification-robust method for estimating the sep- arate effects of housing and financial wealth on per-capita consumption, using the Anderson-Rubin procedure. Using this procedure,we provide a re-estimation of the model proposed in Carroll, Otsuka and Slacalek (2011); a model that exploits consumption persistence to concurrently identify both short-run and medium run wealth effects. Using U.S. data for two separate periods; one ter- minating prior to 2007-Q4 and the second containing data for the post-financial crisis period, we find conflicting results. For the period preceding the financial crisis, our results tend to support prior studies that found evidence of a relatively dominant housing wealth effect; however, the respective roles of financial and housing wealth in terms of per-capita consumption growth are reversed with the use of our updated data set. Ultimately, we argue that the method employed in this study yields estimates that based on theory, are statistically superior to the results of previous studies