- Income Asymmetries and the Permanent Income Hypothesis.
Abstract: Within the context of the Permanent Income Hypothesis (PIH), the predictions for consumption depend crucially upon the process for income. In this paper, we consider an unobserved components model that allows for both asymmetric transitory movements and correlation between permanent and transitory innovations. Using aggregate U.S. data, we show that this model fits labor income data significantly better than common alternatives. However, we find that consumption is excessively smooth relative to the predictions of our model. To reconcile these predictions with the data, we explore the possibility of imperfect information. A delayed information version of the model fits the data better but consumption is excessively sensitive compared to the predictions of this model. We are able to match the data when we consider an economy in which 60 – 65% of consumers behave according to the PIH with full information and the remaining consumers have delayed information.
Abstract: This paper explores the possibility of weak identification in estimated Taylor rule regressions with interest rate smoothing. We argue that the presence of smoothing renders the information for the estimates of Taylor rule parameters dependent on its true value in such a way that as it approaches unity inference could be spurious. In the literature, quarterly estimates of the smoothing parameter are typically in the order of 0.7 – 0.9. Therefore, we conduct a series of Monte Carlo experiments for empirically relevant sample sizes and values of the smoothing coefficient. Our results show that the actual size of a nominal 5% test is always oversized, hitting rejection rates of up to 20%. Altogether, our results suggest that evidence supporting the Taylor Principle could be spurious.
