摘要：
The purpose of this paper is to revisit the evidence on the excess smoothness of consumption within the permanent
income model, by using recently available monthly data. Two formulations of the univariate process of personal disposable
income are adopted: in the levels and in the log-levels. More than one sample is studied. Three different impacts
are defined and measured. In theory, the three of them should be equal. The conclusion that is strongly supported is that
these three impacts are significantly different from each other, implying that excess smoothness is still a feature of the
data. However a weak version of the permanent income hypothesis is endorsed which is that consumption changes by the
annuity value of revised expectations of future income. In other terms, permanent income innovations have a significant,
although relatively small, effect on consumption.
