Reinterpreting a Recent Temporally Aggregated Consumption-Cap Model

Abstract

Recent literature has found that the estimated values of the coefficient of relative risk aversion in consumption-based asset pricing models are much higher than indicated by the theory. Moreover, these values still remain implausibly high, although significantly reduced, when the phenomenon of time-averaging is taken into account when estimating the model with quarterly or annual data. As a possible solution to the puzzle, this paper suggests that the more realistic IMA(l,l) representation be assumed as the generating mechanism of consumption. In this case, a necessary and sufficient condition for reducing the estimated value of relative risk aversion is that the coefficient of the moving average component be negative. Some recent empirical and theoretical work in a related area seems to support this suggestion.