Simple econometric tests reported in the literature consistently report what appears to be a bias in inflation expectations. These results are commonly interpreted as constituting evidence overturning the hypothesis of rational expectations. In this paper, we investigate the validity of such an interpretation. The main tool utilized in our investigation is a computational dynamic general equilibrium model capable of generating aggregate behavior similar to the data along a number of dimensions. By construction, the model embedded the assumption of rational expectations. Standard regressions run on equilibrium realizations of inflation and inflation expectations nevertheless reveal an apparent bias in inflation expectations. In these simulations, the null hypothesis of rational expectations is incorrectly rejected in a large percentage of cases; a result that casts some doubt on conventional interpretations of the evidence.
