The damage rested in the people's faith that the government could carry out economic policy. Specifically in the context of the government being able to do this without manipulation. During this time a remarkable suggestion of the replacement of adaptive expectations by rational expectations was the "Lucas Critique," This critique illustrated that expectation parameters, and endogenous variable dynamics, depend on policy parameters. (Muth p. 315) The presentation and discussion of this critique is taken into consideration for purposes of this discussion from the vantage point of the issue. The issue for this discussion is the issue of bounded rationality, where for transparency it is modeled to bounded
rationality by means of simple adaptive expectations. The examination of this critique will illustrate that for a range of processes, monetary policy remains subject to the Lucas critique. (Cooley p 64) Nonetheless, "there are also regimes in which the expectation
The adaptive expectations hypothesis was introduced by Cagan (1956) and Friedman (1957) as a plausible and empirically meaningful approach to modeling expectations of future variables in a world of uncertainty. "Their apparent empirical success led to widespread utilization of the adaptive expectations hypothesis before it was ultimately swept away by the rational expectation revolution, initiated by Muth (1961) and advanced by Lucas (1976) and Sargent and Wallace (1975). Rational expectations has the great advantage of providing optimal expectations; under the standard of optimality, adaptive expectations suffers by comparison and should be rejected. (Cooley 1973) One of the most salient implications of rational expectations is the critique of traditional policy making presented in Lucas (1976).
The traditional theory of economic policy is characterized as treating the time series process followed by the economy as fixed and invariant with respect to exogenous changes in policy. Under rational expectations, however, the forecast or expectation (Muth 1973) rule will be affected by policy changes and, if the economy is in
turn affected by expectations, these will alter the time series process followed by the economy. Lucas provided examples of this phenomenon based on prominent macroeconomic models.
Our objective here is to reconsider the Lucas critique in the context of
adaptive expectations. The starting point of our argument is Muth (1960). In that paper Muth showed that adaptive expectations, with an appropriate adaptation parameter, are fully rational if the variable being forecasted follows an exogenous IMA(1,1) stochastic process, i.e. if the first difference of the variable is a first-order moving average process. Rational expectations, however, assumes that the true process generating the data is known, an assumption that many feel to be implausibly strong. Recently Evans and Honkapohja (1993, 2001) and Sargent (1999) have argued that adaptive expectations
may be a reasonable, if not fully rational, forecast method when
the true process is unknown.
We consider a simple macroeconomic model, inspired by Lucas (1973) and
Fischer (1977), in which aggregate output is affected by unanticipated price
level changes. Let aggregate supply be specified as follows:
qt = (pt pe
t ), (1)
where qt and pt