When I think about the notion of rational expectations, I am reminded of a character in a Kurt Vonnegut, Jr., novel who, after becoming “unstuck in time,” is condemned to live and relive his life in random order until his death. In so doing, he is able to know the past and the future, though not perfectly. His omniscience is incomplete because some key factors are not revealed to this hapless character, whose futile efforts are unable to avert a horrible destiny. Ironically, his actions to escape his fate are the very trigger needed to bring it about. Thus, the long-run future is completely determined, even though unanticipated surprises may occur at any instant. With a rational expectations framework, the agents' forecasts of prices and other essential economic variables are correct in the long run, though there are unanticipated short-run surprises along the way.

Rational expectations has come to mean diverse things to different economists. To its critics, the rational expectations hypothesis casts economic agents as being perfectly informed of the likelihood of future events and, thus, capable of maximizing their well-being in the presence of random events – known as shocks. Agents are rational in the sense that they know the underlying model of the economy, which is no easy feat. According to Kenneth Arrow (1978, p. 160), “in the rational expectations hypothesis, economic agents are required to be superior statisticians, capable of analyzing the future general equilibria of the economy.”