Rational expectations in a changing world

Expectations can either involve a consideration of what will happen in the future or be unrelated to time. Both play an important role in economics, but the temporal aspect will be the focus here. When we are concerned about how an economy will evolve over time, we have to make assumptions about how the individuals that make up the economy anticipate the future. In a world with uncertainty, it is not possible to model the evolution of the economy without making assumptions about the expectations that individuals hold. Economists assume that, based on their expectations, people make the ?best? choice available to them. The ?rational expectations? hypothesis assumes that people somehow all know correctly the process that governs the evolution of the economy. This paper argues that such an approach is not only unrealistic and inconsistent with the empirical evidence but incompatible with the non-ergodic evolution of our complex socio-economic system. Given the impossibility to construct a full-blown theory with ?rational? or optimal behaviour we may need to base our analysis on the sort of heuristics that people and even animals use to plan their future actions. It is the interaction between agents using these heuristics that generate aggregate behaviour not the sophisticated calculations of those agents. JEL codes: B41, B50, D84