Neo-institutionalism: Bounded Rationality

The various models discussed in the preceding chapter rest on a certain definition of rationality, which can be characterized as ‘unbounded’ or ‘hyperrationality’. In these models investment decisions are made according to calculations that span all possible states of the world for all relevant time periods. Rationality is thus all-encompassing, in the sense that there can be no endogenous surprises, though there can be exogenous shocks.1 Barring events outside the explanatory power of the model, the menu of possibilities is given. In the growth and investment models presented in the previous chapter, unbounded rationality means that the technology and preference sets change only in predictable and preordained ways. Therefore possible courses of action, that is, sequences of investments, and possible future states that follow from these investments, are defined and known. As was the case for David Ricardo, in modern neoclassical models investment opportunities are obvious, and it follows as a matter of course that the optimal path will be chosen. By contrast, the approach described in this chapter recognizes the limits to calculation. Instead of taking the choice set as given, it asks how the choice set is defined, what possible courses of action an agent will consider, and how agents’ choices reshape the choice set.