How Much Can Firms Know

Two key stylised facts have been established about the extinction patterns of firms. First, the probability of extinction is highest at the start of the firm’s existence, but soon becomes more or less invariant to the age of the firm. Second, a very recent finding, that the relationship between the size and frequency of firm extinctions is closely approximated by a power law. Economic theory has a great deal to say about many aspects of firm behaviour, but offers relatively little on the deaths of firms. An agent based model of firm evolution and extinction has been developed which has properties which conform closely to the stylised facts. However, in this model, firms are unable to acquire knowledge about either the true impact of other firms’ strategies on its own fitness, or the true impact of changes to its own strategies on its fitness. We examine the effects of allowing firms different amounts of knowledge about the effects of strategy in the context of the agent-based evolutionary model. We investigate the emergent properties of the system in terms of the relationship between the size and frequency of firm extinction, and the survival patterns with respect to the age of agents. In other words, we compare the properties of the model with the two key stylised facts. There are very considerable returns in the model to acquiring knowledge. There is a sharp increase in the mean agent age at extinction for agents with even a small amount of knowledge compared to those without. Indeed, we find that as both the amount of knowledge available to firms increases and as the number of firms capable of acquiring such knowledge rises, the lifespan of agents begins to approach the limiting, full information paradigm of neo-classical theory in which agents live for ever. However, even with relatively low levels of knowledge and numbers of agents capable of acquiring it, the model ceases to have properties which are compatible with the two key stylised facts on firm extinctions. The clear implication is that firms have very limited capacities to acquire knowledge about the true impact of their strategies.