X-Efficiency Evolution: An Old-New Hypothesis about the Sources of Productivity Growth, from an Agent-Based Simulation

Abstract

While mainstream economics assumes that firms are always on the production frontier, there is an evolutionary interpretation of competition according to which firms have different degrees of what Leibenstein (1964) called "X-efficiency," and competition selects for higher X-efficiency. This concept of competition is implemented in an "agent-based simulation." Firms producing a generic consumption good are endowed with X-efficiencies, which undergo mutation. Even when mutations are mostly negative, selection effects can make productivity rise over time. There is no logical limit to the increase in productivity due to "X-efficiency evolution," which is thus a potential explanation of long-run productivity (or "TFP") growth.