Growth Cycles in a Discrete, Nonlinear Model

This paper develops a discrete, nonlinear growth cycle model for a macroeconomy. The nonlinearities, which correspond to empirical relationships between profitability and capacity utilization in the postwar U.S. economy, can produce stable, periodic and chaotic behavior. These behaviors are established analytically, and further investigated through simulation. Data from the simulations are used to show that chaotic attractors can produce time series which are useful representations of business cycles.