Between the 1760s and the 1980s, three industrial revolutions occurred with almost equal intervals. We explain this phenomenon, which is left untouched in the existing literature. Although the standard theory of discrete-time chaos provides a perfect tool to explain business cycles in a single macroeconomic variable, it is of little use fully to explain innovation dynamics; innovation dynamics inherently involves two state variables representing monopolistically supplied new products and competitively circulating existing products separately. This study unveils two-dimensional ergodic chaos and develops a model of innovation dynamics with these variables. It demonstrates that if the first mover advantage from new products lasts for about 8 y, an industrial revolution-like phenomenon will occur about every 100 y.
Since the 1760s, at least three industrial revolutions have occurred. To explain this phenomenon, we introduce two-dimensional (2D) constrained chaos. Using a model of innovation dynamics, we show that an industrial-revolution-like technology burst, driven by investment/saving motives for R&D activities, recurs about every one hundred years if the monopolistic use of a new technology lasts about 8 y.