Why Does Technology Sometimes Regress? A Model of Knowledge-Diffusion and Population Density

Abstract

In this paper we develop a model of invention and knowledge-diffusion. We show, in a setting of imperfect knowledge-transfers from one generation to the next, that there is a tension between the tendency for old knowledge to be lost on the one hand, and the tendency for preserved knowledge to be relearned and new ideas to be created on the other. The steady state of technology is ultimately determined by the interaction intensity of the inhabitants. Our model captures facets of the historical record on productivity that are otherwise hard to explain. In particular, our model can account for the puzzling phenomenon of technological regress, which is well documented historically, but has no satisfactory explanation in the literature. Although we view our model primarily as a means of understanding certain historical patterns, we argue that the framework we present is general enough to shed light on current issues as diverse as the harmful effects of ethno-linguistic fractionalization on productivity, the trade-off between quality and quantity in fertility decisions, and the interaction between employment-density and productivity in modern economies.

Keywords: Very long run economic growth, Innovation, Knowledge-diffusion, Human Capital

JEL Classification: O40, O31