Innovation Diffusion and the Evolution of Regional Disparities

An important part of modern theory on regional disparities identifies in the innovative activities the source of competitive advantage and, consequently, of the observed differences in economic development. However, since differently from a standard physical factor of production, knowledge can be used in many different places and many productions at the same time, innovation is the basic but not the only feature of the process: an essential role is also played by imitation, the acquisition of external knowledge through devoted efforts, and diffusion, the non costly acquisition of external knowledge. If, for example, the diffusion of technology were instantaneous and not protected by any form of patenting, the physical place in which innovation takes place would be irrelevant. Since knowledge is instead 'sticky', the innovative region get an advantage, but the extent of this advantage still depends on the easiness of diffusion. This mechanism of knowledge creation and diffusion has been affected by two important changes: the first one is the ‘technological revolution’, as some scholars called the process that, in the past century, changed innovation from being the outcome of initiatives of single entrepreneurs, to the outcome of specifically designed R&D departments of the firms; this has determined an increase in the innovation pace. The second change is the recent fast development of new means of communications, called ICTs, which have made easier, faster and cheaper the transfer of blueprinted knowledge across places. This is determining an increase in the speed of diffusion of knowledge, even if, in order to use it, it is always necessary to have people able to understand and implement it. This article investigates the effect that the interaction between the creation and the diffusion of technology brings on regional disparities. We will show that an increase in the pace of innovation, as the one that took place in the XXth century, can engender regional income disparities; we will also show that if, afterwards, the speed of diffusion also increases enough, these disparities can fade out. We will not, at this stage, guess which effect will prevail. To analyse the problem, we will first build a simple model with technological disparities as the source of income disparities and a set up aimed at representing all knowledge flows. The basic model will then be used in two different ways for the study of innovation and diffusion mechanisms. We will show that the most important variable to determine if income disparities exist is the ratio between the speed of diffusion and the speed of innovation. In particular, when this ratio is low, the most likely prediction is an equilibrium with both technological and income disparities. For intermediate values, technological disparities will exist, but will not be large enough to generate income disparities; for higher values, there will not be technological disparities anymore and, consequently, no income disparities.