Pervasive Technical Change Driven by “General Purpose Technologies”

A more comprehensive discussion regarding the effects of innovations for the economy is provided in the literature on general purpose technologies (GPTs), which spur sustained growth via their pervasive use in the economy.

Introduced by Bresnahan and Trajtenberg (1995), the initial focus was on the prob­lems of coordinating innovation activities between up- and downstream sectors in a partial equilibrium model. The lack of technological information flowing between the GPT-providing sectors and the user sectors prevent externalities from coming to full play. Not last in an attempt to explain the productivity slowdown in the US in the 1980s, models were developed around the notion of GPTs emphasizing the impact of major technological change on economic structure and long-term growth. The general content of this discussion was not new, but had close links to previous concepts, such as techno- economic paradigms or macro inventions seeking to explain radical technological change in an evolutionary framework (see Lipsey et al. 2005 for a review of GPT models and related theories). Nevertheless, the notion of GPTs is by no means re-inventing the wheel, not least because of its emphasis on the generality of purpose of a technological breakthrough.

At the theoretical level, the approach by Helpman and Trajtenberg (1998), embed­ded in a model of expanding product variety (Romer 1990; Grossman and Helpman 1991), and the Schumpeterian framework of Aghion and Howitt (1998), became the workhorse models in this strand of research: its broad applicability ensures a new GPT to be demanded in a wide range of sectors, but the crudity in design hampers its adop­tion immediately upon arrival (characteristics 1 and 2). Thus, firms need to engage in the development of complementary components required for the efficient utilization of the technology (characteristic 3).

The shift of resources from the manufacturing sector towards R&D therefore may cause a slump in output as an “integral feature” (Helpman and Trajtenberg 1998: 71) of this type of technological change. Sooner or later, productivity grows due to the increas­ing variety of complementary components available for the new GPT. Other approaches draw on quality-ladder models to explain the cyclical productivity pattern.

The GPT models extend those of Lucas and Romer by explicitly considering the inter­action between technological breakthroughs and complementary (sub-)technologies. However, in all these theories the GPT enters the economy in the form of a productivity parameter, the technology itself and innovations in GPTs being left out of the picture. Furthermore, the focus on the development of new technical solutions to the efficient utilization of the GPT implies a rather simple view of the complex process of adoption of a radically different technology across heterogeneous agents. Yet it is their inherent potential to diffuse over the whole economic system that assigns to GPTs a crucial role in spurring long-term growth. Addressing some of these criticisms, a comprehensive, evolutionary framework for dealing with the variety and coexistence of GPTs was pro­posed by Carlaw and Lipsey (2011). Focusing more on a meso-macro level, one popular framework - building partly on evolutionary economics - is the “systems of innovation” approach.