Steam as a General Purpose Technology: A Growth Accounting Perspective

In recent years there has been an upsurge of interest among growth economists in General Purpose Technologies (GPTs). A GPT can be defined as "a technology that initially has much scope for improvement and evntually comes to be widely used, to have many uses, and to have many Hicksian and technological complementarities" (Lipsey et al., 1998a, p. 43). Electricity, steam and information and communications technologies (ICT) are generally regarded as being among the most important examples.An interesting aspect of the occasional arrival of new GPTs that dominate macroeconomic outcomes is that they imply that the growth process may be subject to episodes of sharp acceleration and deceleration. The initial impact of a GPT on overall productivity growth is typically minimal and the realization of its eventual potential may take several decades such that the largest growth effects are quite long-delayed, as with electricity in the early twentieth century (David, 1991). Subsequently, as the scope of the technology is finally exhausted, its impact on growth will fade away. If, at that point, a new GPT is yet to be discovered or only in its infancy, a growth slowdown might be observed. A good example of this is taken by the GPT literature to be the hiatus between steam and electricity in the later nineteenth century (Lipsey et al., 1998b), echoing the famous hypothesis first advanced by Phelps-Brown and Handfield-Jones, 1952) to explain the climacteric in British economic growth.[...]