Introduction

It is frequently argued by economic historians that Britain's technological decline relative to countries which industrialised later can be dated to the Victorian period (Habakkuk 1962, Landes 1969). One factor which has been suggested to explain this relative decline is the disadvantages which were due to British industry's early start, making a switch to newer methods of production more costly (a theme discussed by Jervis 1947, Frankel 1955, Saville 1961, Kindleberger 1961, Habakkuk 1962, Ames and Rosenberg 1963, Aldcroft and Richardson 1969, amongst others). Generally underlying the early start thesis is the belief that production methods in different parts of an industry or an economy are interrelated, such that it is costly to change one component of a production process or the methods prevailing in an individual sector without complementary changes elsewhere.

This idea can be allied to an argument which has become popular more recently in the literature on the economics of technological change. The proposition has been advanced that technological innovation proceeds as a cumulative process, and that as a consequence it tends to iock in' to a particular course, once that course has become established. This proposition has been seen as relevant in a variety of contexts. One suggestion is that where there is competition between rival technologies for adoption it is possible that adopting firms become ‘locked in’ to a potentially less efficient technology following the choice of the earliest adopters (Arthur 1989). The decisions of the earliest adopters affect the decisions of the firms that come later. One reason is the role of technological interrelatedness, whereby a supporting infrastructure of complementary technologies become better developed, lowering the costs and increasing the benefits of adopting a technology which is already more widely diffused.