The discussion of technological and industrial development in this book provides us with several important observations and lessons. As a way of concluding the book, these will be now discussed. We begin by discussing them in relation to economic theories.
It has been more than three decades since Nelson ( 1959) and Arrow ( 1962) elucidated the peculiarities of technology (or, more broadly, information) as economic goods. Most importantly, they emphasized the presence of non-rivalry in consumption and the difficulty in fully appropriating the returns. Arrow ( 1962: 618) also discussed the 'interdependence of inventive activities, which reinforces the difficulties in achieving an optimal allocation of the results. Information is not only the product of inventive activity, it is also an input--in some sense, the major input apart from the talent of the inventor.' This last observation implies that the interdependence of inventive activity takes place both spatially and intertemporally, because it should be the information accumulated up to now that would be utilized as an input into the current inventive activity.
This intertemporal interdependence is exactly what we found to be the conspicuous nature of science and technology. How much technology a nation can absorb and how much it can generate are dependent in a crucial way upon the accumulated quantity and quality of technological capabilities. Some of these capabilities may be consciously created through innovation and education, while some may be acquired rather by chance, as the adoption of the QWERTY keyboard layout illustrates ( David, 1985). Consequently, the process of technological progress and economic change is path-dependent. The presence of sunk investment reinforces this dependence. For instance, getting used to typing on a QWERTY keyboard is a sunk investment. Switching costs, that is, the costs of switching to other keyboards, are therefore substantial, and the future course of technological development is constrained by this fact.