Opportunities, Incentives and the Collective Patterns of Technological Change

In a few pages, in the introductory article for this debate, Vernon Ruttan succeeds in the unlikely task of flagging many issues at the core of the economic interpretation of technological change (and by doing so he implicitly addresses also a few sensitive spots for the economic discipline at large). Here I will not attempt to provide any thorough view of what I consider the achievements and limitations of the current state-of-the-art (see, however, Freeman (I 994) for a critical survey which I largely share). Rather, I shall put forward a series of remarks which, hopefully, could be seen as part of a much wider, and still largely unexplored, research programme. As a premise, just note that some form of attention to technology-related phenomena has spread over the last 20-30 years from small enclaves of heterodox economists and economic historians to become an important concern even for quite mainstream theories of e.g. the firm, industrial organisation, macroeconomics, trade, growth, etc. From 'patent races' all the way to 'New Trade' and 'New Growth' theories and even Real Business Cycle macro models, technical change has drawn a consideration which is less by default than it was, say, in Robert Solow's original growth models orJoe Bain's barriers-to-entry views on industrial structures. And, in parallel, empirical studies at the levels of specific technologies, firms, sectors and countries have boomed (for overviews cf. Dosi (I988), Nelson (I993), Freeman (I982, I 994), Cohen (I 995) and Granstrand (I 994), among others). So, at least in this sense, those 'decreasing returns' to the economics of technical change suggested by Ruttan do not seem to have set in. However, Ruttan's paper raises also substantive questions about the relative success of three interpretative perspectives which he calls 'induced', 'evolutionary' and 'path-dependent'. In the following I shall argue that an evolutionary perspective, broadly defined, can easily accommodate both inducement effects and path-dependent patterns of technical change. Let me start from what I consider the major building blocks of an