Accounting for change in national systems of innovation: A friendly critique based on the U.S. case

This paper advances a friendly critique of the national systems of innovation approach and offers some suggestions for its future development. I argue that the approach has difficulty accounting for bounded change in national systems. I review three recent changes in the U.S. innovation system - the Internet boom and bust of the late 1990s and early 2000s, the response to the terrorist attacks of September 11, 2001, and the acceleration of productivity growth since the mid-1990s - in order to assess the strengths and weaknesses of the framework in this respect. Future research might be enriched, at least in the case of large national innovation systems, by absorbing concepts developed in other strands of institutionalist literature, such as "intercurrence" and "embeddedness".