In 1987 a book by Chris Freeman with the title “Technology policy and economic performance. Lessons from Japan” was published. This book turned out to be the birth certificate of the systems of innovation perspective which came to enjoy a tremendous popularity the next three decades. The topic of this thesis is within the field of systems of innovation.  This approach is extensively used today and the systems of innovation is an extensive field of studies; there is a growing interest and a well-established group of researchers deal with the concept and set the research agenda. Despite the great interest and the extensive work so far, the concept is challenged by noteworthy conceptual and methodological ambiguities and limitations.

The aim of this thesis is to advance a theoretical framework of the systems of innovation perspective by adopting an actor-based perspective. From a theoretical perspective, the dissertation pinpoints the system of innovation at the interplay between two rationales: a territorial rationale and a functional rationale. The intention of this theoretical framework is to shed light upon the variety of actors operating within a system of innovation. Based on different logics or rationalities, actors understand and behave differently which has an impact on the behaviour and performance of the system. The assumption is that the different rationalities influence the innovation process, and how activities are organized and carried out. In the empirical backdrop of the thesis, the Swedish system of innovation is analysed in terms of how innovation policy discourse and practice have developed over time. The territorial based system of innovation is analysed through three instances: VINNOVA, the Swedish National Innovation Strategy and the Innovation Council, all considered to be relevant in understanding the embeddedness of ideas on innovation within Swedish politics and practice. The functional based system of innovation is addressed through the role of the large firm Ericsson in the systems of innovation. Ericsson has been chosen as an actor in a system of innovation, and therefore attention is paid to the interplay with the territorial actors, such as the state and universities. Several conflicts of interests characterize the relation between Ericsson on the one hand and the university and government on the other. The interplay between the two types of systems of innovation is further concretized in the analysis of the Mobile Heights case, an innovation cluster programme. Three main analytical conclusions have been emerged from the empirical research, discussed in terms of policy makers, policy implementers and practitioners pinpointing to a fairly loose system where different interests, networks and practices can only be partially and temporarily aligned.