UCLA

The global integration of markets for many goods and services has fueled the rise of the knowledge economy and pushed innovation to the center of models of competitive advantage. At the same time, the knowledge landscape has become more uneven with new technology increasingly generated in relatively few, large cities characterized by deep pools of skilled workers and dense agglomerations of technology-intensive firms.The standard model of innovation has also been reworked. Less and less do we think of new knowledge as produced by individual economic agents engaged in varied forms of learning and in-house research and development. Rather, new technology combinations are seen as emerging from local and non-local collaborations among groups of firms and other economic actors and through more general forms of interaction with buyers and suppliers. The more fractured geography of ideas has granted multi-locational firms clear advantages over their single-plant competitors. First, they are able to locate establishments in different knowledge environments to access multiple, localized forms of buzz and spillovers. Second, embedding plants in different knowledge clusters assists multi-plant firms overcome constraints on effective knowledge flows that are imposed by different forms of proximity. Third, local embedding also gives multi-plant firms the ability to shape the character of knowledge development over space, as well as some control over who is able to exploit specific knowledge assets. Economic geographers have been rather slow in exploring the connections between firm organization and the production of ideas within the knowledge economy. These connections have been more extensively developed within the fields of international business and management. However, scholars in these cognate fields are focused almost exclusively on multi-national enterprises that distribute activities across national borders. Relatively little attention is paid to multi-locational firms that largely operate within a single country. It is these firms upon which most attention is placed in this dissertation. Three broad questions are explored in the main chapters of this dissertation that focus on knowledge production in the domestic operations of Chinese multi-locational firms. The first question is who are the “main agents of change” that drive innovation within Chinese city-regions. Within contemporary economic geography, considerable research examines how local structures condition regional development possibilities. I challenge that logic, exploring whether capabilities are more likely to emerge within the firm and to flow across spatial boundaries than they are to be built within the region flowing across firm boundaries. Analysis targets technological diversification within the establishments of multi-locational firms that are distributed across Chinese cities. Results suggest that the knowledge structure of firms is more important than the knowledge structure of cities in shaping the character of diversification within the establishments of multi-unit firms. The second question explored in the dissertation asks whether multi-plant firms, that distribute their R&D efforts over several cities, produce different kinds of technological knowledge within these R&D units, whether that knowledge is linked to the broader knowledge stocks of the cities where they are located, and whether firms benefit from the geographical distribution of knowledge production. To explore these issues, a linked firm-patent dataset is constructed covering the period 2001-15. Those data reveal significant differences in the character of knowledge production at different R&D sites of multi-unit firms. Furthermore, those differences can be linked to the underlying knowledge bases of the cities where the R&D units are placed. This confirms arguments regarding the knowledge sourcing behavior of multi-locational firms. Finally, I show that the average complexity value of the patents produced by multi-unit firms increases as they distribute their knowledge production across a larger number of cities. The third research question in the dissertation asks how knowledge recombination within Chinese cities scales with city-size and the number of technology components found at the city level. That question is then extended to examine how new technology combinations diffuse across the Chinese urban system. Once more, patent data form the basis for the empirical analysis of the questions raised. Individual technologies are defined by the sub-classes identified in the International Patent Classification. New technology combinations are captured by novel pairings of these technology classes that are found on the same patents. The location of these pairings is traced at the city-level and over time. Analysis of technology recombinations shows that they scale super-linearly with city-size and the number of technologies available at the city-level. This means that new knowledge combinations are increasingly likely to be found in larger cities that contain more diverse sets of knowledge. Subsequent analysis shows that new technology recombinations diffuse across the Chinese urban system conditioned by the geographical proximity of cities, by the technological proximity of different urban areas and as the social proximity between cities increases.