Association for Public Policy Analysis & Management

Answers to the question of why some regions succeed in nurturing high-technology industries and entrepreneurship, while other regions, that are seemingly rich in factors that should produce such growth, fail remain elusive. Our current theoretical understanding of the processes of economic development is therefore incomplete. This creates a policy problem, because governments devote considerable resources to try to create vibrant local economies. This paper argues one reason behind divergence in regional entrepreneurial development is due to the degree of coherence in the regional system. It advances the topic of knowledge spillovers between organizations as a mechanism for coherent regional development, arguing that within regions, different types of research organizations, such as universities, firms, and government labs, develop technological expertise that may be more or less similar and evolve at different speeds. Furthermore, spillovers from one organization to another and from one region to another will be characterized by temporal lags and leads with, for example, universities leading and another organization type following. While knowledge spillovers are studied extensively since Jaffe et al. (1993), examination of temporality and across organization types has been mostly overlooked.

The research uses econometric analysis to test the regional knowledge spillover hypothesis by examining the spillovers between different organizations in and across United States (US) regions over time. Specifically, novel techniques for identifying emerging technologies through patents using US Patent and Trademark Office (USPTO) data are tested, evaluated, and extended. Next, the spillover of knowledge codified in these patents is followed prospectively using patent forward citations, while prior knowledge leading to the emerging technology is traced retrospectively using patent and non-patent references. The research distinguishes patent assignee organization types (universities, government, global/nonglobal firms, other research institutes) to examine spillovers across geographic space, organization type, and temporally. For econometric analysis we construct a data set matching the USPTO data to Web of Science articles, COMPUSTAT firm data, and federal research grants. With this data the paper is able to:

(1) develop a typology of intra- and inter-regional knowledge spillover;

(2) answer the question of whether university patenting leads to industry uptake of a technology, or rather, if the entry of new technologies from industry or other types of research organizations, such as government labs, creates opportunity to which universities respond; and,

(3) importantly, better understand the mechanisms of spillover timing across organizations and regions.

Understanding these phenomena would be a contribution to the literature, which for several decades has attempted to establish the presence of spillovers with limited means to measure them. Findings will have policy implications as policymakers interested in supporting economic growth by strengthening technological development will have evidence of the pathway and timing of spillover, leading to better information for targeted policy interventions. Findings will also shed light on the interrelations of universities, industry, and other research organizations in supporting economic growth, providing a better understanding of the innovation systems policymakers often aim to strengthen.