The Management, Organization, and Geography of Novel Innovation

Thumbnail Image



Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



This dissertation develops new theory and evidence on the antecedents and consequences of innovation for firms, and includes three empirical studies focusing on various facets of the management, organization, and geography of novel innovation.

Chapter 1 examines the role of relationships in mitigating change to firm boundaries for new firms entering the medical device industry, focusing in part on how the timing of novel innovation influences whether firms integrate their sales function. Using a new dataset on more than 1,600 new medical device manufacturers that documents both their full product portfolios and sales governance modes over time, this paper finds evidence that relationships develop and influence sales governance choices only when they cross firm boundaries. Further, launching a novel innovation has a nuanced relationship with integration: early in a firm's life, it increases the likelihood of sales integration, but this relationship diminishes over time. This research offers new insights into the limits of relational governance, and contributes to our understanding of the nuanced impact of novel innovation on firm boundaries.

Chapter 2 examines the “dual role” of local inventive activity in firm innovation. On one hand, a vibrant, local research community provides inputs into internal research and development activities: the seeds of internal invention. On the other, external inventive activity provides inventions which can substitute for internally generated inventions: the fruit. Furthermore, inventive activities also provide fertile ground for imitation. This paper develops a simple model of how geographically proximate inventive activity—or clusters—affect firms' innovation choices. Firm invention capability importantly conditions the value to a firm of local innovation-related inputs. The paper employs a recent survey of product innovation and the “division of innovative labor” among nearly 5,000 US manufacturing firms. Absent a direct measure, invention capability is treated as a latent, unobserved variable, and a latent class multinomial model is used to infer its value. Consistent with the model's predictions, more inventive firms make use of the richer soil whereas less inventive firms pick the fruit. Further, more capable firms make use of higher value external sources in clusters. This research expands our understanding of how location shapes both who innovates and how they innovate, and provides a novel method for identifying latent capability.

Chapter 3 examines how the novelty of a startup's invention conditions its likelihood of venture capital (VC) financing. In it, I argue novelty increases uncertainty about commercial viability, thus requiring startups to search more extensively to find willing VCs to fund them. Two factors lower the cost of search: prior startup experience and a thick VC market. Because these factors make extensive search cheaper, novel startups will disproportionately benefit from experience and cluster location. To test this, I build a hand-collected dataset of 4,700 patenting US medical device startups, and follow them from “birth” (first patent), to VC investment (if any), to eventual success or failure. While novelty has no impact on funding or success on average, firms with novel technologies who also have a lower cost of finding and attracting potential partners are much more successful than those with more incremental technologies. Further, thick markets are less useful for firms pursuing novel technologies if they lack prior startup experience, and experienced founders are not especially advantaged in thin markets. Advancing theories of innovation and entrepreneurship, this study highlights when, where, and for whom novelty pays.





Cunningham, Colleen Mary (2017). The Management, Organization, and Geography of Novel Innovation. Dissertation, Duke University. Retrieved from


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.