External knowledge sourcing, knowledge spillovers, and internal collaboration: The effects of intrafirm linkages on firm‐university co‐authorship linkages

Abstract

Abstract Research Summary Existing literature shows the benefits of firm‐university collaborations. Yet, it has overlooked factors that mitigate concerns with outgoing knowledge spillovers associated with this form of external knowledge sourcing. This study argues that the deterrent effects produced by intrafirm collaborations make a firm more likely to form co‐authorship linkages with a university when its corresponding R&D unit exhibits a higher degree of intrafirm collaboration linkages, especially when that university maintains more linkages with rivals. Analysis at the firm unit‐university‐year level using data on co‐authorship linkages between 157 pharmaceutical firms and the top 400 global universities during a 25‐year period supports these propositions. This study advances research on knowledge sourcing by showing the role of internal collaborations in enabling external knowledge sourcing. Managerial Summary A firm whose employees co‐author scientific articles with university‐affiliated researchers gains access to external knowledge it can build on to create innovations. But, these collaborations with academia can also produce knowledge spillovers benefiting rivals, thereby undermining a firm's efforts to protect its knowledge. Using data from the pharmaceutical industry, this study shows that, because collaboration between inventors in a firm's R&D unit with inventors in other units helps protect that firm's knowledge, a firm creates more co‐authorship linkages with a university when its respective R&D unit exhibits higher levels of internal collaboration with other units. This effect underscores the role of internal collaboration in helping a firm boost innovativeness via external collaboration with universities while mitigating concerns with the erosion of knowledge protection.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 40 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.