Cascading innovation: R&D team design and performance implications of mobility

Abstract

Abstract Research Summary Given the high cost of external hiring and uncertainty related to performance benefits, how can organizations foster an environment that maximizes the post‐mobility performance of external hires and their collaborators? In this article, I posit that R&D team design is an important factor that could shape the post‐mobility performances of both groups. Building on the interfirm mobility, innovation, and teams literatures, I argue that technological knowledge diversity within teams and across teams could differently impact innovation performance. Analyzing 63,976 interfirm moves of engineers and scientists, I find that the post‐mobility performances of external hires and teammates are conditioned by team design. A high level of within‐team diversity improves the performances of both groups, while a high level of across‐team diversity hurts their innovation outcomes. Managerial Summary In the war for talent, firms often offer premium wages to source external hires. Yet there have been unclear expectations about the post‐mobility outcomes of these hires and the implications for team performance. To better assess the value of hiring, managers should look beyond the performance of external hires and also consider team member performance. In the context of knowledge production activities, external hires, on average, experience an improvement in innovation performance after they move. Working with an external hire reduces the productivity of immediate collaborators but leads to more breakthrough innovations with greater technological impact. Most importantly, the performances of external hires and their teammates can be further improved by effectively designing R&D teams. Recomposing teams such that a high level of diversity exists within a team but minimal divergence across teams creates an environment that appears to enhance post‐mobility innovation activities.

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 — 38 citations exist upstream.

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