Shifting grounds: how industry emergence changes the effectiveness of knowledge creation strategies – the case of the US automotive airbag industry

Abstract

This paper investigates the effect industry life cycle phase shifts have on the effectiveness of firms’ knowledge creation strategies. Building on literature streams on strategic knowledge management and industry life cycles, we develop theoretical arguments for why the best knowledge search strategy should be different before the emergence of an industry compared to afterwards. Testing our hypotheses empirically in the emerging US automotive airbag industry confirms the powerful forces of industry emergence: the best knowledge search strategy is initially one that looks inward into the organisation but outside of the technology area, and later shifts to one that is looking outward from the organisation and the technology. As practical implication we derive that R&D managers should (i) adjust their teams’ knowledge search strategies depending on the industry life cycle phase in which they find themselves, and (ii) especially look for new applications of their firm's existing knowledge in related fields.

Keywords:

• industry life cycle
• knowledge management
• search processes
• R&D
• patents

Acknowledgements

The authors would like to thank Carliss Baldwin, Michael Heeley, J. Peter Murmann, Karl Ulrich, seminar participants at Wharton and MIT, and three anonymous Academy of Management Annual Meeting Reviewers for their helpful comments on earlier versions of this paper. All errors remain our responsibility. The authors acknowledge financial support for this research from the National Science Foundation (SES-0620487).

Notes

Because we will define an industry by using technological knowledge categories, our use of the ‘product technology’ dimension becomes identical to what the industry life cycle theory (Utterback 1994) has labelled ‘industry’ or ‘product’.

We were unable to use a product-technology impact measure for comparison across the pre- and post-sales take-off phases as the dataset is too small in the pre-take-off period.