Recruiting for ideas: How firms exploit the prior inventions of new hires

摘要

When firms recruit inventors, they acquire not only the use of their skills but also enhanced access to their stock of ideas. But do hiring firms actually increase their use of new recruits' prior inventions? Our estimates suggest they do, quite significantly in fact, by approximately 219% on average. However, this does not necessarily reflect widespread "learning by hiring." In fact, we estimate that a recruit's exploitation of her own prior ideas accounts for almost half of the above effect, with much of the diffusion to others being limited to the recruit's immediate collaborative network. Furthermore, although one might expect the recruit's role to diminish rapidly as her tacit knowledge diffuses across her new firm, our estimates indicate that her importance is surprisingly persistent over time. We base these findings on an empirical strategy that exploits the variation over time in hiring firms' citations to the recruits' premove patents. Specifically, we employ a difference-indifferences approach to compare premove versus postmove citation rates for the recruits' prior patents and corresponding matched-pair control patents. Our methodology has three benefits compared to previous studies that also examine the link between labor mobility and knowledge flow: (1) it does not suffer from the upward bias inherent in the conventional cross-sectional comparison, (2) it generates results that are robust to a more stringently matched control sample, and (3) it enables a temporal examination of knowledge flow patterns.