Under what conditions does digital technology adoption increase cross location knowledge flows within firms? We investigate this question by studying the impact of adopting basic Internet access on cross-location knowledge flows within the same firm. We construct a large data set of Internet adoption and patent citations among dyadic pairs of firm-locations between 1992–۱۹۹۸٫ We find that when both locations in the pair adopt basic Internet there is an increase in the likelihood of a citation between the citing and (potential) cited location. In contrast, we find no significant effect of Internet adoption at only the citing location. We further study how this effect varies according to the proximity of the research activities between the source and recipient of knowledge and specialization of the research activities within the recipient. We find that the likelihood of a citation increases more after dyadic Internet adoption when the pair is working in similar research areas and when the research areas in the citing location are less specialized. These results, which are robust to a range of robustness analyses, suggest that digital technologies such as Internet connectivity are able to facilitate knowledge flows between locations only when they share a common knowledge base.

Introduction

Research and development operations are often physically dispersed within firms (Leiponen and Helfat, 2011; Miller et al., 2007; Penner-Hahn and Shaver, 2005; Singh, 2008). Such geographic dispersion confers benefits, including the ability to tap into local knowledge bases. However, even within firms, it is often difficult to effectively transfer and use knowledge produced elsewhere within an organization (Allen, 1977; Teece, 1977). The increasing digitization of innovation processes has the potential to increase the flows of knowledge across locations within geographically disparate organizations. However, there are often significant barriers to the transfer of knowledge through digital systems.1 While the sources and consequences of these barriers have been explored in the literature, a lack of systematic empirical evidence has made it difficult to quantify their effects and assess the conditions in which digital technology can help overcome the well-known localization of knowledge (e.g. Jaffe et al., 1993), leaving open a number of important questions. One question of particular interest is whether digital technology helps users to identify and absorb knowledge that they do not already know, or simply reinforces existing areas of expertise. One view has held that IT systems can enable users to access new sources of knowledge outside of their own local cohort or group (e.g., Offsey, 1997), leading to novel recombinations of ideas that generate significant scientific impact (Uzzi et al., 2013) and that are important for economic growth (Weitzman, 1998). Another has held that individuals use IT systems to disproportionately reach out to existing contacts and those who share similar tastes or knowledge bases (Gaspar and Glaeser, 1988; Rosenblat and Mobius, 2004; Van Alstyne and Brynjolfsson, 1996, 2005).