<P>Network Effects and the Role of Influence in Technology Adoption</P>
<P>Catherine Tucker
November 18, 2004</P>
<P>Abstract
Communications networks may not be symmetric. People may use a communications
network technology to talk with only a few other people, and their conversations
may vary in importance. Such asymmetries may be reflected in the structure of network
effects. This research addresses three questions: Are network effects limited to
the people that an individual communicates with? Do network effects vary in size
according to how important communication is? When deciding whether to adopt, do
people anticipate that their installation will affect whether those they wish to talk
to will adopt in the future? To answer these questions, this research uses extensive
data on all potential adopters of a firm’s internal video-messaging system and their
video-messaging patterns. The technology can also be used to watch TV. Exogenous
shocks to the benefits of watching TV are used to identify the causal (network) effect
of changes in the installed base on installation decisions. The first finding is that network
effects are individually localized: Potential installers only react to installation by
people they wish to communicate with. The second finding is that network effects are
heterogenous: Installation by managers and workers in “information broker” positions
has a large impact on the installation decisions of employees who wish to communicate
with them. Installation by ordinary workers has a far smaller impact. The third finding
is that these “influential” installers do not anticipate the positive effect their installation
has on the future installation decisions of those they wish to communicate with.
Consequently, they do not internalize the externality associated with their installation.
This suggests that those managing a firm’s technology adoption should not rely on
potential adopters, even when they know each other well, to internalize network externalities.
Instead, knowledge of the firm’s communication structure should be employed
to ensure that key employees adopt. In general, optimal technology policy should re-
flect the structure of network effects and focus adoption incentives on influential users
of a network product, rather than subsidizing everyone’s adoption equally.</P>
<P>
</P>