All Networks are not Equal: Multiple Ties and Alliance Formation in High Tech Industries

Abstract

In this paper we examine the contingent effects of embeddedness in different kinds of networks on alliance formation rates. To form alliances, firms have to be aware of their opportunity set. This awareness can come from various sources; embeddedness in different networks are such sources. Prior alliance networks of firms and possession of technological knowledge have been shown to be important contextual factors (sources of information) which affect alliance formation rates. We posit embeddedness in technical committee networks through joint participation and embeddedness in director interlocks networks are mechanisms that allow firms to exchange information pertinent to alliance formation beyond that allowed by the contextual factors. We note that extant literature has not focused much on the influence of multiple networks on firm strategic outcomes, or on their contingent effects. Our objective in this study is twofold. Firstly, we propose to study the effects of embeddedness/participation in multiple networks on alliance formation, both as inducements and opportunities of technical alliances. Secondly, we seek to gain insight on the contingent value of participation in networks for alliance formation. Thus, our question is: what role does embeddedness in technical committee networks and director interlock networks play for firms when they have lower or higher levels of embeddedness in prior alliance networks? Similarly, what role do these networks play when the firms have lower or higher levels of technological knowledge? We propose that for firms embedded in an alliance network, participation in technical committee networks is not a very important source of information for alliance formation (Thus we are arguing for a redundancy effect of the new channel of information). Similarly for firms with significant technological knowledge, participation in technical committee networks is not a very significant source of information for alliance formation. Further, we propose that director interlocks play an important role for firms even if they possess significant technological knowledge and are embedded in an alliance network. (Thus we are arguing a non-redundancy effect for director interlocks; implying that endorsement and influence from the top-most management level in the organization is not substituted by any other information channel). Results from a ten year panel on the US cellular industry show that the influence of technical committee network is indeed greatest for dyads with less visible patents. The influence for dyads not embedded in an alliance network, however, is not as expected. Technical committee networks act as a complement rather than as a substitute to prior alliances. The effect of director interlocks does not depend on the context factors, as expected. Implications of the results are discussed.