| We investigate the relationship between alliances and learning by examining patterns of alliance ties and patent citations in the global pharmaceutical industry. The empirical studies anchor on the evolutionary perspective, suggesting ‘local search’ for and ‘experiential learning’ from a technologically related and socially connected partner.; Our study found strong evidence that learning takes place as a result of forming alliances. With respect to alliance formation decision, firms are more inclined to ally with partners, (1) which are similar in technological foundations in a network of patent citations, (2) which are already connected in a network of prior ties, and (3) which are in the same region. There is a tension, however, concerning the extent to which firms should be similar or different in order to learn most effectively. The results suggest that forming alliance between technologically similar firms doesn't add any learning because most learning has already occurred by the time the alliance comes into effect. It highlights the importance of selecting partners with related but still distinctive capabilities. The results also suggest that the ‘international’ or cross-border variable produces a significant increment of learning possibly because firms from different nationality have the most to learn from each other's idiosyncratic technological environments.; There is a discrepancy, as our results indicate, between partner selection and learning from the selected partner. Local search for a technologically similar and socially familiar partner is a natural response of firms but there may an optimum level of diversity that one should seek in a partner when it comes to learning.; The studies lead us to a ‘co-evolution model of alliance and technology networks’ in which the current alliance affects future alliances and technology, and current technology affects future technology and alliances. Alliance and technology networks co-evolve in the sense that prior position of a firm in an alliance and technology network at time 0 guides its subsequent alliance formation and amount of learning. The dynamics of learning and trust play a pivotal role in bridging across networks, moving them forward and reshaping them. |
