Product innovation in heterogeneous R&D networks pathways to exploration and exploitation

The questions of whether the effective internalization of external variance can help firms (re)balance exploration and exploitation by initiating and interlacing cycles of possibility and action, create superior value, attain a competitive advantage, and/or better cope with market or technological requirements present critical interest for organizational studies scholars. However, they have so far generated comparatively little systematic theoretical inquiry or empirical investigation. This thesis focuses on the effects of ego-centric R&D networks and examines how and under what contingencies diverse external alliances help focal firms grasp and convert varied external information into desirable firm-level outcomes. It answers three related questions: 1) can more heterogeneous R&D networks stimulate innovation, 2) can different types of external R & D ties provide a source of competitive advantage, and 3) can diverse R&D alliance sets improve performance under specific environmental conditions? The three complementary studies included in this dissertation suggest that relational ties spanning different pools of expertise channel a mosaic of naive insights, pseudocapabilities, and fresh opportunities into the focal firm. Together, they show that heterogeneous R & D alliances can offer a source of superior exploration and exploitation revenues, a source of above-average innovativeness and competitive advantage, and a source of improved fit to different environments. The first study argues that participation in heterogeneous R&D networks contributes to both exploration and exploitation up to a certain threshold and that several intraorganizational capabilities operate as intervening mechanisms, mediating between external heterogeneity and innovation outcomes. The second study suggests that interorganizational strategies provide an important but under-studied origin of competitive advantage. It proposes that distinct types of R&D networks generative, horizontal, vertical, and cross-border ties help firms initiate an exploration path and achieve above-average growth, although small firms may benefit less from interoganizational strategies than their larger peers. The third manuscript treats external R&D networks as levers or buffers for idiosyncratic in-house capabilities under specific environmental conditions. It argues that organizationallyand geographically-diverse R&D alliances sets can temporarily support higher levels of performance than internal capabilities may allow, either by strenghtening the performance effects of environmentally-fit capabilities or by weakening the effects of unsuitable ones. The hypotheses were tested on a large sample of Canadian manufacturing firms. Descriptions of network configurations, internal capabilities, and innovation activities over a 3-year period, 1997 to 1999, were extracted from the Survey of Innovation 1999, administered by the Science, Innovation, and Electronic Information Division of Statistics Canada. To obtain a representative cross-section of manufacturing activities across Canada, the design of the Survey of Innovation was stratified by 31 separate manufacturing industry groupings and 12 geographic regions (Canadian provinces and territories). The survey achieved a response rate of 95%, sampling a total of 5,455 manufacturing firms, which were representative of 91.5% of Canadian manufacturing activities in 1997 in terms of manufacturing shipments, 89.7% in terms of manufacturing value-added, and 89.9% in terms of total value-added. The primary data collection was supplemented with independent measures of ex-ante and ex-post firm-level performance from the 1997 and the 1999 Survey of Manufacturers annual censuses of manufacturing establishments conducted each year by the Manufacturing, Construction, and Energy Division of Statistics Canada and with a comprehensive set of secondary industry-level measures gathered from multiple data sources, at different points in time, which helped control for discrepancies in firms' competitive, economic, and technological environments over the 3-year period of the study. The analyses lend support to the majority of the hypothesized effects. The first manuscript shows that crafting R & D networks with moderate levels of heterogeneity fosters both exploration and exploitation. Heterogeneity has a linear positive influence on exploration revenues and a quadratic, inverse U-shaped impact on exploitation revenues, whereby some heterogeneity fosters, but excessive heterogeneity diminishes, focal firms' emphasis on commercializing improved products. Four different subcomponents of absorptive capacity acquisition, assimilation, transformation, and deployment capabilities channel part of the effect of external heterogeneity on both exploration and exploitation revenues, mediating respectively the linear effects of R&D network heterogeneity on exploration and its curvilinear effects on exploitation. Furthermore, this first study shows that the innovation benefits derived