Abstract
In an increasingly competitive global market, knowledge has become the cornerstone of enterprise development. This study investigates how a firm's position within a knowledge network—specifically its ability to bridge structural holes—influences its innovation performance. By analyzing Chinese manufacturing firms in the electronics sector from 2008 to 2020, the research identifies technological diversification as a critical mediator that translates network advantages into innovative success.
Introduction
Modern innovation is rarely an isolated endeavor; it is the product of complex "knowledge networks" where information and collaborative learning flow between entities. However, the mere existence of a network connection is insufficient for success. This research addresses the structural question: does occupying a "bridge" position (a structural hole) between unconnected knowledge clusters enhance innovation? Furthermore, it explores the mechanism of this impact, specifically whether network positioning improves innovation directly or by fostering technological diversification first.
Firms occupying structural holes act as gatekeepers, accessing non-redundant information from disparate groups. This unique positioning prevents "path dependency" and provides the "information advantage" necessary to anticipate market shifts and the "resource advantage" to acquire new technologies efficiently.
Methodology
The study employed a longitudinal quantitative approach focusing on 352 Chinese manufacturing enterprises in the computer and electronics sectors. Data was sourced from the incoPat database for patent classifications and CSMAR for financial metrics. The final filtered dataset included 11,127 patent data points from companies maintaining continuous R&D activity throughout the 2008–2020 period.
Network and Variable Construction
- Knowledge Network: Built using International Patent Classification (IPC-4) codes. Connections were established when two technical categories appeared on the same patent.
- Structural Hole Efficiency: Measured using the Structural Hole Efficiency Index to quantify how effectively a firm connects otherwise isolated knowledge elements.
- Technological Diversification: Calculated using the Entropy Index based on the variety of IPC-4 codes held by an enterprise.
- Analysis: The researcher utilized rolling 4-year time windows to capture the evolving nature of the network and applied multiple regression analysis to test the hypotheses.
Key Results
The empirical analysis validated the study's core hypotheses, demonstrating a clear path from network structure to innovation performance.
- Direct Impact of Embedding: Structural hole efficiency has a significant positive effect on innovation performance. As shown in Model 2, the regression coefficient is 1.844, indicating that firms bridging knowledge gaps are more innovative.
- Driving Diversification: Stronger network embedding leads to higher levels of technological diversification (Model 4, coefficient: 0.005).
- Diversification as a Catalyst: Technological diversification itself is a powerful driver of innovation, with a substantial regression coefficient of 9.593 in Model 6.
- The Mediation Effect: Model 11 confirms that technological diversification partially mediates the relationship between network embedding and innovation. By diversifying, firms improve their "absorptive capacity"—the ability to recognize and apply external information—thereby reducing R&D risks and costs.
Discussion and Conclusion
The findings suggest that technological diversification is not merely a byproduct of growth but a functional engine for innovation. By positioning themselves at the intersections of different knowledge fields, enterprises can effectively acquire heterogeneous resources. This synergy allows firms to balance capabilities across fields, fostering sustainable innovation.
Limitations and Future Work
While providing robust evidence for structural embedding, the study did not account for relational embedding—the quality and trust within individual connections. Future research should integrate these qualitative factors and consider external variables like policy shifts and market volatility to provide a more comprehensive view of the innovation landscape.
Summary updated