摘要
结合双重创新理论和创新价值链理论,从双重创新价值链视角构建多维创新能力概念框架,将国家创新能力解构为知识创造、知识吸收共享、技术商业化和技术聚集辐射4个子维度创新能力。利用多维创新能力指标体系评价2010—2022年全球66个国家的创新能力,采用空间自相关模型、多尺度地理加权回归模型和空间收敛模型揭示国家创新能力的空间异质性。研究发现:①国家创新能力存在显著的空间依赖性特征,并逐步形成以西欧、北美、亚太为核心的三极空间分布格局,发达国家知识吸收共享能力领先于知识创造、技术商业化及技术聚集辐射能力,中国、印度等发展中国家则相反;②不同驱动因素的空间作用尺度存在一定分异,其中多元化生产水平、法制水平和政治稳定性对国家创新能力空间异质性的影响较大,而信息技术发展水平、人口密度和经济发展水平对国家创新能力空间异质性的影响较小;③国家创新能力呈现收敛异质性特征,全样本国家和发展中国家的国家创新能力存在显著为负的空间绝对β收敛和空间条件β收敛,而发达国家的国家创新能力只存在空间条件β收敛且不存在空间绝对β收敛。
Mounting evidence indicates that the objective and accurate measurement of national innovation capabilities,along with the exploration of pathways for their enhancement,has become a prominent and longstanding concern in academia.This urgency is particularly pronounced in the context of growing international competition,where innovation has become a cornerstone of national competitiveness and economic resilience.However,despite extensive research into the frameworks,concepts,and metrics for evaluating innovation,the spatial heterogeneity of innovation capabilities remains underexplored.The concept of spatial heterogeneity in national innovation capabilities highlights the non-uniform distribution and aggregation of innovation levels across different geographic areas.It also addresses the diverse relationships and impacts of driving factors,which are contingent upon their spatial attributes.Furthermore,it encompasses the parameter heterogeneity evident in spatial convergence processes,which reflect disparities in development velocities,trajectories,and influencing factors among regions.Neglecting these spatial dimensions not only limits the understanding of how innovation diffuses geographically but also exacerbates analytical challenges in modeling the complex,evolving nature of national innovation systems.Moreover,traditional linear“black-box”frameworks for evaluating innovation capabilities fail to capture the multistage,sequential,and interactive nature of innovation,leading to incomplete assessments that neglect external exploitation capabilities,including the absorption,adaptation,and utilization of global knowledge and technology.As a result,there is a pressing need to move beyond one-dimensional assessments and adopt more comprehensive frameworks that integrate spatial heterogeneity and multistage innovation dynamics.To address these gaps,the study adopts a dual innovation value chain lens,deeply rooted in dual innovation theory and the innovation value chain framework.This perspective underscores that the essence of innovation capabilities extends beyond internal procedures to include the strategic harnessing of global knowledge networks for external exploitation.By dividing the innovation value chain into internal innovation development and external innovation exploitation,the framework captures the multifaceted nature of innovation capabilities.Internally,innovation encompasses activities such as knowledge creation,research,and technology development,while externally,it involves the absorption,adaptation,and dissemination of innovations within global value chains.To operationalize this framework,national innovation capabilities are deconstructed into four sub-dimensions—knowledge creation,knowledge absorption and sharing,technology commercialization,and technological agglomeration and radiation—capturing the comprehensive stages of innovation generation,dissemination,and utilization.Using a multidimensional innovation capabilities index system,this study evaluates the innovation capabilities of 66 countries globally from 2010 to 2022.Methodologically,spatial autocorrelation models reveal patterns of regional clustering in innovation,while multi-scale geographically weighted regression models capture how the drivers of innovation vary across different spatial scales.Spatial convergence models,by examining the dynamics of innovation capability disparities and revealing whether countries are converging toward similar innovation levels or diverging further apart,complement a robust analytical foundation that underscores the spatial and temporal dynamics of innovation capabilities while highlighting the interplay between internal development and external engagement.The study yields several key findings regarding national innovation capabilities.First,national innovation capabilities exhibit strong spatial dependence,forming a three-pole distribution centered in Western Europe,North America,and the Asia-Pacific region.Spatial disparities within this framework reveal that developed countries excel in knowledge absorption and sharing,while developing nations like China and Russia face challenges in this dimension,highlighting systemic gaps in collaboration,cross-sectoral synergy,and technology diffusion.Second,the drivers of innovation capabilities display spatial heterogeneity in their scale and significance.Factors such as production diversification,rule of law,and political stability contribute significantly to spatial differentiation,while others like information technology development and population density have more modest impacts.Lastly,convergence heterogeneity is evident,as developing nations and the global sample demonstrate significant spatialβ-convergence,reflecting a“catch-up”effect where low-capability nations narrow the gap with high-capability counterparts,while developed countries show only conditional convergence without absoluteβ-convergence trends.These findings underscore the complex and differentiated trajectories of innovation systems,highlighting the need for tailored strategies to foster balanced growth and global competitiveness.
作者
覃雄合
陈东成
卢函
Qin Xionghe;Chen Dongcheng;Lu Han(School of Economics,Hefei University of Technology,Hefei 230009,China;School of Economics and Management,Shanghai Maritime University,Shanghai 201306,China)
出处
《科技进步与对策》
北大核心
2025年第24期48-60,共13页
Science & Technology Progress and Policy
基金
国家自然科学基金项目(42001124,42401216)
教育部人文社会科学研究基金项目(23JHQ060)
中央高校基本科研业务费专项资金资助项目(JZ2024HGTB0252)。
关键词
双重创新价值链
国家创新能力
空间收敛
空间异质性
多尺度地理加权回归
Dual Innovation Value Chain
National Innovation Capabilities
Spatial Convergence
Spatial Heterogeneity
Multi-Scale Geographically Weighted Regression
