| 程时雄,何铭文.外部供应风险、技术结构与自主创新能力提升[J].数量经济技术经济研究,2026,(10):166-191 | | 外部供应风险、技术结构与自主创新能力提升 | | External Supply Risks, Technological Structure, and the Enhancement of Indigenous Innovation Capabilities | | | | DOI： | | 中文关键词: 外部供应风险 技术结构 技术对外依赖 自主创新能力 | | 英文关键词: External Supply Risks Technological Structure External Technological Dependence Indigenous Innovation Capabilities | | 基金项目: | | | 中文摘要: | | 在全球技术体系深度重构与“脱钩断链”风险上升的背景下，企业面对外部技术供给不稳定时如何调整技术来源结构与技术对外依赖程度，已成为统筹开放合作与产业链供应链安全的重要现实问题。本文利用授权专利对外引用数据测算企业技术对外依赖程度，基于年报文本采用词嵌入方法测度供应链中断风险，考察外部供应链中断风险对企业技术对外依赖程度的影响及其作用机制，并进一步分析技术对外依赖调整的创新后果。研究发现，面对供应链中断风险上升，企业会显著降低技术对外依赖程度。机制分析表明，企业主要通过推动技术来源多元化和知识结构自主化，进而降低技术对外依赖程度。异质性分析发现，上述作用在高科技行业和出口竞争力较弱企业中更为明显，且企业对美国、英国等国技术来源的依赖下降更为显著。进一步分析表明，降低企业技术对外依赖并未削弱其创新能力，反而有助于提升企业自主创新能力。本文从企业主动应对外部技术供给约束的视角，揭示企业通过调整技术来源结构降低技术对外依赖的微观逻辑，为全球技术体系重构背景下统筹开放合作、技术自主与产业链供应链安全提供经验证据。 | | 英文摘要: | | Against the backdrop of restructuring in the global technological system and rising risks of decoupling and supply chain fragmentation, how firms adjust their technology sourcing structures and technological external dependence has become important for coordinating openness, technological autonomy, and supply chain security. Using Chinese A-share listed firms from 2010 to 2023, this paper constructs a firm-level measure of technological external dependence based on citations to foreign patents and measures supply chain disruption risk using word embedding techniques applied to annual reports. It examines how supply chain disruption risk affects firms’ technological external dependence, the mechanisms underlying this adjustment, and its subsequent innovation consequences.
The results show that supply chain disruption risk significantly reduces firms’ technological external dependence on average. As uncertainty increases regarding the availability, continuity, and controllability of key external technologies, firms reassess their technology sourcing structures and reduce reliance on high-risk external technological paths. Instrumental-variable estimation, double machine learning, and identification based on predetermined foreign-technology exposure provide further support for the baseline result. The effect also varies with firms’ initial foreign-technology exposure. The marginal effect of supply chain disruption risk becomes more negative as baseline exposure rises. Firms with low initial exposure may continue to use accessible foreign technologies because switching costs and continuity considerations remain important, whereas highly exposed firms face greater disruption losses and therefore stronger incentives to reduce concentrated external dependence.
Mechanism analysis identifies two main adjustment channels. First, supply chain disruption risk promotes technology source diversification. After standardizing foreign citation portfolios to a common citation scale, firms expand the number of foreign source countries and non-US foreign source countries, while the relative position of non-US sources rises and concentrated dependence on the United States declines. These results indicate that firms broaden their technology sourcing portfolios rather than simply substituting one foreign source for another. Second, supply chain disruption risk promotes greater autonomy in firms’ knowledge structures. The relative position of self-owned and controllable knowledge increases, and the absolute stock of self-owned knowledge also rises after accounting for overall patent citation activity. Thus, knowledge-structure adjustment reflects both a relative shift away from foreign knowledge and additional internal knowledge accumulation.
Heterogeneity analysis shows that the negative effect of supply chain disruption risk on technological external dependence is stronger among firms in high-tech industries and firms with weaker export competitiveness. When dependence is decomposed by source country, firms’ dependence on technologies from the United States, the United Kingdom, and Russia declines more significantly. The country-level evidence further shows that changes in relative source shares should not be interpreted as simple substitution across foreign countries, because they may also result from different rates of contraction in absolute citation volumes.
Further analysis examines the innovation consequences of dependence adjustment. Lower technological external dependence is significantly associated with stronger subsequent indigenous core innovation and a higher probability of entering a high indigenous innovation state. This relationship remains stable over longer forward-looking horizons, suggesting that dependence adjustment is accompanied by improved subsequent indigenous innovation performance rather than a weakening of firms’ innovative capacity.
This paper contributes to the literature in three respects. First, it links supply chain disruption risk directly to firms’ technology sourcing decisions and provides micro-level evidence on how firms adjust technological external dependence under rising external supply uncertainty. Second, it identifies technology source diversification and greater knowledge-structure autonomy as two specific channels through which firms reduce external dependence. Third, by combining annual-report text with patent citation information, it develops decomposable firm-level measures that distinguish overall external dependence, source-country structure, and internal knowledge accumulation. The findings provide empirical evidence for understanding how firms can preserve access to global knowledge while reducing concentrated technological dependence and strengthening supply chain security. | | 查看全文 相关附件： 下载数据代码附录 |
| | |
