增强发展韧性是我国制造业抵御外部不确定性冲击、实现制造大国向制造强国跨越的必然要求。基于2010—2022年我国内地30个省份面板数据，综合运用动态门槛效应模型、有调节的中介效应模型和差分GMM模型，研究数字化转型对制造业发展韧性的影响。结果发现：考察期内我国制造业发展韧性存在差距，但整体上呈现缩小趋势；数字化转型对制造业发展韧性具有显著提升作用且存在动态门槛效应；数字化转型对制造业发展韧性的影响存在企业、行业和区域异质性。从作用机制来看，技术创新具有中介效应，融资约束和外源融资发挥有调节的中介效应。进一步分析，制造业数字化转型存在显著的行业与地区同群效应。

Since the implementation of the "Made in China 2025" strategy in 2015, the size of China's manufacturing industry has grown substantially, and its overall capacities have greatly improved. China has held the title of the world's largest manufacturing nation for 13 straight years. Meanwhile, the world is undergoing the most significant changes, characterized by a cooling global economy, the emergence of new trade protectionism and deglobalization trends, increasing geopolitical conflicts, and other global economic uncertainties. China's manufacturing industry is facing a progressively complex external environment. Enhancing its manufacturing resilience is an inevitable requirement for China's manufacturing industry to resist external uncertainties, promote high-quality development of the manufacturing industry, and smoothly realize the transition from a "big manufacturing country" to a "powerful manufacturing country". Digital transformation is the process by which enterprises utilize digital technologies to enhance efficiency, improve business processes, and innovate value creation methods. It can enhance manufacturing resilience through various means, such as promoting production process reengineering, enhancing collaborative innovation capabilities, and improving resource allocation efficiency and total factor productivity. Therefore, there are significant theoretical and practical implications for exploring how digital transformation can boost the resilience of manufacturing.
There is limited research on manufacturing resilience through the lens of digital transformation.Using panel data samples from 30 provinces from 2010 to 2022, this study adopts the dynamic threshold effect model, the moderated mediation effect model and the differential GMM model to analyze the impact of digital transformation on the development resilience of the manufacturing industry. It is found that (1) from a perspective of evolution, the resilience of China's manufacturing industry has slightly declined. Although the difference in manufacturing resilience varies among provinces, it shows an overall shrinking trend. (2) The regression results of the benchmark model demonstrate that digital transformation has a significant enhancing effect on manufacturing resilience. Furthermore, this effect is not linear. There exists a dynamic threshold effect, and when the level of digital transformation meets or exceeds the threshold, its impact on manufacturing resilience will exhibit a marginal decline. (3) The results of heterogeneity test indicate that the impact of digital transformation on manufacturing resilience varies among enterprises, industries, and regions. Compared to private enterprises and small and medium-sized enterprises, digital transformation has a more significant impact on enhancing the manufacturing resilience of state-owned or large enterprises; in low threshold areas, except for textiles, there is a significant enhancing effect on the manufacturing resilience of various subsectors of the manufacturing industry, while there are differences in effects within high value areas; compared to eastern and central China and low human capital areas, digital transformation has a more significant impact on enhancing manufacturing resilience in the western region and high human capital areas in China. (4) During the process of enhancing manufacturing resilience, digital transformation has a moderated mediating effect.Technological innovation is the mediating variable, and financing constraints and external financing are the moderating variables. Simultaneously, this digital transformation has significant industry and regional homogeneity effects. Hence, China should pay more attention to enhancing manufacturing resilience, and continuously strengthening the foundation of digital transformation by vigorously promoting a "new infrastructure". To enhance targeted digital transformation and manufacturing resilience from a heterogeneous perspective and consider technological innovation and financing convenience, it is necessary to fully leverage the peer effects of digital transformation and accelerate the process of digital transformation in the manufacturing industry through a digital transformation resource sharing platform.
The marginal contribution of this article lies in three aspects. The study first adopts a single indicator method to measure manufacturing resilience, which avoids the causal inversion problem that may occur when using a multi-factor indicator system to measure manufacturing resilience. Then it analyzes the nonlinear relationship between digital transformation and manufacturing resilience based on a dynamic threshold effect model, which enriches the theoretical connotation of the relationship between digital transformation and manufacturing resilience. Lastly, it advances the understanding of the relationship between digital transformation and manufacturing resilience by clarifying the internal mechanism of the relationship between digital transformation and manufacturing resilience, and the peer effects of digital transformation on manufacturing in different industries and regions.