数字化转型对高端装备制造企业经营绩效的影响研究——基于运营效率中介效应的实证分析

Research on the Impact of Digital Transformation on the Operating Performance of High-End Equipment Manufacturing Enterprises—An Empirical Analysis Based on the Mediating Effect of Operational Efficiency

摘要: 本文聚焦中国资本市场中高端装备制造这一战略性新兴产业，以2015~2023年沪深A股高档数控机床、智能制造装备、航空航天装备、海洋工程装备、轨道交通装备及新能源装备等细分领域上市公司为研究样本，构建“数字化转型–运营效率–经营绩效”的逻辑分析框架，运用双向固定效应面板模型与逐步回归中介效应检验法，系统探究数字化转型对高端装备制造企业经营绩效的影响及其传导机制。实证结果显示：第一，数字化转型对企业经营绩效具有显著正向促进作用，数字化转型指数(即年报中数字化相关关键词总频次加1后取自然对数)每提升一个单位(约相当于关键词频次扩大e ≈ 2.72倍)，企业总资产净利润率提升约0.38个百分点(t = 3.71, p < 0.01)；第二，Sobel中介检验Z值为4.74 (p < 0.001)，运营效率(总资产周转率)发挥显著的部分中介作用，中介效应占总效应的比例为25.25%；第三，分组回归显示，国有企业子样本中数字化转型系数(0.0054, t = 3.19)显著高于非国有企业子样本(0.0029, t = 2.25)，呈现明显的产权异质性特征。稳健性检验方面，替换被解释变量(ROE)及解释变量滞后一期处理后，核心结论依然成立。本文丰富了高端装备制造业数字化转型经济后果的微观实证证据，可为企业数字化战略制定及差异化产业扶持政策设计提供决策参考。

Abstract: This paper focuses on high-end equipment manufacturing—a strategic emerging industry in China’s capital market. Using a panel dataset of A-share listed companies from 2015 to 2023 in six sub-sectors (high-grade CNC machine tools, intelligent manufacturing equipment, aerospace equipment, marine engineering equipment, rail transit equipment, and new energy equipment), this study constructs a “Digital Transformation-Operational Efficiency-Operating Performance” analytical framework and applies a two-way fixed effects panel model together with stepwise mediation tests. The empirical results show that: first, digital transformation significantly improves operating performance—a one-unit increase in the digital transformation index (defined as the natural logarithm of one plus the total frequency of digital-related keywords disclosed in annual reports, hence equivalent to an e-fold rise in keyword frequency) raises ROA by approximately 0.38 percentage points (t = 3.71, p < 0.01); second, the Sobel test (Z = 4.74, p < 0.001) confirms that operational efficiency (measured by total asset turnover) plays a significant partial mediating role, accounting for 25.25% of the total effect; third, subsample regressions reveal pronounced ownership heterogeneity—the digital transformation coefficient for state-owned enterprises (0.0054, t = 3.19) is significantly higher than that for non-state-owned enterprises (0.0029, t = 2.25). The conclusions remain robust after replacing the dependent variable with ROE and lagging the explanatory variable by one period. This study contributes micro-level evidence on the economic consequences of digital transformation in the high-end equipment manufacturing industry and provides reference for corporate strategy formulation and differentiated industrial policy design.