混合所有制改革、政治关联与企业创新 ——基于双重/无偏机器学习方法

Mixed-ownership Reform, Political Connection and Enterprise Innovation Based on Double/Debiased Machine Learning Method

摘要:

企业提高创新能力是推动高质量发展的重要举措，那么混合所有制改革是否促进了国有企业创新？本文基于2008—2019年国有上市企业混合所有制改革样本数据，使用双重/无偏机器学习方法，研究国有企业混合所有制改革对企业创新水平的影响。研究表明，国有企业混合所有制改革有效促进了企业创新水平的提升；同时，引入具有政治关联的民营股东能进一步提升企业的创新水平。进一步机制研究表明，国有企业混合所有制改革通过提高创新投入强度与改善创新效率两途径促进企业创新水平。上述研究结果对时下进行的国有企业混合所有制改革具有参考借鉴意义。

Abstract:

Increasing the innovation capability of enterprises is an important measure to promote high-quality development, so does the reform of mixed ownership promote the innovation of state-owned enterprises? Based on a sample of mixed ownership reform of state-owned listed companies from 2008 to 2019, this paper used the Double/Debiased Machine Learning (DML) method to study the impact of state-owned enterprise mixed ownership reform on enterprise innovation. The results of the study found that the mixed ownership reform of state-owned enterprises has effectively promoted the improvement of enterprise innovation; meanwhile, the introduction of politically connected private shareholders has a more significant role in promoting corporate innovation. Further mechanism research shows that the mixed-ownership reform of state-owned enterprises promotes enterprise innovation through two ways: increasing the intensity of innovation input and improving innovation efficiency. The above research results have reference significance for the current state-owned enterprise mixed ownership reform.