Can the Input of Data Elements Improve Manufacturing Productivity? Effect Measurement and Path Analysis
本文利用中国制造业上市公司2010-2023年数据，通过文本分析和随机前沿模型，发现数据要素投入主要通过提升技术效率来提高全要素生产率，而数据采集环节无显著作用，且数字化对产业结构的影响尚未促进技术进步。
We first used text analysis methods to define and measure the level of data element input. We qualitatively demonstrated that data element input can improve total factor productivity (TFP) by constructing a new classical economic growth model by adding data elements. On this basis, we built a translog stochastic frontier model to incorporate data elements into the production function and TFP measurement model. Using data from Chinese manufacturing listed companies from 2010 to 2023, we quantitatively measured and dynamically evaluated the impact of data element input on manufacturing TFP and the role of technical efficiency and technological progress. The results revealed the following: 1) Data element input as a whole is beneficial for improving manufacturing TFP, but the main path is the improvement of technical efficiency. Additionally, data processing and application significantly improve TFP, whereas data acquisition does not. 2) The impact of digitalization on the current industrial structure has not affected technological progress, but it has restricted improvements in technical efficiency. Data elements are increasingly becoming the critical material basis for the manufacturing industry's digital transformation. In this context, this study has the following practical value: 1) It helps better identify the critical path of data elements to empower manufacturing industry TFP to implement more targeted digital transformation in practice; and 2) it contributes to a more comprehensive understanding of the impact of digitalization on the manufacturing industry structure to fully leverage the positive role of data elements in enhancing enterprise productivity.