Can Industrial Intelligence Improve the Quality of Information Disclosure? Evidence from Manufacturing Listed Firms in China
TU Manman, CAO Chunfang, DU Shanzhong
School of Economics & Management, South China Normal University; School of Business, Sun Yat-sen University; International School of Business & Finance, Sun Yat-sen University
Summary:
With the extensive adoption of industrial intelligence, technologies such as industrial robots are rapidly permeating various manufacturing processes and driving enterprises' strategic upgrading. According to data from the International Federation of Robotics, China continues to lead the global market, with 290,258 new industrial robots installed in 2022, accounting for more than 50% of the global total, and maintaining a compound annual growth rate of 32.55% over the past decade. Although previous studies indicate that industrial robots significantly enhance firms' productivity and flexibility, profoundly reshaping labor markets, its influence on corporate information disclosure quality and the underlying mechanisms remains underexplored. Given that high-quality information disclosure is essential for effective capital market functioning, investigating this issue has both theoretical and practical significance. We measure the quality of management earnings forecasts from accuracy and precision, and construct firm-level industrial intelligence adoption based on the identification of related construction-in-progress and R&D expenditures projects from financial statement disclosures. Using a panel of manufacturing listed firms in Chinese A-share markets from 2011 to 2022, our results show that higher industrial intelligence adoption significantly improves the accuracy and precision of management earnings forecasts. This positive effect is primarily driven by reduced uncertainty in production costs associated with automated production, and it is more pronounced in regions with higher statutory minimum wages, greater high-skilled labor supply, and firms with higher inventory ratios. Additionally, industrial intelligence reduces income volatility through greater production controllability, especially in firms experiencing higher product demand volatility and more diversified product portfolios. Further heterogeneity analysis indicates that the positive impact of industrial intelligence adoption is significantly stronger in labor-intensive firms or those facing intense market competition. This study contributes to the literature in three main ways. First, it extends the literature on the economic consequences of industrial intelligence adoption by focusing on the quality of management earnings forecasts. While prior research has primarily examined how robot exposure affects firms' labor decisions and cost management behaviors, this paper investigates how industrial intelligence adoption improves cost controllability and reduces income volatility, thereby enhancing disclosure quality. Second, it enriches the set of determinants of management earnings forecasts' quality. Existing studies have emphasized external factors such as litigation risk and internal governance mechanisms like internal controls, which mainly influence intentional bias in forecasts. In contrast, this study highlights that forecast quality also depends on the quantity and reliability of information available to managers. It shows that industrial intelligence provides managers with more precise and stable operational data, thereby improving the accuracy and precision of earnings forecasts. Third, the study introduces a replicable, fine-grained metric of firm-level industrial intelligence investment derived from financial statement disclosures, offering a new template for researchers interested in tracing technological adoption. The policy implications of this study are as follows: Corporate managers should promote the deep integration of intelligent manufacturing and accounting information systems to ensure more timely and accurate access to information, thereby improving decision-making efficiency. Firms should adopt intelligent strategies tailored to their labor structures and competitive conditions. Labor-intensive firms should proactively adopt automation to mitigate labor-market uncertainties, while highly competitive firms can leverage intelligent systems to enhance forecasting and disclosure quality. For policymakers, targeted policy support should be implemented based on regional industrial characteristics and market conditions. On one hand, improving talent policies and support services can help ensure an adequate supply of skilled labor for “human-machine matching.” On the other hand, fiscal and financial incentives should be provided to firms in highly competitive sectors. Finally, when formulating labor policies such as minimum wage standards, governments should carefully consider the “human-machine substitution effect” driven by rising labor costs, in order to strike a balance between labor protection and industrial intelligence adoption, and to achieve coordinated progress in intelligent upgrading and employment stability.
涂漫漫, 曹春方, 杜善重. 工业智能化应用能提高企业信息披露质量吗？——来自制造业上市公司的证据[J]. 金融研究, 2025, 543(9): 115-132.
TU Manman, CAO Chunfang, DU Shanzhong. Can Industrial Intelligence Improve the Quality of Information Disclosure? Evidence from Manufacturing Listed Firms in China. Journal of Financial Research, 2025, 543(9): 115-132.
[1]步丹璐、文彩虹和Rajiv Banker,2016,《成本粘性和盈余稳健性的衡量》,《会计研究》第1期,第31～37页。 [2]方巧玲、余怒涛和徐慧,2024,《数字化转型的治理效应研究:会计信息质量视角》,《会计研究》第3期,第34～50页。 [3]何小钢、朱国悦和冯大威,2023,《工业机器人应用与劳动收入份额——来自中国工业企业的证据》,《中国工业经济》第4期,第1～19页。 [4]金献坤、徐莉萍和辛宇,2023,《企业数字化与业绩预告可靠性研究》,《会计研究》第2期,第52～64页。 [5]李常青、陈泽艺和黄玉清,2018,《内部控制与业绩快报质量》,《审计与经济研究》第1期,第21～33页。 [6]李晓溪、饶品贵和岳衡,2019,《年报问询函与管理层业绩预告》,《管理世界》第8期,第173～188页。 [7]鲁桐和党印,2014,《公司治理与技术创新:分行业比较》,《经济研究》第6期,第115～128页。 [8]陆瑶、施新政和刘璐瑶,2017,《劳动力保护与盈余管理——基于最低工资政策变动的实证分析》,《管理世界》第3期,第146～158页。 [9]彭博和贺晨,2022,《“互联网+销售”有助于改善管理层预测的质量吗——基于上市公司开设电商店铺的数据分析》,《会计研究》第6期,第75～89页。 [10]权小锋和李闯,2022,《智能制造与成本粘性——来自中国智能制造示范项目的准自然实验》,《经济研究》第4期,第68～84页。 [11]孙新波、钱雨、张明超和李金柱,2019,《大数据驱动企业供应链敏捷性的实现机理研究》,《管理世界》第9期,第133～151页。 [12]王丹、孙鲲鹏和高皓,2020,《社交媒体上“用嘴投票”对管理层自愿性业绩预告的影响》,《金融研究》第11期,第188～206页。 [13]王永钦和董雯,2020,《机器人的兴起如何影响中国劳动力市场?——来自制造业上市公司的证据》,《经济研究》第10期,第159～175页。 [14]温素彬、张金泉和焦然,2022,《智能制造、市场化程度与企业运营效率——基于A股制造业上市公司年报的文本分析》,《会计研究》第11期,第102～117页。 [15]吴义爽、盛亚和蔡宁,2016,《基于互联网+的大规模智能定制研究——青岛红领服饰与佛山维尚家具案例》,《中国工业经济》第4期,第127～143页。 [16]许辉,2019,《“世界工厂”模式的终结?——对“机器换人”的劳工社会学考察》,《社会发展研究》第1期,第143～162页。 [17]闫雪凌、朱博楷和马超,2020,《工业机器人使用与制造业就业:来自中国的证据》,《统计研究》第1期,第74～87页。 [18]尹洪英和李闯,2022,《智能制造赋能企业创新了吗?——基于中国智能制造试点项目的准自然试验》,《金融研究》第10期,第98～116页。 [19]袁航和朱承亮,2018,《国家高新区推动了中国产业结构转型升级吗》,《中国工业经济》第8期,第60～77页。 [20]赵家悦和卢锐,2024,《企业数字化转型的融资价值:基于数字化转型的新度量》,《会计研究》第6期,第55～69页。 [21]周广肃、李力行和孟岭生,2021,《智能化对中国劳动力市场的影响——基于就业广度和强度的分析》,《金融研究》第6期,第39～58页。 [22]Acemoglu, D. and P. Restrepo, 2020, “Robots and Jobs: Evidence from US Labor Markets”, Journal of Political Economy, 128(6), pp. 2188～2244. [23]Ajinkya, B. and Gift, M., 1984, “Corporate Managers' Earnings Forecasts and Symmetrical Adjustments of Market Expectations”, Journal of Accounting Research, 22,pp.425～444. [24]Anderson, M., R. Mashruwala, Y. Wang and R. Zhao, 2023, “Cost Uniqueness and Information Uncertainty”, Contemporary Accounting Research, 40(4), pp. 2226～2255. [25]Banker, R. D. and L. Chen, 2006, “Predicting Earnings Using Model Based on Cost Variability and Cost Stickiness”, The Accounting Review, 81(2), pp. 285～307. [26]Bates, T. W., F. Du and J. J. Wang, 2020, “Workplace Automation and Corporate Financial Policy”, Available at SSRN 3556935, pp. 1～70. [27]Bena, J., H. Ortiz-Molina and E. Simintzi, 2022, “Shielding Firm Value: Employment Protection and Process Innovation”, Journal of Financial Economics, 146(2), pp. 637～664. [28]Beyer, A., D. A. Cohen, T. Z. Lys and B. R. Walther, 2010, “The Financial Reporting Environment: Review of the Recent Literature”, Journal of Accounting and Economics, 50(2), pp. 296～343. [29]Cheng, H., R. Jia, D. Li and H. Li, 2019, “The Rise of Robots in China”, Journal of Economic Perspectives, 33(2), pp. 71～88. [30]Cheng, Q., T. Luo and H. Yue, 2013, “Managerial Incentives and Management Forecast Precision”, The Accounting Review, 88(5), pp. 1575～1602. [31]Cheng, X., E. Lyandres, K. Zhou and T. Zhou, 2024, “Labor-Replacing Automation and Finance”, Management Science, pp. 1～59. [32]Ciftci, M. and F. M. Salama, 2018, “Stickiness in Costs and Voluntary Disclosures: Evidence from Management Earnings Forecasts”, Journal of Management Accounting Research, 30(3), pp. 211～234. [33]Gihleb, R., O. Giuntella, L. Stella and T. Wang, 2022, “Industrial Robots, Workers' Safety, and Health”, Labour Economics, 78, pp. 1～12. [34]Healy, P. M. and K. G. Palepu, 2001, “Information Asymmetry, Corporate Disclosure, and the Capital Markets: A Review of the Empirical Disclosure Literature”, Journal of Accounting and Economics, 31(1), pp. 405～440. [35]Kasznik, R. and B. Lev, 1995, “To Warn or Not to Warn: Management Disclosures in the Face of an Earnings Surprise”, The Accounting Review, pp. 113～134. [36]Mann, K. and L. Püttmann, 2023, “Benign Effects of Automation: New Evidence from Patent Texts”, Review of Economics and Statistics, 105(3), pp. 562～579. [37]Morikawa, M., 2019, “Uncertainty over Production Forecasts: An Empirical Analysis using Monthly Quantitative Survey Data”, Journal of Macroeconomics, 60, pp. 163～179. [38]Qiu, J., C. Wan and Y. Wang, 2024, “Labor-Saving Innovations and Capital Structure”, Journal of Corporate Finance, 84, pp. 102510. [39]Rogers, J. L. and P. C. Stocken, “Credibility of Management Forecasts”, The Accounting Review, 80(4), pp. 1233～1260. [40]Simintzi, E., V. Vig and P. Volpin, 2015, “Labor Protection and Leverage”, Review of Financial Studies, 28(2), pp. 561～591. [41]Toorajipour, R., V. Sohrabpour, A. Nazarpour, P. Oghazi and M. Fischl, 2021, “Artificial Intelligence in Supply Chain Management: A Systematic Literature Review”, Journal of Business Research, 122, pp. 502～517. [42]Trajtenberg, M., 2018, “AI as the Next GPT: A Political-Economy Perspective”, National Bureau of Economic Research, pp. 1～11. [43]Weiss, D., 2010, “Cost Behavior and Analysts' Earnings Forecasts”, The Accounting Review, 85(4), pp. 1441～1471.
