|
|
|
| Multimodal Accounting Information and Investor Attention——Evidence from Official Weibo of Listed Companies |
| LI Qingyuan, LI Houyuan, HU Longyin
|
School of Economics and Management, Wuhan University;
School of Economics and Management, Hubei University of Technology |
|
|
|
Abstract Amid the wave of “digital intelligence” transformation, enterprises are actively integrating cutting-edge technologies such as big data, machine learning, and multimedia into finance and accounting. This transformation has profoundly impacted the form of information disclosure. Traditionally, enterprise information disclosure has primarily relied on single structured data, but it is now gradually shifting to multimodal accounting information disclosure encompassing various forms such as text, images, audio, and video. This transition undoubtedly poses new challenges for financial accounting supervision. Against this background, this study aims to utilize Natural Language Processing (NLP) and Optical Character Recognition (OCR) technologies, relying on the social platform Sina Weibo, to construct multimodal accounting information disclosure indicators applicable to Chinese listed companies and delve into its impact mechanism on investor attention.
The marginal contributions of this paper are as follows: Firstly, from the perspective of investor attention, this paper examines the economic impact of multimodal accounting information disclosure. Existing research mainly focuses on the analysis of structured information and textual information, without fully exploring the value of non-textual information such as images and videos, resulting in omitted variable endogeneity. By employing NLP and OCR technologies, this paper identifies and captures texts, images, and videos containing accounting information posted on the official Weibo accounts of Chinese A-share listed companies on announcement days. Through dimensionality reduction and mechanism testing, it provides empirical evidence for the impact of multimodal accounting information disclosure on investors among Chinese listed companies, complements research on investor attention in the context of multimodal accounting information disclosure in China, and offers technical support and theoretical foundations for understanding the economic consequences of multimodal accounting information disclosure among Chinese listed companies. Secondly, by integrating information theory from a computer science perspective with accounting information theory, this paper explores the crucial role of information characteristics in accounting information transmission. Existing research emphasizes the link between management disclosure and investor attention, often neglecting the impact of the content and form of corporate accounting information disclosure on investor attention. This study improves this based on computer science's information theory. While focusing on information transmission, this paper also examines how information characteristics affect its dissemination effect in the accounting field, contributing to a comprehensive understanding of how investors interpret and respond to different forms of accounting information disclosure. Thirdly, this paper examines the heterogeneity and economic consequences of multimodal accounting information disclosure in enhancing investor attention, providing empirical evidence for governments to formulate relevant accounting information disclosure policies. This paper finds that although multimodal accounting information disclosure can increase investor attention, it also widens market bid-ask spreads and volatility, reducing market liquidity, which has practical significance for further improving information disclosure regulatory mechanisms.
The research results indicate that multimodal accounting information disclosure can attract investor attention through abundant incremental information and significant visual effects. Further analysis reveals that the effectiveness of multimodal accounting information disclosure is closely related to investors' basic characteristics. Specifically, when the institutional investor ownership ratio of a company is relatively low, multimodal accounting disclosure has a more significant effect on enhancing investor attention. However, despite the positive role of multimodal accounting information disclosure in increasing investor attention during the information acquisition stage, it may also lead to widened market bid-ask spreads and increased volatility, thereby reducing market liquidity.
Based on the research conclusions, this paper proposes to improve financial accounting supervision policies in the new era, standardize the supervision of Internet financial accounting information, establish guidelines for enterprise informal disclosure, ensure the authenticity, accuracy, and completeness of disclosed information, and prevent enterprises from misleading investors using multimodal disclosure. Penalties for corporate financial fraud should be increased, and market mechanisms should be further improved to leverage market competition, increase the cost of opportunistic manipulation of accounting information by enterprises, prompt enterprises to enhance the quality of accounting information disclosure, and avoid overuse or abuse of this disclosure method, which could distract investors' attention or cause overreactions. On the basis of improving the supervision mechanism, investor education should be strengthened, improve the professionalism of institutional investors and enhance their information acquisition and screening capabilities, thereby efficiently allocating their resources. Meanwhile, protection for retail investors should be strengthened, penalties for corporate violations should be increased, a transparent and fair market atmosphere should be created, and investor confidence should be bolstered.
|
|
Received: 20 November 2023
Published: 01 November 2024
|
|
|
|
|
|
|
|
