公司自媒体可以提供更多的指导性信息吗?——基于企业微信公众号技术创新披露的文本分析

Can Corporate Self-media Offer More Guiding Information?Based on the Textual Analysis for Companies’Wechat Posts on the Disclosure of Technological Innovation

近年来公司自媒体信息披露的传递效率和增量效应日益凸显,与定期报告等正式渠道相比,这一新兴渠道能否纾解企业创新专有性信息披露的困境?本文运用机器学习的方法,对生物医药和汽车行业企业2020—2022年官方微信公众号推文进行了文本分析,从融资有效性的视角探讨了自媒体技术创新披露的文本内容对投资者的增量指导作用。研究发现,公司自媒体技术创新披露水平越高,融资约束程度越低;当自媒体技术创新披露(定性信息)和专利成果(定量信息)相互佐证时(即“言行一致”),会显著强化对融资约束的缓解作用。作用机制检验的结果表明,自媒体技术创新披露主要通过缓解信息不对称程度进而来降低融资约束。进一步研究发现,在双元创新披露策略方面,相较于单一化的披露策略,平衡式的技术创新披露策略可以显著缓解企业融资约束;在融资渠道来源方面,相比于股权融资,自媒体技术创新披露更能吸引债权投资人的注意力;在文本特征方面,更积极的情感语调、更乐观的前景预期和更详实的文本披露对融资约束的缓解作用更加有效。本研究从多个维度构建了公司层面的自媒体技术创新披露指标,证实了自媒体这一新兴的非正式披露渠道的有效性,为完善技术创新披露对融资约束的影响研究提供了理论支持和经验证据。

In formal channels,companies often struggle to balance innovation transparency with proprietary costs,facing the dilemma of innovation disclosure.Recently,the efficiency and impact of information disclosure via informal channels,such as self-media,has become increasingly prominent.It is worth exploring whether this emerging channel can mitigate innovation disclosure challenges and provide more guidance.This paper uses machine learning to analyze the text of high-tech companies’official WeChat posts from 2020-2022,testing whether self-media’s technological innovation disclosures provide incremental guidance information.Specifically,the article constructs firm-level indicators of self-media technological innovation disclosure from multiple dimensions—disclosure level,strategy,and textual features—to explore its effect on mitigating firms’financing constraints.The main findings of this paper are as follows.First,corporate self-media’s disclosure of textual information on technological innovations offers incremental guidance to investors,thereby reducing financing constraints through the alleviation of information asymmetry.Additionally,when the qualitative text disclosed through self-media is supported by quantitative patent outcomes(i.e.,alignment between words and actions),the alleviation of financing constraints is significantly enhanced.This enhancement is attributed to the logical consistency between results and evidence,which bolsters the credibility of these disclosures,enabling these informal channels to offer effective decision-making guidance to investors.Second,this study demonstrates that,compared to radical or incremental technological innovation disclosure strategies,a balanced strategy significantly mitigates financing constraints more effectively.In terms of financing sources,disclosures of technological innovations through self-media garner greater attention from debt investors than from equity financing.Concerning textual characteristics,a more positive emotional tone,more optimistic future projections,and more comprehensive descriptions further enhance the effectiveness in mitigating financing constraints.Third,the paper explores the variable impacts of corporate characteristics and the intensity of industry competition.It finds that disclosures about technological innovations via corporate self-media can more effectively break through financing barriers when management’s shareholding is lower,the company’s size is smaller,earnings management is less prevalent,and industry competition is milder.The potential marginal contributions of this paper are primarily in the following areas.Firstly,it uniquely applies machine learning to capture technological innovation information from corporate self-media more precisely,confirming the method’s scientific validity through multiple verifications.Despite previous studies focusing on the informational effects of corporate self-media on capital markets,this paper pioneers a detailed examination of content adequacy on Wechat,a crucial informal channel.Additionally,it integrates machine learning with expert interviews to enhance the lexicon of innovation information,substantiating the disclosure channel’s role in improving financing efficiency and alleviating constraints.Secondly,the study innovatively merges qualitative self-media indicators with quantitative annual report data within a unified analytical framework,enhancing research on how information disclosure affects financing constraints.It shifts from traditional single-source assessments to explore the interplay of qualitative and quantitative data on financing,delineating how various disclosure forms impact investor decisions and providing actionable insights for management on leveraging both information types.Thirdly,it offers a comprehensive analysis of self-media disclosures,focusing on content depth,novelty,and structure,thus expanding the dimensions of corporate innovation research.Furthermore,the study differentiates between breakthrough and incremental technology disclosures,developing variables to measure and compare these approaches’impacts on economic outcomes,thereby offering fresh perspectives for subsequent investigations.