Interconnected Ownership and Earnings Management: Supplier-Customer Collusion in China’s IPO Context

1. Research Hypothesis & Overview This dataset underpins an empirical study of how the supplier-customer ownership connections influence supplier firms’ earnings management in the pre-IPO period. The core hypothesis is that equity-based ownership ties between pre-IPO supplier firms and their major customers facilitate collusive behavior in the IPO process, leading to higher accrual-based earnings management before listing. We also examine whether regulators and investors recognize and discipline such behavior through IPO rejection, IPO pricing, and post-IPO stock performance. The data structure (as detailed in the accompanying Data Guide) is a firm-year panel for Chinese IPO applicants from 2004–2017, covering the three fiscal years prior to the IPO application and subsequent IPO- and post-IPO-level outcomes. 2. Key Findings The constructed variables support several main findings: Pre-IPO earnings management: Firms with stronger ownership connections to their top five customers exhibit significantly higher pre-IPO abnormal accruals. Regulatory discipline: The China Securities Regulatory Commission (CSRC) is more likely to reject IPOs when high earnings management coincides with strong supplier-customer ownership ties, suggesting regulators respond to the interaction of connections and manipulation. Investor reaction: Ownership-connected supplier firms enjoy higher IPO price premiums but underperform in the long run, indicating that initial optimism is later revised. Heterogeneity: The association between ownership connections and earnings management is stronger for firms without politically connected executives, firms in durable-goods industries, and firms in regions with weaker legal environments. 3. Data Collection Methodology Data are assembled via a rigorous, multi-step process: Sample and pre-IPO window: All firms that submitted domestic IPO applications between 2004 and 2017 are identified. For each, financial and governance data for the three pre-IPO fiscal years are collected. Supplier-customer links and ownership ties: Top-five customers and their sales shares are hand-collected from IPO prospectuses (available for download at Juchao http://www.cninfo.com.cn/new/index). Ownership links between suppliers and customers are manually traced and coded based on data at Tianyancha (https://www.tianyancha.com). This hand-collected approach to constructing the ownership-connection variable ensures that our measure accurately captures the underlying ownership structure along the supply chain for the sampled firms. Financial, governance, and market variables: Accounting, governance, IPO outcome, pricing, and post-IPO return variables come from commercial databases (e.g.,China Stock Market & Accounting Research Database, WIND database).1. 研究假设与概况 本数据集支撑一项实证研究，旨在探究IPO前阶段内，供应商与客户间的股权关联如何影响供应商企业的盈余管理行为。核心假设为：IPO前供应商企业与其主要客户之间的股权关联纽带，会助推IPO进程中的合谋行为，进而导致上市前应计盈余管理水平升高。本研究同时检验监管机构与投资者是否会通过IPO否决、IPO定价以及IPO后股价表现，识别并约束此类盈余管理行为。 本数据集的结构（详见配套数据指南）为2004至2017年中国IPO申请企业的公司-年度面板数据，覆盖IPO申请前的三个完整会计年度，以及后续的IPO相关与IPO后层面的各类结果变量。 2. 核心研究发现 经构建的变量支撑以下几项主要研究结论： IPO前盈余管理：与前五名客户股权关联度更高的企业，其IPO前异常应计利润水平显著更高。 监管约束：当企业存在高盈余管理且同时具备较强的供应商-客户股权关联时，中国证券监督管理委员会（证监会，CSRC）更倾向于否决其IPO申请，这表明监管机构会对关联纽带与盈余操纵的交互行为作出响应。 投资者反应：存在股权关联的供应商企业能够获得更高的IPO溢价，但长期股价表现却不及市场平均水平，这表明市场初始的乐观情绪后续会得到修正。 异质性特征：在未拥有政治关联高管的企业、耐用品行业企业以及法治环境较弱地区的企业中，股权关联与盈余管理之间的正向关联更为显著。 3. 数据收集方法 本数据集通过严谨的多步骤流程进行汇编： 样本与IPO前窗口期：筛选出2004至2017年间提交境内IPO申请的所有企业，并收集每家企业IPO前三个会计年度的财务与治理数据。 供应商-客户关联与股权纽带：前五名客户及其销售占比数据从IPO招股说明书（可于巨潮资讯网http://www.cninfo.com.cn/new/index下载）手工收集得到；供应商与客户间的股权关联则基于天眼查（https://www.tianyancha.com）的数据进行人工追溯与编码。 本次构建股权关联变量所采用的手工收集方法，确保了我们的衡量指标能够精准捕捉样本企业供应链上的实际股权结构。 财务、治理与市场变量：会计、治理、IPO结果、定价以及IPO后收益率等变量均来自商业数据库（如中国股票市场与会计研究数据库、万得（WIND）数据库）。