| 张玮艺,陆毅,王勇,孙震.个体数字信用的声誉价值——基于在线二手商品交易数据的实证研究[J].数量经济技术经济研究,2026,(6):163-184 | | 个体数字信用的声誉价值——基于在线二手商品交易数据的实证研究 | | The Reputational Value of Digital Credit: An Empirical Study Based on Online Second-hand Transactions | | | | DOI： | | 中文关键词: 数字信用 信用评级 声誉价值 信息不对称 | | 英文关键词: Digital Credit Credit Rating Reputation Information Asymmetry | | 基金项目: | | | 中文摘要: | | 健全社会信用体系是构建高水平社会主义市场经济体制的重要基石，也是畅通国内大循环、构建新发展格局的关键环节。随着数字经济的发展，数字信用已从金融风控工具演变为降低市场交易成本、优化营商环境的新型基础设施。现有研究多局限于信贷场景，缺乏关于数字信用缓解P2P商品交易中道德风险的因果证据。本文借助P2P商品交易平台数据，使用断点回归设计，剥离信用评级的分数效应与标签效应，识别数字信用评级的声誉价值。研究发现,当卖家的信用评级由优秀提升至极好时，商品成交率显著提高3.68%，证实数字信用评级具备显著的声誉价值。声誉价值在跨区域、跨群体及非标准化商品等高信息不对称场景中作用更为凸显，表明数字信用能够突破地缘与社会距离限制，建立广泛的数字化制度信任。数字信用作为通用声誉与平台内生的本地声誉呈显著互补关系，共同构成了多层次的市场信任生态。上述发现表明，数字信用能够作为一种市场化的声誉工具，有效约束P2P交易中的机会主义行为，为拓展数字信用在非金融场景下的应用提供了重要的实证依据，也为在数字经济下深化社会信任建设、降低制度性交易成本指明了可行的方向。 | | 英文摘要: | | A robust social credit system is a fundamental cornerstone of a well-functioning modern market economy, as it is crucial to facilitate efficient economic circulation and foster sustainable development. With the rapid expansion of the digital economy, Digital Credit has evolved from a conventional financial risk-management tool into a critical digital infrastructure that curtails market transaction costs and optimizes the broader business environment. In peer-to-peer commodity markets, transactions are predominantly one-off interactions. Consequently, traditional local reputation mechanisms relying on historical feedback frequently fail, as sellers struggle to accumulate sufficient reviews, leading to severe moral hazard and substantial transaction frictions. Digital credit offers a viable solution by functioning as a transferable General Reputation that encapsulates user behavior across multiple domains, amplifying the potential cost of opportunistic behavior and providing buyers with a centralized, low-cost mechanism for trust verification.
Despite its increasing application, existing economics literature remains largely confined to credit market scenarios, primarily focusing on how Digital Credit mitigates adverse selection through signaling. As such, limited causal evidence exists on how Digital Credit alleviates moral hazard in non-financial P2P commodity transactions. Furthermore, previous empirical studies faced the fundamental identification challenge of distinguishing whether an increased transaction success rate is driven by the Score Effect (which reflects the seller’s intrinsic continuous quality) or by the Label Effect (which captures the pure reputation signal derived from the categorical credit rating itself). It is crucial to clarify this distinction, in order to understand whether Digital Credit merely sorts high-quality sellers or actively disciplines market behavior.
To address this issue, this study utilizes comprehensive micro-transaction data from Xianyu, a prominent P2P online marketplace, and employs a Regression Discontinuity Design to evaluate the reputation value of Digital Credit. By exploiting the threshold where a seller’s Zhima Credit rating discretely upgrades from “Excellent” to “Outstanding” at exactly 700 points, this study cleanly disentangles the Label Effect from the Score Effect. Because the credit scoring algorithm operates as a black box, users are unable to precisely manipulate their scores to cross the threshold. Consequently, those sellers situated within a narrow bandwidth on either side of the cutoff possess nearly identical observable and unobservable characteristics, thus enabling a robust causal estimation of the discrete impact of credit label.
This study has three main findings. First, the upgrade of a seller’s credit rating from “Excellent” to “Outstanding” was found to significantly increase their transaction rate by 3.68%. This improvement is quantitatively equivalent to the effect of a seller lowering their listing price by 0.74%, confirming that Digital Credit ratings carry substantial reputation value. Second, this reputation value is significantly moderated by the degree of information asymmetry within the trading environment. The reputation value is significantly more pronounced in those transactions involving non-standardized commodities, cross-city interactions where geographical distance limits offline verification, and cross-gender transactions where social distance is greater. This demonstrates that Digital Credit can effectively transcend geographical and social barriers to establish broad institutional trust. Third, as a general reputation, Digital Credit significantly complements the platform’s endogenous local reputation derived from historical sales records, jointly forging a multi-level market trust ecosystem. Moreover, the efficacy of the Digital Credit is constrained by specific transaction attributes and institutional environments. Its reputation value diminishes in the exchange of high-value commodities and within cities with lower degrees of marketization, where buyers rely more heavily on personalized trust and in-depth verification.
Ultimately, these findings suggest that Digital Credit can serve as a strong, market-oriented reputation tool to effectively constrain opportunistic behavior in P2P transactions. This study provides robust empirical evidence for the expansion of the application of Digital Credit infrastructure into non-financial scenarios, thereby offering a viable pathway to expand social trust and reduce institutional transaction costs in the digital era. | | 查看全文 相关附件： 下载数据代码附录 |
| | |
