社会信任对债券评级质量的影响研究

LZUFE-KMS > 金融学院

作者	孙旭
姓名汉语拼音	Sun Xu
学号	2021000005073
培养单位	兰州财经大学
电话	17852040991
电子邮件	934165339@qq.com
入学年份	2021-9
学位类别	专业硕士
培养级别	硕士研究生
一级学科名称	金融
学科代码	0251
第一导师姓名	丁汝俊
第一导师姓名汉语拼音	Ding Rujun
第一导师单位	兰州财经大学
第一导师职称	教授
题名	社会信任对债券评级质量的影响研究
英文题名	Research on the impact of social trust on bond rating quality
关键词	社会信任信用评级主观偏信声誉效应
外文关键词	social trust ; credit rating ; subjective preference ; reputation effect
摘要	最近几年我国的信用评级市场正在经历着较大的波动，大量不利事件的冲击使得本就不明朗的市场更加暗流涌动，在此影响下，信用评级机构的效力与口碑正在饱受投资者的质疑，同时由于信用评级机构在对市场主体评级时存在不够及时、不够准确、质量低下等问题，于是信用评级机构的时效性与准确性问题被推到了风口浪尖之上，因此在目前复杂多变的经济背景下研究信用评级质量是被什么因素影响的、如何影响的是十分要紧的。自 2022 年来山东大量民企倒闭破产，过程中呈现出信用评级级别较高、集中、连锁等特点，“山东担保圈”现象愈演愈烈。企业之间存在互相联保互保的现象，尤其是在疫情之后我国经济大环境不明朗的背景下，单个企业如果出现了较大的资金缺口甚至资金链断裂，那么相关区域内所有与之互保联保企业的信用状况都将受到严重影响；另外，企业并没有完全披露对外担保信息，对应的企业的信用风险就无法完全通过信用评级体系呈现出来，这也就使得企业存在评级虚高的问题，继而相关区域整体的信用评级质量都会受到影响。2014 年发生的“超日债”违约事件对评级市场来说是一场震惊行业的地震，此后开始许多评级较高的债券开始违约，出现问题，其中不乏大量国有企业，“刚性兑付”的预期落空；2018 年的大公国际“停评”事件更是暴露出我国信用评级机构存在管理无序、从业人员不符合资质、利益冲突严重等问题，这些都导致了我国信用评级机构评级的低质量。本文选用 2012-2022 年发行的债券作为研究样本，以 2008-2022 年社会综合调查的与信任有关的问卷调查结果来描述地区的社会信任水平，中债市场隐含评级与评级机构评级之间的评价分歧来定义债券评级质量，使用普通最小二乘法进行基准回归，探究社会信任与债券评级质量之间的相关关系以及是通过何种途径进行此关系的。本文研究发现，社会信任对债券评级质量有显著的负影响，表现为评级质量的膨胀，为了检验稳健性，本文使用了 2007 年社会综合调查对陌生人信任的调查问题作为社会信任的代理变量、删除了样本数据中占比较大的省份的影响、使用了地区无偿献血指标作为社会信任的工具变量，回归结果仍然稳健。机制检验通过了主观偏信为主要影响机制，具体表现为评级机构会对社会信任水平较高的区域形成区域内企业违约概率较低的预期，使得评级机构高估企业经营状况，低估企业违约可能，最终导致评级水平的膨胀，评级质量的降低。主观偏信产生自债券评级分析机构与发债的公司或企业长时间的博弈产生的分析惯性。本文进一步检验了主观偏信存在情况下的利益冲突、道德风险、声誉效应是否为社会信任影响债券评级质量的机制。检验发现，在评级机构已经形成主观偏信的情境下，利益冲突机制表现为评级机构其出于增扩当地的市场份额的目的，往往采取取悦客户、膨胀评级的行为，债券评级质量由此降低；本文假设的道德风险机制表现为评级机构采取懒惰行为，自身较少的进行信息的加工与甄别，并过度引用与依赖企业的公开信息，企业美化过的公开信息与真实信息的偏离可能会使评级进一步膨胀扩大，但实证研究并未支持该假设；声誉效应机制表现为评级机构存在主动追求良好声誉的逐利行为与监管压力下不得不维持声誉的行为，以及拥有良好声誉的的企业往往具有较高的内部控制质量，有效的内部控制提高了企业的公开信息披露质量，这会促使评级机构严谨评级，降低社会信任对债券评级质量的膨胀效果，从而纠正评级质量。
英文摘要	In recent years, China's credit rating market has been experiencing significant fluctuations, and the impact of a large number of adverse events has made the already uncertain market more turbulent. Under this influence, the effectiveness and reputation of credit rating agencies are being questioned by investors. At the same time, due to the problems of untimely, inaccurate, and low quality of credit rating agencies in rating market entities, the timeliness and accuracy of credit rating agencies have been pushed to the forefront of the storm.Therefore, it is very important to study what factors affect the quality of credit ratings and how they affect it in the current complex and changing economic context. Since 2022, a large number of private enterprises in Shandong have gone bankrupt, showing characteristics such as high credit rating, concentration, and chain operation. The phenomenon of "Shandong Guarantee Circle" has become increasingly severe. There is a phenomenon of mutual protection and mutual protection among enterprises, especially in the context of the uncertain economic environment in China after the epidemic. If a single enterprise has a significant funding gap or even a broken funding chain, the credit status of all mutual protection and mutual protection enterprises in the relevant region will be seriously affected; In addition, if enterprises do not fully disclose external guarantee information, the corresponding credit risk of the enterprise cannot be fully presented through the credit rating system, which leads to the problem of inflated ratings for the enterprise, and subsequently affects the overall credit rating quality of the relevant regions. The "super day bond" default event that occurred in 2014 was a shocking earthquake for the rating market. Since then, many higher rated bonds have started to default, causing problems, including many state-owned enterprises, whose expectations of "rigid redemption" have been dashed; The 2018 Dagong International rating suspension incident further exposed the problems of disorderly management, unqualified employees, and serious conflicts of interest in China's credit rating agencies, all of which led to the low quality of credit ratings. This article selects bonds issued from 2012 to 2022 as research samples, and uses the results of a trust related questionnaire survey from the 2008 to 2022 social comprehensive survey to describe the level of social trust in the region. The evaluation divergence between implicit ratings in the Chinese bond market and ratings from rating agencies is used to define the quality of bond ratings. The ordinary least squares method is used for benchmark regression to explore the correlation between social trust and bond rating quality, as well as the channels through which this relationship is established. This study found that social trust has a significant negative impact on the quality of bond ratings, manifested by the inflation of rating quality. To test robustness, this study used the 2007 CGSS survey on stranger trust as a proxy variable for social trust, removed the influence of provinces with a larger proportion in the sample data, and used regional voluntary blood donation indicators as instrumental variables for social trust. The regression results remained robust. The mechanism test has passed the subjective bias as the main influencing mechanism, which is manifested in the expectation that rating agencies will form a lower probability of corporate default in regions with high social trust, leading to an overestimation of the company's operating conditions and an underestimation of the possibility of corporate default, ultimately leading to an expansion of rating levels and a decrease in rating quality. Subjective bias arises from the analytical inertia of long-term games between bond rating analysis agencies and issuing companies or enterprises. This article further examines whether conflicts of interest, moral hazard, and reputation effects in the presence of subjective bias are mechanisms by which social trust affects the quality of bond ratings. The test found that in the context of subjective bias formed by rating agencies, the conflict of interest mechanism is manifested as the fact that rating agencies often adopt behavior of pleasing customers and inflating ratings for the purpose of expanding local market share, resulting in a decrease in the quality of bond ratings; The moral hazard mechanism assumed in this article is characterized by rating agencies engaging in lazy behavior, engaging in less information processing and screening, and excessively citing and relying on publicly available information from companies. The deviation between publicly available and truthful information that companies have beautified may further inflate and expand ratings, but empirical research does not support this hypothesis; The reputation effect mechanism is manifested by the profit seeking behavior of rating agencies actively pursuing good reputation and the behavior of having to maintain reputation under regulatory pressure, as well as the fact that enterprises with good reputation often have higher internal control quality. Effective internal control improves the quality of public information disclosure of enterprises, which will encourage rating agencies to rigorously rating and reduce the inflationary effect of social trust on bond rating quality, thereby correcting rating quality.
学位类型	硕士
答辩日期	2024-05-26
学位授予地点	甘肃省兰州市
语种	中文
论文总页数	57
参考文献总数	60
馆藏号	0005802
保密级别	公开
中图分类号	F83/615
文献类型	学位论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/36612
专题	金融学院
推荐引用方式 GB/T 7714	孙旭. 社会信任对债券评级质量的影响研究[D]. 甘肃省兰州市. 兰州财经大学,2024.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
2021000005073.pdf（1101KB）	学位论文		开放获取	CC BY-NC-SA	浏览下载