2024届硕士专业学位研究生学位论文学校代码：10269
学号：51244407031
我国城投债信用风险测度研究--基于
Logistic-KMV 模型
院系：经济与管理学院
专业学位类别：金融硕士
专业学位领域：金融
论文指导教师：张玉鹏副教授
论文作者：王嫚
2023年5月
STER DISSERTATION 2024UNIVERSITY CODE：10269
STUDENT NO：51244407031
EAST CHINA NORMAL UNIVERSITY
Research on Credit Risk Measurement of
Urban Investment Bonds in China --Based
on Logistic KMV Model
College：Faculty of Economics and Management
Major：Master of Finance
Specialty：Finance
Advisor：Zhang Yupeng Associate Professor
Candidate：Wang Man
May of 2024
王嫚硕士学位论文答辩委员会成员名单
姓名职称单位备注
龙翠红教授华东师范大学主席
汪莉副教授华东师范大学
鲁文龙高级经济师兴业银行
摘要
2023年，随着下半年“一揽子化债方案”逐步落地，城投债市场呈现了一
定的边际走强态势，但当前市场仍然面临着多重挑战和不确定性。宏观经济环境
的不稳定性，以及经济增速的放缓和财政收入的下降等因素，给地方政府带来了
更大的偿债压力。在这种情况下，一旦发生债务违约事件，将不仅引发金融市场
的波动，还可能导致信用风险的传导，对地方政府和金融市场造成严重的影响和
挑战。因此，对我国城投债信用风险进行准确的测度和评估具有重要的意义。
首先，本文详细梳理了我国城投债产生的背景和发展历程，结合当前城投
债市场的现状和特点进行了深入分析。其次，通过对我国城投债信用风险主要影
响因素的识别和分析，本文建立了相应的分析框架。在此基础上，我们对各种信
用风险测度模型进行了比较，最终选取了Logistic-KMV 模型作为研究的主要工
具。为了验证该模型的有效性，我们收集了截至2022年末我国271个地级市仍
有存续债券的2076个城投主体的财务数据，并综合考虑了相应地级市的宏观经
济指标和政府财政情况，以及基于修正的KMV 模型计算出的违约距离等因素，
将这些作为自变量纳入模型中进行分析。最后，基于回归结果，我们建立了一套
完善的评分体系，以便更好地评估和管理城投债信用风险，为相关决策提供科学
依据。
实证结果显示，首先，修正的KMV 模型计算出的违约距离能够在一定程度
上反映各地级市城投债的风险水平。其次，Logistic-KMV 模型的应用揭示了城
投主体的信用风险与其自身的财务状况、所处地级市的宏观环境以及地方政府的
财政状况密切相关。该模型在我国城投债信用风险的测算与预警方面表现出更好
的适用性。最后，通过评分模型对各地级市的城投债进行评分，可以更准确地直
观反映城投债的信用风险情况。基于以上结论，本文从政府、城投主体和投资者
三个层面提出了建议，旨在建立风险预警机制，预防和化解城投债的信用风险，
以确保我国城投债市场和整个资本市场的健康稳定运行。
关键词:城投债Logistic-KMV 模型风险预警信用风险
I
ABSTRACT
In 2023,with the gradual implementation of the "comprehensive debt restructuring
plan"in the latter half of the year,the local government financing vehicle (LGFV)
bond market has shown marginal signs of strengthening.However,the current market
still faces multiple challenges and uncertainties.The instability of the macroeconomic
environment,coupled with the slowdown in economic growth and the decline in fiscal
revenue,has imposed greater debt repayment pressure on local governments.In such
a scenario,any occurrence of debt default events could not only trigger financial
market volatility but also potentially propagate credit risks,posing significant
challenges and impacts on local governments and financial markets.Therefore,
accurately measuring and assessing the credit risks of LGFV bonds in China is of
paramount importance.
To begin with,this paper comprehensively reviews the background and development
trajectory of LGFV bonds in China and conducts an in-depth analysis of the current
status and characteristics of the LGFV bond market.Subsequently,by identifying and
analyzing the main influencing factors of credit risks associated with LGFV bonds in
China,we establish a corresponding analytical framework.Based on this framework,
we compare various credit risk measurement models and ultimately select the
Logistic-KMV model as the primary tool for our study.To validate the effectiveness
of this model,we gather financial data from 2076LGFV entities in 271
prefecture-level cities across China that still have outstanding bonds by the end of
2022.We also consider macroeconomic indicators and government fiscal conditions
at the corresponding prefecture-level cities,as well as default distances calculated
based on the modified KMV model,and incorporate them as independent variables
for analysis.Finally,based on the regression results,we develop a comprehensive
scoring system to better assess and manage the credit risks of LGFV bonds and
provide scientific insights for relevant decision-making processes.
Empirical results indicate that,firstly,the default distances calculated by the modified
KMV model can partially reflect the risk levels of LGFV bonds in various
prefecture-level cities.Secondly,the application of the Logistic-KMV model reveals
that the credit risks of LGFV entities are closely related to their own financial
conditions,the macroeconomic environment of the prefecture-level cities they operate
in,and the fiscal conditions of local governments.This model demonstrates better
applicability in measuring and warning against credit risks associated with LGFV
bonds in China.Lastly,by using the scoring model to assess LGFV bonds in different
prefecture-level cities,we can more accurately and intuitively reflect their credit risk
situations.Based on these conclusions,this paper proposes recommendations from the
perspectives of governments,LGFV entities,and investors,aiming to establish risk
warning mechanisms to prevent and mitigate the credit risks of LGFV bonds,thereby
II
I
I
I