【作者】 周青；

【导师】 杨秀苔；

【作者基本信息】 重庆大学 ， 技术经济及管理， 2011， 博士

【摘要】 研究地方政府投融资平台风险管理与风险度量对中国经济具有重要的理论与现实意义。在地方政府投融资平台投资带动下,地方政府有效缓解了财力不足、建设资金短缺的发展瓶颈,并且带动了企业投资,活跃了市场,有力地推动了各地城市化和基础设施建设。为保增长、扩内需、调结构、促民生,以及迅速扭转经济下滑趋势发挥了重要的作用。然而,地方政府投融资平台在拉动中国经济迅速回升的同时,也埋下了很多隐患。地方政府投融资平台建立的初衷是借政府拥有的优质资源建立而成,是政府实现产业调整、解决发展难题、推动经济可持续发展的重要举措,投融资平台采用市场化的运作手段进行内部经营管理,同时得到政府的大力支持和政策指导。一方面,其主要职能受到政府规范,也得到政府赋予的大量优质资源和经营性资产,在未来经营发展方面得到政府的大力支持;另一方面,其主要的投资决策遵循地方产业政策集中于基础设施建设、重大项目配套和先导产业投资等,在其职能定位上也将充分显示政府的宏观调控意图。本文首先阐述了风险及其管理的基本理论,对国内外地方政府投融资平台的发展模式进行了综述,在国外城市基础设施建设投融资模式分析方面,主要以英美法三国为例,同时从融资风险和投资风险两个方面对地方政府投融资平台的风险进行了分析。从经济学的角度分析了地方政府投融资平台过度投融资行为的动因,同时也从其他几个方面分析了投融资平台的风险机理,阐述了投融资平台的风险传导路径。然后研究了地方投融资平台风险影响因素,从行业界定、地方投融资平台行业风险因素两方面探讨地方投融资平台风险特征,从宏观、中观、微观三方面分析地方投融资平台风险影响因素。建立了地方投融资平台风险预警模型。这部分内容包括建立地方投融资平台风险预警程序,分析风险预警研究方法。依据地方投融资平台风险影响因素分析,从宏观、中观、微观三个方面提取指标构建地方投融资平台风险预警指标体系,选取模糊综合评价方法和层次分析法构建地方投融资平台风险预警模型,并对该模型进行了应用分析。建立了地方政府投融资平台投融资风险度量模型。投融资平台风险度量模型分融资风险模型和投资风险模型,对于融资风险,采用传统的CreditMetric模型并对其进行改进使之更适合我国地方政府投融资平台的实际情况对融资做一维风险度量。对融资多维风险,考虑到多种融资间的相关关系,应用更为实用可行的Copula理论来进行度量。对于投资风险,采用随机波动模型与极值理论结合,提出了基于POT-SV-t的动态VaR模型来进行一维投资风险度量,在此基础上采用Copula理论考察多维变量间的相关关系来进行多维风险度量。在地方政府投融资平台风险管理与度量的实证研究方面,分析了重庆地方投融资平台现状,并结合前面章节的地方投融资平台风险预警研究方法及模型,以重庆地产集团为研究对象,按照地方投融资平台风险预警模型,对其相关数据采用模糊层级分析法进行风险预警系统的实证分析,并对重庆交旅集团的融资债券采用CreditMetric模型进行一维风险度量的实证计算。最后得出研究结论,并提出相关的政策建议与未来展望。本文特色在于紧密围绕我国地方政府投融资平台发展的现实,研究地方政府投融资平台的风险管理与度量,具有强烈的现实针对性,并在理论方法上具有较好的深度与创新。研究内容上深入,研究方法严谨。本论文的创新体现在如下几个方面:第一,地方政府投融资平台的风险形成机理方面,从经济学的角度剖析投融资平台过度投融资的动因。利用财政机会主义分析方法与模型对过度投融资行为的动因进行了分析;从双向羊群效应的相互作用、影响和强化出发,分析我国地方政府投融资平台数量和融资规模急剧增长以及风险迅速积聚的原因;从地方财政收支失衡、公共风险承担、平台运作等方面阐述了地方政府投融资的风险机理;从地方政府债务风险、中央或国家财政风险、投融资平台的先天性缺陷和运作不规范、宏观经济运行、平台融资主渠道单一及金融系统的脆弱性等方面分析了投融资平台的风险传导路径。第二,地方投融资平台风险预警研究方面,从宏观、中观、微观三方面分析地方投融资平台风险影响因素,宏观方面主要从宏观经济环境、宏观经济政策、法律因素三方面寻找影响因素;中观方面主要从区域经济环境与地方政府的信用等两大方面进行探索;微观方面主要从投融资平台内部财务因素与非财务因素两方面寻找分析影响因素;并据此确定地方投融资平台风险预警目标、风险预警程序、提取指标构建地方投融资平台风险预警指标体系,利用模糊综合评价方法和层次分析法构建了地方投融资平台风险预警模型。第三,在地方政府投融资平台风险度量模型方面,从融资风险和投资风险两个方面进行了研究。首先深入分析已有投融资风险管理模型与方法的适应性与局限性,提出了基于Credit Metrics模型的投融资平台一维信用融资风险度量方法;通过分析地方政府投融资平台的融资结构与融资模式的多元性,通过构建信用曲线、选择合适的Copula函数及计算联合违约概率分布,提出并建立基于Copula函数的投融资平台多笔贷款的信用风险测度方法。在投资风险度量方面,从动态角度考虑VaR时间序列的特征,应用SV-t模型与极值理论相结合拟合金融资产收益的尾部特征,建立了一种新的投资风险测度模型——基于POT-SV-t的动态VaR模型,以此进行一维投资风险度量,并结合Copula理论来进行多维投资风险度量。另外,实证研究方面,全面分析了重庆市级及区县级的投融资平台风险及影响因素并进行风险的相关定量测度。利用重庆地产集团的相关数据,对风险预警系统模型进行实证,计算分析此投融资平台的宏观、中观、微观的风险指标预警阈值,设定各指标权重,计算出总体标准风险分数值,并参照预警信号表输出预警信号和风险状况评价。利用重庆交旅集团的实际数据,通过设定信用等级转移矩阵、估算未来不同信用等级下的贷款远期价值、推导贷款价值变动的远期分布,对此平台进行一维信用风险度量计算。更多还原

【Abstract】 Study on the risk management and measurement of investment and financing platform of local government (IFPLG) has an importantly theoretical and practical significance for Chinese economy. Driven by investment of IFPLG, local government alleviate the development bottlenecks of resources and development fund shortage effectively. Moreover, the IFPLG has vivified business investment and the market and given a strong impetus to the urbanization and infrastructure construction, playing an important role in maintaining growth, expanding domestic demand, adjusting structure, promoting the people’s livelihood and reversing the economic decline quickly. The IFPLG has made a significant contribution to the rapid rise in China’s economic. However, it has planted a lot of hidden dangers. The original intention for establishing the IFPLG is relying on the government-owned high-quality resources, and it is an important way for the government to solve the problem of industrial adjustment and economic development, promotes sustainable development. The IFPLG uses market-oriented operation for internal management and receives supports and policy guidances form government. On the one hand, its main functions are regulated by government, and the government gives a lot of high quality resources and management assets to support its future development. On the other hand, the main investment decisions of the IFPLG are in accordance with the local industrial policy which is focused on infrastructure construction, major projects and forerunner industrial investment. Supported by the government, its functional location will demonstrate the government’s macro-control intention.This study firstly describes the basic theory of risk as well as its management and development models of domestic and foreign IFPLG. In the analysis of foreign urban infrastructure construction’s investment and finance model, it takes three countries-England, America, France-as examples to analyse the IFPLG’s risks from the two aspects of financial risks and investment risk. It analyses the drivers of the IFPLG to invest and finance excessively from an economic view. Moreover, it researches risk mechanism of investment and finance platform from other aspects and describes transmission pathway of investment and financial platform risk. And then, it studies influence factors of the IFPLG risk. By probing into the IFPLG risk’s characteristics from the two aspects of industry definition and risk, analysing influence factors of the IFPLG risk based on the macro, meso and micro aspects, this paper establishes risk early warning model of the IFPLG. This section includes the procedure of establishing risk early warning model and methods of analysing of risk early warning. Based on local analysis of the IFPLG risk factors, it collects indicators from aspects of macro, meso and micro to establish the risk warning indicator system. Choosing the fuzzy comprehensive evaluation method and the AHP to build the IFPLG risk warning model and made application analysis of the model, then stablished risk measurement models. The IFPLG risk measurement models contains investment risk measurement model and finance risk measurement model. For the financing risk, using the traditional model of CreditMetrics and improving it to make it more suitable for the actual situation of IFPLG in China and make one-dimensional risk measures. Otherwise, for multidimensional risk of financing, taking the relationship of a variety of financing into account , this paper applies Copula theory which is more practical and feasible. For investment risk, stochastic volatility models combined with the extreme value theory was used and dynamic VaR model based on POT-SV-t was proposed for one-dimensional risk measurement. On the basis, Copula theory was used for multidimensional risk measurement by analysing relationship between the variables. In empirical research of the IFPLG risk management and measurement, this paper analysed the status of IFPLG in Chongqing, combined with the previous chapters, take Chongqing Real Estate Group as the research object, in accordance with risk early warning model of local investment and financing platform, took its data to made an empirical analysis on early warning systems by fuzzy comprehensive evaluation method. Besides, this paper also calculated one-dimensional empiricaliy to Chongqing Travel Group’s financing bonds with CreditMetric model. Finally, based on the statistical data it made the research conclusion, and provided some useful advice to relevant policy and future prospects.This study is featured by researching the IFPLG’s risk management and measurement with focus on the development reality of the IFPLG in China. It has a strong practical relevance and a good depth and innovation on theoretical methods. It researches in a in-depth content and strict method. The innovative points of this project can be summarized the following four aspects: First, for the aspect of the IFPLG risk formation mechanism, it analysed the drivers of the IFPLG to invest and finance excessively from an economic view. Using fiscal opportunism analysis method and model, it analysed the motives for excessive investment and financing behaviour. From the interaction, influence and strengthening of bidirectional herding effect, the paper analysed reasons that why the IFPLG in China has a sharp growth on quantity and financing scale and why risk accumulated rapidly. Risk mechanism of local government investment and financing in China was described from views of the disequilibrium of local revenues and expenditures, public risk-taking and platform operation and so on. From local government debt risk, central or state financial risk, congenital defect non-standard operation of investment and financing platform and the macro-economy, single main financing channel single, vulnerability of inancial systems and other aspects to analyse transmission pathway of investment and financial platform risk. Second, to the IFPLG risk early warning research, the paper analysed influence factors of IFPLG risk from the macro, meso, microscopic three aspects. Macroscopic aspect, it collected influencing factors from the macroeconomic environment, macro economic policies mainly, and meso mainly from the two aspects of regional economic environment and the local government credit, for the micro aspect, however, it mainly focused on internal financial factors and non-financial factors of platform. And according this, it determined the risk early-warning target, risk alarming program, collecting indicators to establish the risk warning indicator system. Choosed the fuzzy comprehensive evaluation method and the AHP to build the IFPLG risk warning model. Third, to the IFPLG risk measurement model, the two aspects of financing risk and investment risk were studied. The paper analysed the the appropriateness and limitations of existing investment risk management models and methods firstly and put forward Metrics model-based one-dimensional Credit financing risk measurement model of financing platform. Through the analysis of diversity of the IFPLG’s financing structure and financing mode, by constructing credit curve, choosing appropriate Copula function and calculating joint default probability distribution, this paper put forward and developed the credit risk measurement methods and models of several loans on investment and financing platform based on Copula function. In the investment risk metric aspects, considering VaR time series feature from a dynamic view, combined SV - t model and extreme value theory to fit financial asset returns tail characteristics, established a new investment risk measuring model -dynamic VaR model based on POT-SV-t for one-dimensional risk measurement. And and combined with Copula theory to multidimensional investment risk measurement. To empirical research, this paper analysed the investment and financing platform risk and affecting factors in Chongqing and made risks related quantitative measurement. By using relevant data of Chongqing Real Estate Group, the paper did an empirical analysis to risk early-warning system model, calculated and analysed the macro, meso, micro risk index warning threshold of this investment and financing platform, set weight for each index, calculated the overall standard risk score value, and then output signal and risk assessment by consulting with the early warning signal table. Using the actual data from Chongqing Tour Group, by setting credit rating transfer matrix, and estimating forward price of loans under different credit rating and deducting long-term distribution of loan value changing, one-dimensional credit risk measurement of this platform is computed.更多还原