地方政府债务测度指标体系及风险预警研究

发布时间：2018-11-22 07:29

【摘要】：地方政府债务对促进地区经济的发展具有举足轻重的作用,但随着债务规模的扩大,地方财政的风险隐患日益凸显,已成为威胁地方经济安全和社会稳定的重要因素。正确地认识地方政府债务与地方经济发展的关系,建立规范的地方政府债务管理制度和科学有效地债务风险预警体系,基于地方经济发展水平建立科学的债务承载阈值,对于完善地方政府债务管理,保证经济安全和社会稳定,促进社会经济健康持续发展具有重大意义近些年,地方债务迅速增长,因为偿债能力的欠缺、资金应用不恰当等原因,一些地方政府不得不盲目加大债务量,从而形成债务恶性循环。不少地区政府债务赤字已经严重影响到地方经济的发展。那么,怎样确定恰当的债务水平才能够促进经济的持续快速发展?建立怎样的测度模型来确定两者之间的量化关系从而达到相互促进、相辅相成的效果?这都是刻不容缓需要解决的问题。本研究以地方债务为研究对象,重点从地方政府偿债能力、资金再生能力、债务存量、债务增速、债务资金投向以及地方GDP发展等方向入手研究,建立了全口径指标评价体系,利用定性分析和定量分析相结合的方法来分析债务风险状况。在指标体系中引入资金投向这一概念,对地方政府债务资金投向合理性进行研究,弥补了以往研究的不足。将指标体系进行聚类分析,通过聚类分析剔除重复指标,得到有效的指标体系,达到指标独立性和最少性的原则。然后通过因子分析得出地方债务风险的三个主要因子,对以往的地方政府债务状况进行了系统的评价。同时,本研究将BP神经网络引入研究中,以因子分析的结果为训练集,对地方政府债务进行预警。BP神经网络模型是多输入多输出的逆向传播模型,适合对复杂影响因素对象进行分析和预估,BP神经网络输入集分为训练集和测试集,本研究中以因子分析结果为训练集,以最新一期数据为测试集。通过预警模型的处理,进一步判断地方政府债务风险处于的风险区间。文章最后提出了有针对性的政策建议。最后,本研究以山东省地方政府在债务状况为例,利用研究成果进行了预警及其风险能力分析。
[Abstract]:Local government debt plays an important role in promoting the development of regional economy, but with the expansion of debt scale, the hidden risks of local finance become increasingly prominent, which has become an important factor threatening local economic security and social stability. To correctly understand the relationship between local government debt and local economic development, to establish a standardized local government debt management system and a scientific and effective debt risk warning system, and to establish a scientific threshold for debt bearing based on the level of local economic development. It is of great significance to improve local government debt management, ensure economic security and social stability, and promote the healthy and sustainable development of social economy. In recent years, local debt has grown rapidly because of the lack of solvency and improper use of funds. Some local governments have to blindly increase debt, thus forming a vicious cycle of debt. Many regional government debt deficits have seriously affected the development of the local economy. So, how to determine the appropriate level of debt to promote sustained and rapid economic growth? What kind of measurement model is established to determine the quantitative relationship between the two so as to achieve the effect of mutual promotion and complementarity? These are urgent problems that need to be solved. This research takes the local debt as the research object, focuses on the local government solvency, capital regeneration ability, debt stock, debt growth rate, debt capital investment and the development of local GDP, and establishes a full-caliber index evaluation system. Qualitative analysis and quantitative analysis are used to analyze debt risk. The concept of capital investment is introduced into the index system, and the rationality of local government debt investment is studied, which makes up for the deficiency of previous research. The index system is analyzed by cluster analysis and the repetitive index is eliminated by clustering analysis. The effective index system is obtained and the principle of independence and minimum of index is achieved. Then, through factor analysis, three main factors of local debt risk are obtained, and the former local government debt situation is systematically evaluated. At the same time, the BP neural network is introduced into the research, and the result of factor analysis is used as the training set to forewarn the local government debt. The BP neural network model is a multi-input and multi-output reverse propagation model. The input set of BP neural network is divided into training set and test set. The result of factor analysis is the training set and the latest data is the test set. Through the treatment of early warning model, the risk interval of local government debt risk is further judged. At the end of the article, some policy suggestions are put forward. Finally, this study takes the debt status of local governments in Shandong Province as an example and makes use of the research results to carry out early warning and risk analysis.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F812.5

【相似文献】