中国政府支出对居民消费影响的理论和实证研究

发布时间：2018-03-04 21:39

本文选题：政府分类支出　切入点：居民消费　出处：《湖南师范大学》2013年硕士论文　论文类型：学位论文

【摘要】：改革开放以来,尤其近十年中国经济发生了平均10%的增长,也伴随着居民消费占国民生产总值的比重在不断下降,各省区市的经济也表现出了快速增长,区域差异也相继出现,在地里空间上表现出了相关性或差异性。2008年受全球经济危机的影响,过去在中国经济拉动中发挥重要作用的出口和民间投资已经明显锐减,消费需求增长乏力。同时,2013年各地区的环境恶化,不仅影响居民生活水平的提高,而且考验着我国政府支出对全局的考量。扩大内需,促进居民消费来拉动经济增长是主力,近年来我国政府支出也不断在增加,在基本建设、民生工程和行政管理支出上加大了投入力度,这样就可以改善投资环境、消费环境和生活环境,使得私人消费和居民消费增加,促进经济增长。本文试图通过对政府财政支出发展现状做深入分析,采用理论和实证研究分析中国政府支出对居民消费是怎样影响的,这对我国财政政策具有很好的意义,同时也对刺激我国居民消费和提高居民生活质量具有实际意义。本文利用非线性有效消费函数建立消费效用最优模型,通过理论模型的分析得到政府的三类支出与代表性消费者的关系参数,将会对居民消费产生一定的影响。三类支出参数的动态变化会对消费产生抑制或促进作用,以及这三个参数的变化也表明了如何才能使政府支出规模达到一种最优状态。同时在实证分析上,采用对数模型和SVAR模型,加入产业结构和对外负债的因素来分析对消费的动态效应。首先,经验分析得到政府基本建设支出对居民消费有一定挤出效应；民生支出对居民消费有促进作用,通过脉冲效应分析,在短期内民生支出规模的效应不太明显；行政经费支出对居民消费在一定时期里有挤出作用；产业结构调整升级对消费有一定的正效应；还有对外负债在一定时期中对居民消费有一定的负效应。接着采用了空间面板数据来自31个省份的人均数据,样本数据时间是从1997年到2010年,空间滞后模型(SAR模型)和空间误差模型(SEM模型)都采用了极大似然法估计,都分别用了地区固定效应、时间固定效应和地区时间效应估计地理加权矩阵的空间面板数据模型。SAR模型中的空间滞后项、解释变量人均收入、基本建设支出、民生支出和行政管理支出系数都显著,表明在人均收入和政府分类支出对居民消费的影响下,居民消费支出存在着显著的空间相关性,空间滞后项系数为负。SEM模型中的空间误差项、解释变量人均收入、基本建设支出、民生支出和行政管理支出系数都显著,表明居民消费支出不仅受到了人均收入和政府分类支出的影响,而且受到相邻省份的解释变量人均收入、基本建设支出、民生支出和行政管理支出的影响和空间相邻的不可观测因素的影响。
[Abstract]:Since the beginning of reform and opening up, China's economy has grown by an average of 10% in the past decade, and along with the decline in the proportion of residents' consumption to the gross national product, the economy of various provinces, districts, and cities has also shown rapid growth, and regional differences have also appeared one after another. In 2008, under the influence of the global economic crisis, exports and private investment, which used to play an important role in stimulating the Chinese economy, have been significantly reduced. At the same time, in 2013, the deterioration of the environment in various regions not only affected the improvement of residents' living standards, but also tested the overall consideration of our government expenditure. Expanding domestic demand and promoting residents' consumption to stimulate economic growth is the main force. In recent years, our government expenditure has also been continuously increasing. We have increased investment in capital construction, people's livelihood projects, and administrative management spending, so that we can improve the investment environment, the consumption environment and the living environment. This paper attempts to make a deep analysis of the current situation of the development of government fiscal expenditure, and use theoretical and empirical research to analyze how Chinese government expenditure affects residents' consumption. This is of great significance to our country's fiscal policy, but also to stimulating our country's residents' consumption and improving the residents' quality of life. In this paper, the nonlinear efficient consumption function is used to establish the optimal model of consumption utility. Through the analysis of the theoretical model, the relationship parameters between the three types of government expenditure and the representative consumers are obtained. The dynamic changes of three kinds of expenditure parameters will inhibit or promote consumption. And the change of these three parameters also shows how to make the scale of government expenditure reach an optimal state. At the same time, the logarithmic model and SVAR model are used in the empirical analysis. First of all, the empirical analysis shows that government capital construction expenditure has a certain crowding out effect on residents' consumption; people's livelihood expenditure can promote residents' consumption. Through the pulse effect analysis, the scale effect of the people's livelihood expenditure is not obvious in the short term, the administrative expenditure has the extrusion function to the resident consumption in a certain period, the industrial structure adjustment and upgrading has the certain positive effect to the consumption; There are also external liabilities that have a certain negative effect on residents' consumption in a certain period of time. Then we use spatial panel data from 31 provinces' data per capita. The sample data period is from 1997 to 2010. The spatial lag model (SAR model) and the spatial error model (SEM model) are both estimated by maximum likelihood method. Time fixed effect and region time effect estimate Spatial Panel data Model of Geographic weighting Matrix. Spatial lag in SAR model, explain variable per capita income, capital expenditure, people's livelihood expenditure and administrative expenditure coefficient are significant. It shows that under the influence of per capita income and government classified expenditure on residents' consumption, there is a significant spatial correlation between residents' consumption expenditure, and the coefficient of spatial lag is negative. SEM model is the spatial error term, which explains the variable per capita income. Capital construction expenditure, people's livelihood expenditure and administrative expenditure coefficient are significant, indicating that residents' consumption expenditure is not only affected by per capita income and government classified expenditure, but also by the explanatory variables of neighboring provinces, such as per capita income, capital construction expenditure, etc. The influence of people's livelihood expenditure and administrative expenditure and the unobservable factors adjacent to each other.
【学位授予单位】：湖南师范大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F126.1;F812.45

【参考文献】