【作者】 王宪勇；

【导师】 杜两省；

【作者基本信息】 东北财经大学 ， 国民经济学， 2009， 博士

【摘要】 改革开放以来的三十年中,中国保持了近10%的年均高速经济增长率,同时也经历了宏观经济的数个大起大落。中国宏观经济波动有何特征?中国经济波动的原因何在?本文的研究试图回答这两个问题。本文的研究在动态随机一般均衡的框架下进行。自20世纪80年代诞生以来,该框架经过不断地发展和完善,在经济波动的研究中已取得很大成功,目前已成为研究发达国家经济波动问题的主流方法。为了站在巨人的肩膀上研究中国的经济波动,本文的第2章概述了以该方法为框架所发展出经济波动理论的发展脉络及最新进展。概括中国经济波动的特征是解释中国经济波动成因的前提。本文的第3章使用HP滤波、BP滤波对中国主要宏观经济变量时间序列进行分解,获得序列的周期成分,然后计算这些变量的二阶矩特征,以此来反映经济波动的易变性、粘持性和共动等中国经济波动的静态特征。具体地说,这些静态特征主要包括：(1)消费的波动小于产出的波动,是产出波动的50%,这表明了中国居民平滑消费的特点。在消费的构成中,城镇消费和农村消费的波动相同,都为0.04。(2)投资的波动大于产出的波动,在投资的构成中,存货投资的波动最大,是产出波动的近7倍,房地产投资的波动也很大,是产出波动的近3倍。(3)进出口的波动大于产出的波动,进口的波动和出口的波动以及净出口的波动都大于产出的波动,进口和出口的波动约为GDP波动的2倍,净出口的波动是产出波动的近5倍。(4)产出具有一定的粘持性,相邻两期的相关系数为0.58。(5)多数经济变量与产出具有高度共动性,资本、消费、投资、进出口等经济变量具有强顺周期的特点,它们与产出的同期相关性都在0.7以上。同时,本文也使用SVAR方法分解出技术冲击、货币冲击和财政冲击对中国经济的动态影响。这些动态特征主要包括：(1)三种冲击引起对产出波动的相对重要性。三种冲击同时发生时,技术冲击对产出波动的贡献在观察期内一直占有绝对优势,其贡献一直在80%以上。技术冲击对产出的影响在初期便已达到83.5%,并且随着时间的推移而逐渐增加,在第25期达到了982%。货币供给冲击对产出的影响最小,货币扩张对产出的变动是中性的。政府支出冲击在初期对产出的贡献率为16.4%,随后逐渐减少,在第25期已不足1%。政府支出冲击在需求冲击中对产出的影响最大,其贡献率占需求冲击总贡献率的90%以上。(2)三种冲击引起对通胀波动的相对重要性。除了第1期外,从第2期开始,技术冲击对通胀的贡献为65.3%,虽然在第三期略有下降,但是以后各期基本稳定在66%的水平上。需求冲击对产出变动的贡献率只有10%左右,需求冲击对通胀的贡献则大得多,贡献率在35%左右。其中,货币冲击的贡献率在20%左右,而财政冲击的贡献率则在15%左右。本文的第4章至第6章分别考察了五个DSGE模型对中国经济波动的解释力,其中前两个模型为基准DSGE模型,模型经济中没有冲击放大机制,而后三个模型考察不同的冲击放大机制在中国经济波动的作用。基准DSGE模型包括基本实际商业周期模型(RBC model),基本货币商业周期模型(MBC model),这两个模型不带有任何具有市场不完全性的因素；后三个DSGE模型是对基准DSGE模型的扩展,在模型中加入三种冲击的放大机制——名义粘性、实际粘性和金融加速器。结果发现,基本RBC模型和基本MBC模型只能在复制现实经济波动的部分静态特征,二者相比,MBC模型的解释力略胜一筹。但是这两个模型在本文中却又具有重要的意义,它们可以作为我们观察后续三个模型的基准模型,因此来确定不同放大机制的具体作用。在MBC模型中加入实际粘性和名义粘性后,与基本MBC模型相比,模型的解释力并无明显改进。因此,从我们的研究结果看,实际粘性和名义粘性并不是中国经济波动的一个重要的放大机制。而在MBC中加入金融加速器后,模型较好地复制了中国经济波动的动态特征,因此金融加速器是放大中国经济波动的重要机制。金融加速器能够成为重要的放大机制的关键在于,在信息不对称的情况下,银行根据可观察到企业净资产规模来来判断企业的还款能力和违约风险,因此企业的外部融资升水和其净资产规模是负相关的。其中外部融资升水为企业外部融资和内部融资成本之差,而净资产规模为个体流动性资产以及非流动性资产之和减去债务得到的差值。当经济受到有利冲击,例如正向的技术冲击、扩张性的货币冲击和财政冲击等,资产价格上升,所以企业的净资产增加,外部融资升水降低,企业贷款增加。当经济受到不利冲击,例如利率增加冲击,资产价格下降,企业净资产下降,外部融资升水增加,企业贷款减少。因此,随着经济受到不同的冲击,企业净资产不断变化,进而出现了信贷波动。信贷波动通过两种渠道对宏观经济产生影响：一种为投资渠道。贷款被企业直接用于投资购置新资本来扩张生产,从而形成信贷增加与净资产增加的相互加强机制,由于能力不断增加,形成新一轮经济周期的繁荣期。另一种为资金配置渠道。大量的银行信贷被注入投机市场,在银行信贷总额一定的情况下,市场性企业出现融资困难,资金未得到有效配置,经济结构发生扭曲。本文的模型中主要考虑的是信贷波动对宏观经济影响的投资渠道。企业的净资产可以通过直接影响投资成本引起投资的波动,这一成本为企业外部融资所需成本,外部融资升水的变动就是投资成本的变动。基于上述分析,我们可以得到以下的政策含义：第一,财政政策、货币政策在实施时,应考虑技术冲击对经济的影响特点。技术冲击带来产出的增加是倒驼峰形变化的,这就带来了经济增长由高到低,以及由低到高的不稳定变化。这就需要在下降阶段配合积极的财政政策和货币政策,使经济具有持续的高增长。如果产出处于倒驼峰形的高点,出现经济过热,则需要提前配合紧缩性的财政政策和货币政策,避免经济的大起大落。第二,需要重视财政政策与货币政策的协调。由于财政政策与货币政策的时滞不同,而且二者对主要宏观经济变量的影响程度是不同的。所以二者协调使用,才不会出现有时二者之间的作用的发挥在一段时期内被抵消,而在另外一段时期内却又被过度放大的情形。第三,金融加速器这一传导机制的存在,为我们熨平经济周期的波动提供了方向。这一方面要求我们加强信用体系的建设,改变银行贷款的顺周期性；另一方面,发展多层次的资本市场,例如加快创业板,甚至三板、四板市场的建立,为企业开拓出多元化的融资渠道。通过这两个方面的建设,可以使企业在经济衰退期的投资需求在一定的风险水平下得到比较充分的满足。最终使企业的自主投资成为经济快速恢复的一个有力支撑。第四,总需求管理政策目前在中国的特殊重要性。技术冲击是中国经济波动的主要原因,因此最有效的方式是降低技术冲击对经济的影响。要达到此目的可以有两个途径,第一是减小技术冲击的冲击源,第二是减小技术冲击的放大机制。但是这两个途径目前在中国都不具有可操作性。减小正向的技术冲击,虽然可以增加经济增长的稳定性,但却要以牺牲经济高速增长为代价。中国计划经济向市场经济转变中凸现的一些问题,如国有企业的巨额呆账坏账,提高国有企业效率引起的大量下岗工人,社会保障账户的空转,都需要经济的高速增长来解决和消化,所谓的以增量消化存量。因此,采取降低技术冲击的方法平抑宏观经济波动,不但会放慢增长速度,而且可能引发其他的经济和社会问题,对于中国来讲是得不偿失的。中国的金融体系改革由于其巨大的复杂性,绝非一朝一夕能够完成,金融加速器效应在较长时期内依然会在中国存在。因此,在主要的冲击放大机制存在、主要冲击源不能降低的情况下,中国降低经济波动、实现经济平稳增长的可行方式还是总需求管理的政策,即财政政策和货币政策。更多还原

【Abstract】 Since reform and opening up 30 years, China maintained high-speed economic growth nearly 10% per year, but also experienced a number of macroeconomic ups and downs. What are the characteristics of China’s macro-economic fluctuations? What are the reasons of China’s economic fluctuations? This study attempts to answer these two questions.This paper is under a dynamic general equilibrium framework of a random manner,began in 1980’s,which has been great successful in the study of economic fluctuations and the mainstream of economic fluctuations approach in developed countries In order to stand on the shoulders of giants to study China’s economic fluctuations, Chapter 2 of this article provides an overview of the latest developments of the theory of economic fluctuations developed from the theoryThe premise of explaining the causes of China’s economic fluctuations is to Sum up the characteristics of China’s economic fluctuations.Chapter 3 explains China’s major macroeconomic variables for time series decomposition by the HP filter, BP filter, reflects the volatility of economic fluctuations such as visco-holders by obtaining the cycle component sequence, calculating the second-order moment of these variable characteristics. Specifically, the static characteristics include:(1) the volatility of consumption is 50% of output fluctuations,less than the volatility of output,indicating smoothing consumption of the Chinese residents. In the composition of consumption, the urban consumption and rural consumption fluctuations in the same for 0.04. (2) the investment is more volatile than the volatility of output, the inventory investment is the largest among the compositions of investment about 7 times output volatility.The real estate investment is also significant about 3 times output fluctuations. (3) the fluctuations of import and export is more volatile than the output, the fluctuations of the volatility of imports and exports, as well as net exports are greater than the volatility of output, the volatility of imports and exports is about 2-fold fluctuations in GDP, net exports fluctuations is nearly 5 times output volatility. (4) output has a hold of the stick, the two adjacent correlation coefficient of 0.58. (5) the majority of economic variables with a total output of high mobility of capital, consumption, investment, imports and exports and other economic variables have a strong cis-cycle characteristics, and the relevance of the above is 0.7 in the same period At the same time, this article also use SVAR method to break down the impact of technology and monetary shocks and financial shocks on the dynamic effects of China’s economy. The main features of these developments include:(1) the relative importance of the three types of shocks to fluctuations of output. When the three shocks is simultaneous, the contribution of technology shocks is dominant, more than 80%,to the output fluctuations during the observation period. At the beginning technology shocks impact on output has been achieved in the initial 83.5 percent, and increased gradually over time,reached 98.2% in 25 period. The impact of money supply has less effect on output, monetary expansion are neutral on output fluctuations. The impact of government expenditure on output is the greatest, the initial contribution rate is 16.4 percent, followed by a gradual reduction,in 25 period less than 1%. The impact of government expenditure in the demand shock is the greatest impact on output, the contribution of demand shocks account for 90% of the total contribution. (2) the relative importance of the three shocks on inflation fluctuations. From the two period, the contribution of technical impact on inflation is 65.3%, although there is a slight decrease in the third period, the subsequent stages basic remain stability level which is about 66%. the contribution rate of demand shocks on output in only 10%. The contribution demand shocks on inflation is much larger, around 35%.The contribution rate of currency impact is around 20%, while the the contribution rate of financial impact is about 15%.Chapter 4 to Chapter 6 of this article, respectively, review the ability of explaination to China’s economic fluctuations of the five DSGE models. the first two models are the benchmark DSGE model, which don’t impact the economy zoom mechanism, the latter hree models study the impact of different mechanism in China. Benchmark DSGE models include the basic real business cycle model (RBC model), the basic monetary model of the business cycle (MBC model), these two models do not carry any market factors; the latter three DSGE model is the expansion of benchmark DSGE model by adding three magnification mechanism-the name of viscous, the actual viscosity and financial accelerator. It was found that the basic RBC model and the basic MBC model can only copy static characteristics of real econominc fluctuations. MBC model is better than the others in explanatory power However, these two models in this article have great significance, they can serve as a model to observe the follow-up three for us, in order to determine specific effects of the different magnification mechanisms. If we add the name of the actual viscous and sticky in the the MBC model model, the explanatory power is not improved significantly compared with the basic the MBC model. Therefore,,the actual name of viscous and viscous fluctuations is not an important magnification mechanism in the Chinese economy. MBC added in the financial accelerator, can reproduce fluctuations in the economy of China’s dynamic characteristics, so the financial accelerator is an important mechanism to enlarge China’s economic fluctuations.The Financial accelerator is the key amplification mechanism, because banks can confirm repayment ability of enterprises and default risk according to their net asset size when there is asymmetric information. The external financing premium and the net asset size of enterprises is negatively correlated. Which the external financing premium equals difference of external financing and internal financing costs.when the economy gets good shocks, such as the positive technology shocks, the positive monetary and fiscal shocks, asset price increase, so the net assets of businesses increase, reduce the external financing premium, corporate loans increased. When the economy is subject to negative shocks, such as positive interest rates shocks declining asset prices, decline in the net assets of enterprises, increase the external financing premium, corporate loans declined. Therefore, as the economy are responsed to various shocks, the net assets of business is variable, and credit volatility emerged.Based on the above analysis, we can give some policy implications:First, the implementation in fiscal policy and monetary policy should take into account the impact of technology shocks on the economy. Technology shocks increase output in inverted hump-shape. This implies that the recession phase the authorities should adopt the active fiscal policy and monetary policy, the economy has sustained high growth. If the output is inverted hump-shaped high-point of economic overheating, you need to advance with the tightening of fiscal policy and monetary policy, to avoid more economic fluctuation.更多还原