货币政策、企业行为与商业周期

【作者】 刘晴辉；

【导师】 陈学彬；

【作者基本信息】 复旦大学 ， 金融学， 2010， 博士

【摘要】 基于动态随机一般均衡的货币政策分析框架已成为当代货币理论与政策基本的研究范式。但这一模型的微观基础仍处于不断探索和进展之中。这一框架存在的另一个重要问题是,主要针对发达国家或市场经济而建立的模型结构,在应用到不发达国家或市场化程度不高的国家时,模型的效力可能下降甚至不能发挥作用。基于上述研究背景,本文集中和系统地研究货币政策效应的重要微观基础——企业行为,进而在动态随机一般均衡货币政策分析框架内,探讨货币政策、企业行为与商业周期这一过程内在的动态传播机制,并试图通过区分国有和非国有两类企业,扩展基准的动态随机一般均衡货币政策模型。本文首先对货币政策、企业行为与商业周期的文献进行了综述,然后阐述了基准的动态随机一般均衡货币政策分析框架。在此基础上,分析了国有和非国有企业的投资决策行为,并将其纳入基准的动态随机一般均衡货币政策模型,实现对基准模型的第一次扩展。遵循这一从简单到复杂,逐步放松假定的基本研究思路,接着依次从企业融资、企业成本和企业进入退出的角度对国有和非国有企业的决策行为作理论和实证分析,将其依次纳入基准的动态随机一般均衡货币政策模型,进而实现对基准模型的逐次扩展,最终建立基于国有和非国有两元企业结构,包含上述四类企业行为的动态随机一般均衡货币政策模型,并利用中国数据对模型进行数值模拟。文章最后进行总结并提出了相关政策建议。基于上述研究内容和思路,本文可能的创新可概括为,结合中国实际,通过建立基于国有和非国有两元企业结构的动态随机一般均衡货币政策模型,扩展了基准的动态随机一般均衡货币政策模型,进而将预算软约束植入了主流的货币政策分析范式。因此,本文对开发和建立发展中国家或市场化程度不高国家动态随机一般均衡货币政策模型进行了初步尝试。上述创新具体表现为：(1)分别建立了含有内生企业投资和预算软约束；引入预算软约束的金融加速器；以及同时包含企业投资、融资、成本和进入退出,基于国有和非国有两元企业结构的三个动态随机一般均衡货币政策模型,并对模型进行了数值模拟；(2)基于中国企业和行业层面的面板数据,实证研究了货币政策冲击对企业成本和进入退出动态的效应,以及利用中国数据和广义矩(GMM)方法实证分析了利率增广的菲利普斯曲线；(3)对同时包含企业投资、融资、成本和进入退出,基于国有和非国有两元企业结构的动态随机一般均衡货币政策模型的模拟数据和中国实际数据进行了比较分析。基于上述研究,本文得出如下结论：(1)从企业投资角度来看,在基准的动态随机一般均衡货币政策模型内引入内生的企业投资和预算软约束之后,货币政策效应的峰度被削弱但持续性并不受太大影响；国有企业投资的波动性整体上要小于非国有企业,但在经济上行周期,其投资波动要大于非国有企业,而在经济下行周期,其投资波动要小于非国有企业；(2)从企业融资角度来看,引入预算软约束之后,货币政策的金融加速器效应变得更为复杂。当国有企业比例处于较高水平时,金融加速器几乎不起作用,甚至基准模型中的货币政策效应水平也无法达到,但是当国有企业比例下降到一定水平(比如说0.1)时,货币政策冲击在短期内具有极强的效应,甚至超出金融加速器水平,但持续性极低,从而使经济出现大幅度的波动；(3)从企业成本角度来看,一方面,实证分析表明,货币政策冲击无论在总量和微观层面都对企业成本产生显著效应,货币政策的成本渠道成立。另一方面,对利率增广型菲利普斯曲线的实证研究结果表明,利率显著地进入了通胀动态方程,但成本渠道是否存在,对通胀预期的选取形式较为敏感；(4)从企业进入退出角度来看,实证分析表明,货币政策冲击对自我雇佣率的效应显著,但对私人企业数目的效应不显著；货币政策冲击对行业企业数目的效应显著,但信贷冲击的效应最为显著,利率冲击的效应次之,而货币供应量冲击的效应不显著；(5)基于国有和非国有两元企业结构,包含企业投资、融资、成本和进入退出的动态随机一般均衡货币政策模型,能够较好地模拟中国宏观经济波动的经验事实,特别是通货膨胀和国有企业数量的波动。更多还原

【Abstract】 The frame of monetary policy analysis based on dynamic stochastic general equilibrium has become the fundamental research paradigm of contemporary monetary theory and policy. However, the micro-base of this model is still being explored and in progress. Another problem of this frame is that its power may decrease and it may even out of work when it is used into underdeveloped country, because the structure of this frame mainly aims at the developed country or market economy. Considered this background, this article intensively and systemic studies the important micro-base of monetary policy effect: firm behavior, and explores the mechanism of the process of monetary policy, firm behavior and business cycle in the frame of monetary policy analysis based on dynamic stochastic general equilibrium, and tries to extend the basic dynamic stochastic general equilibrium monetary policy model through discriminating state-owned and private-owned firm.The article first surveys the documents about monetary policy, firm behavior and business cycle, and then illustrates the basic dynamic stochastic general equilibrium monetary policy model. Based on this work, the article analyses the behavior of investment decision of state-owned and private-owned firm, and realizes the first extension of the basic model. Following this research rule which is from simple to complication and relaxes presumptions gradually, the article in turn gives the theoretical and positive analysis of the decisive behavior of state-owned and private-owned firm from the perspectives of firm financing, firm cost and firm entry and exit, then in turn realizes the extension of the basic model, and eventually builds an dynamic stochastic general equilibrium monetary policy model which includes the former four kinds of firm behavior and bases on the two elements structure of state-owned and private-owned firm, and based on Chinese data, the article gives numerical simulation of the model. In the end, the article proposes relevant policy suggestions.Based on these contents and consideration of research, the core innovation of the article can be generalized in extending the basic dynamic stochastic general equilibrium monetary policy model through constructing the dynamic stochastic general equilibrium monetary policy model bases on the two elements structure of state-owned and private-owned firm in sight of Chinese practice, and then the article incorporates the soft budget constrain into the mainstream paradigm of monetary policy analysis. So, the article gives primitive try in developing and building dynamic stochastic general equilibrium monetary policy model for developing country. In details, the core innovation of the article can be expressed in:(1) constructing three models which bases on the two elements structure of state-owned and private-owned firm, and which respectively includes endogenous firm investment and soft budget constrain, incorporates soft budget constrain into financial accelerator, includes all the four firm behaviors:investment, financing, cost, entry and exit; (2) Based on the panel data of Chinese firm and industry level, the article positively studies the effect of monetary shock on firm cost and entry and exit, and positively analyses the rate-augmented Philips curve through using Chinese data; (3) Based on Chinese data, the article gives the calibration of parameters value and numerical simulation of the dynamic stochastic general equilibrium monetary policy model which bases on the two elements structure of state-owned and private-owned firm, and includes all the firm behavior of investment, financing, cost, entry and exit, and then gives comparative analysis between real and simulative data.Based on these studies, the article obtains following conclusions:(1) From the perspective of firm investment, when introducing endogenous firm investment and soft budget, the summit of the effect of monetary policy is weakened, but its persistence isn’t influenced. The investment fluctuation of state-owned firm is in all less than that of private-owned firm, however, in the expansive period, the investment fluctuation of state-owned firm is more than that of private-owned firm and in the contractive period, it is less than that of private-owned firm; (2) From the perspective of firm financing, when incorporating soft budget, the financial accelerator effect of monetary policy becomes more complex. When the rate of state-owned firm is high, the financial accelerator hardly works, and the effect of monetary policy cannot even catch up with the level of the basic model. However, When the rate of state-owned firm decreases some level(for example 0.1), the shock of monetary policy generates an very strong effect in short period, even belong the level of the financial accelerator effect, but the persistence is very weak, and then makes the economy fluctuate largely; (3) From the perspective of firm cost, on the one hand, the positive analysis suggests that monetary policy shock generates significant effect on firm cost both on aggregate or micro level, and the cost channel of monetary policy is supported. On the other hand, the positive research of rate-augmented Philips curve suggests that rate significantly enters into the equation of inflation dynamic, but the existence of the cost channel of monetary policy is sensitive to selection of the formation of inflation expectation; (4) From the perspective of firm entry and exit, the positive analysis suggests that monetary policy shock have significant effect on self-employed rate, but have no significant effect on amounts of private-owned firm. Monetary policy shock generates significant effect on amounts of firm in industry level, but credit shock has the most, interest rate shock has the more and monetary supply has no significant effect; (5) The dynamic stochastic general equilibrium monetary policy model which bases on the two elements structure of state-owned and private-owned firm, and includes all the firm behavior of investment, financing, cost, entry and exit, can simulate the stylized facts of Chinese macro-economy quiet well, especially fluctuation of inflation and the amounts of state-owned firm.更多还原