【作者】 李颖；

【导师】 高铁梅；

【作者基本信息】 东北财经大学 ， 数量经济学， 2011， 博士

【摘要】 物价波动一直以来都是经济发展和社会稳定的重要影响要素。1978年以来,我国价格体制已经从计划经济体制转换为市场经济体制。在30多年的价格改革进程中,国内各个领域的物价逐步放开,物价放开的领域在市场力量的推动下逐一蓬勃发展,但也付出了相当大的代价。随着经济全球化的不断发展,国内外的经济运行环境变得日益复杂,维持国内物价稳定已经成为我国政府和中央银行的重要调控目标,而调控手段、调控时机以及政策的有效性显然是重要且极具研究价值的问题。本文收集和研读了国内外与物价波动研究相关的大量文献,并结合我国的实际现状和统计数据,利用国际上前沿的理论和实证研究方法对国内物价的波动特征和影响因素展开研究。综合研究了通货膨胀预期、引起国内物价波动的原因及对其他领域的影响、货币政策调控效果以及物价景气指数的建立等问题,并提出了相应的政策建议。本文的主要研究工作如下：一、基于新凯恩斯主义粘性信息菲利普斯曲线理论,利用滚动方式构建VAR模型并进行样本外动态预测的方法计算出我国的通货膨胀预期序列。在西方经济学的研究中,对通货膨胀预期的关注由来已久,通货膨胀预期不仅会导致通货膨胀的发生,还会引发股票、房地产等市场出现资产价格泡沫。基于前人的研究成果和我国物价波动的具体特征,本文在新凯恩斯主义粘性信息理论框架下,从厂商最优定价的角度出发,根据我国经济的具体运行环境和物价波动特征,选择了10个宏观经济指标构成了宏观经济对物价的影响因素集Ω,利用滚动方式构建VAR模型同时进行样本外预测的方法计算出我国的通货膨胀预期πe。结果显示,由居民消费价格指数所表示的通货膨胀率πe)和通货膨胀预期πe的历史运行轨迹比较相近,但波动幅度存在差异。在物价上升时通货膨胀预期一般较通货膨胀率低；反之当物价下降时,通货膨胀预期一般较通货膨胀率高。二、利用非线性LSTR模型刻画了通货膨胀率π(Q)的动态运行轨迹和影响因素,估计结果表明经济波动、通货膨胀预期tπ和货币政策均对国内物价波动具有显著的非线性影响。本文针对通货膨胀、货币政策、通胀预期和经济波动进行了非线性检验,结果发现,非线性LSTR模型拟合效果优于线性回归模型。为了度量不同货币政策工具的物价调控效果,本文选用1-3年期贷款基准利率r和狭义货币供应量M1增长率作为货币政策工具分别构建LSTR模型。模型的估计结果显示,以利率r作为货币政策工具模型的门限值为2.8%,以Mlr作为货币政策工具模型的门限值为3.9%。当通货膨胀预期低于2.8%时,降低M1增速具有显著抑制通货膨胀的作用；当通货膨胀预期在2.8%-3.9%时,调整利率和M1增速对未来通货膨胀均具有显著的调控作用；而当通货膨胀预期高于3.9%时,调整利率是抑制未来通货膨胀的有效手段。三、以我国各领域物价指数以及与物价密切相关的宏观经济指标为基础,建立了我国物价因子增强型向量自回归(FAVAR)模型,进而得出我国物价波动对其他影响因素冲击的响应效果。本文以31种与物价相关的宏观经济指标和17种物价指数为基础,借助因子分析法的思想提取出4个宏观经济因子和2个物价因子,并据此构建了向量自回归VAR模型,即FAVAR模型。模型结果显示：(1)货币政策仍然是调控物价水平的有效手段,不过需要针对不同的调控对象和调控目标,差异化地选取货币政策工具。(2)经济结构的合理发展、进出口政策的调整,国际贸易和经济增长将不会引发严重的物价上涨。(3)资本市场的大幅波动将在长期内引起生产领域的价格波动,房地产价格对国内物价有着重要的正向冲击效应。(4)对于国内的上下游价格、粮食价格、货币政策以及国际大宗商品价格等各个角度分析表明国内物价的影响因素是多样并且复杂的,建立以大量数据为基础的FAVAR模型是一种较为实用的实证研究方法。四、基于宏观经济景气周期波动理论,构建了我国的物价景气指数以及物价预警系统,并在此基础上利用计量模型对物价走势进行了分析和预测。本文利用国际上流行的经济周期研究方法,以居民消费价格指数为基准指标,在大量与物价相关的宏观经济指标中,筛选出我国物价波动的一致指标组和先行指标组,进而合成了我国物价一致合成指数和物价先行合成指数。根据经济周期波动理论中的“谷-谷”对应法,自1997年以来,我国物价周期己经历了3个完整的循环,2009年7月-2010年10月正处于第4个循环的扩张期内。另外,本文选取了与物价密切相关的8个指标建立了我国物价预警监测系统。结果显示,物价预警综合指数与物价一致合成指数的走势很接近,可以敏感地反映物价的周期性变化,同时直观地显示出价格水平所处的状态,是一种监测国内物价波动的有效工具。本文借鉴国际上前沿的物价理论和实证研究方法,从物价波动的特征和影响因素的多个角度对我国的物价波动问题进行了研究。本文的研究对国内物价波动的变化机理,通货膨胀预期、货币政策效果等问题的研究具有推动作用,可以为中央银行等政策制定部门提供具体的研究结论和政策建议,有助于各级政府职能部门实施合理有效的宏观政策。所以,本文的研究具有重要的理论价值和现实意义。更多还原

【Abstract】 Price fluctuation has always played an important role in both economic development and social stability. China’s Price System has already succeeded in transformation to market economy from planned economy since 1978. However, during the process of prices reform over past three decades, our economy suffered a lot, though prices in every realm have gradually been more connected with market force by which all the industries have become prosperous. As economy globalization goes further, the whole world economy operation environments turn to be more and more complex. Under this circumstance, maintaining the stability of domestic prices becomes one of the most important goals for our government and central bank, and consequently researches of tools, timing and efficiency of the corresponding policy pack turn out to be very important and valuable in order to attain the goals. Through collecting and studying a mass of references home and abroad, this paper deals with the researches on the characteristics and the influencing factors of price fluctuation in China by using international frontier theoretical and empirical method research method and employing corresponding statistic data. The several issues discussed in this paper are inflation expectation, reasons that lead to domestic prices fluctuation and influence on the rest of the economy, effects of regulation and control of monetary policy and the establishment of price climate index, and at last relevant policy suggestions are put forward. The main research works are as follows:Firstly, based on the new Keynesian sticky information Phillips curve theory, this paper estimates the inflation expectation series of China by building a VAR-based model in an iterative way and then constructing out-of-sample forecasts of inflation with the model. Researches in inflation expectation have a long history in western economic field because inflation expectation is such an important factor that it does not only lead to inflation, but also leads to large bubbles in stock and real estate markets. Based on the achievements in this field in the past and the specific features of price fluctuations in China, this paper, from the perspective of optimal pricing-setting, builds a VAR-based model, guided by new Keynesian sticky information framework, in an iterative way by selecting 10 macroeconomics indicators as a whole factors set namedΩwhich can stands the extent to which macroeconomics takes effect on prices and estimates an inflation expectation series of China by constructing out-of-sample forecasts of inflation. The results reveal that the differences between history trends of consumer price index, always a benchmark of inflation, and inflation expectation are comparatively small, and yet the gaps between the two series vary at different time. The inflation expectation is lower than the inflation when prices rise, and on the contrary, the inflation expectation is higher than the inflation when prices fall.Secondly, this paper applies logistic smooth transition regression (LSTR) models to depict dynamic trajectory of inflation rateπt(Q) and its influencing factors and the estimation results show that the factors of economic fluctuation, inflation expectation and monetary policy all have a significant nonlinear effect on the domestic price fluctuation. According to the nonlinear test of inflation, monetary policy, inflation expectation and economic fluctuation, it shows that the fitting effect of nonlinear LSTR model is better than that of linear regression model. In order to measure different effects of different monetary policy tools, this paper builds LSTR models with 1～3 year loan interest rate named r and rate of money supply of narrow sense Ml named M1r, respectively. The estimation results suggest threshold value of the model which pick up interest rate, r, as monetary tools, is 2.8% and the threshold values of the latter model is 3.9%.By means of comparing and analyzing the effects of interest rate and M1r on inflation, we find:if the inflation expectation is less than 2.8%, decreasing money supply of narrow sense (M1) can inhibit inflation; if the inflation expectation is among a range of 2.8%-3.9%, both increasing interest rate and decreasing money supply of narrow sense (M1) are disinflationary; if inflation expectation is larger than 3.9%, interest rate has significant influence on inflation.Thirdly, the paper builds a factor-augmented vector autoregressive (FVAR) model by using the price indices in various fields and some macroeconomic indicators which are closely related to price fluctuations so as to obtain the response effect of price fluctuation to other influencing factors innovation. From 31 macroeconomic factors which are related to prices and 17 price indices, the paper extracts 4 macroeconomic factors and 2 price factors with the method of factor analysis to build factor-augmented vector autoregressive model (FAVAR model).The model results show that:(1) Monetary policy is still effective instrument to control price rises, but it is necessary to select corresponding countermeasures for different objects and targets. (2) The events of the rational pattern of economic structure, the policy adjustment of import and export, the international trade and the economic growth, will not induce a serious inflation. (3)Large swings in capital markets will lead to price fluctuations in the production realm in the long term and especially, domestic house prices have an important positive impact on prices. (4) The analysis of influencing factors of prices from angles of relationships between upstream and downstream prices, food prices, monetary policy and international commodity prices is so diverse and complicated that it’s wise for us to take this effective empirical method of FAVAR model based on large amounts of data.Fourthly, this paper constructs the price composite index and monitors an early warning signal system on price in China based on business cycle theory, and analyzes and forecasts the price trend by utilizing the econometrical model. The paper filters out coincident indicators group and the leading indicators group with the benchmark of the consumer price index, from the large amounts of macroeconomic indices related to prices, and then composes price composite indices- including the leading price composite index and price coincident composite index- with the international mainstream method of business cycle studies. According to "Valley～Valley" corresponding rule in the business cycle theory, it shows that our country has already experienced three round of full circles and has been in an expansion phase of price cycle since July,2009.In addition, this paper selects 8 indices to construct monitor early warning signal system on prices in China. The results show that the trend of early warning composite index is similar to that of price coincident composite index and early warning composite index is an effective tool of monitoring price fluctuation because it’s sensitive to reflect cyclical changes of prices and it can show the state of price level explicitly.By applying the international forward price-setting theory and empirical research methods, this paper studies the issue of price fluctuation in China from angles of characteristics and influencing factors of price fluctuation. The researches in this paper will benefit the understanding of the transformation mechanism of the price fluctuation, inflation expectation and the effects of monetary policy in China. The paper can also provide some specific research conclusions, especially in a quantitative way, and corresponding advices to make proper policies to ensure the healthy and harmonious development of price level for the central bank. Therefore, this paper has both important theoretic value and practical significance.更多还原