【作者】 曹红；

【导师】 张大永；

【作者基本信息】 西南财经大学 ， 金融学， 2014， 博士

【摘要】 石油的使用效率高,并且较煤炭等传统能源清洁,其在我国经济发展过程中的地位越来越重要。然而受国内石油供给不足限制,近年来我国进口石油消费量不断增加,对国际进口石油的依赖度一度非常接近于60%。根据BP世界能源统计年鉴预测,到2030年中国的石油进口依赖度将达到80%。与此同时,一方面,近十年来国际石油价格变化了300%多,油价波动较十年前更加频繁,另一方面,中国国内石油定价机制经过几次大的改革,其市场化特征越来越明显。在这一背景下,国际石油价格冲击开始逐渐受到我国学者与相关政策制定者们的注意。对油价冲击的关注,不仅要求我们了解其对我国宏观经济的影响,也需要分析油价变化对我国股票市场的影响,如影响程度与特征、冲击传导途径等,从而更加了解油价冲击和我国股票市场的运行规律,增强对我国股市变化的预测能力,为股市的监管以及投资者风险预警等提供一定的依据。为理清国际原油价格变化对中国股票市场的冲击效应,结合现有研究中关于油价冲击与我国股票市场关系的探讨还比较有限这一现状,本文就油价对我国行业股市的影响程度与特征,以及两者关系中可能的宏观经济与消费传导途径等层面,进行了实证研究。文章首先分析了国际原油价格变化对中国能源相关类股票的影响,发现两者之间存在着随时间而变化的动态关系,并且在2008年发生了结构性变化。在结构性变化发生之前,油价对我国能源相关类股票的冲击效应几乎不存在,而在结构性变化发生之后,油价冲击对我国股市存在显著的正向影响。对油气、煤炭电力和新能源等不同类别股票投资组合收益的分类分析结果显示,国际油价变化对这些不同类型的能源相关类股票的冲击效应也全部显著为正,并且对煤炭和电力类股票的影响系数最大,对新能源行业股票影响程度略低一些。这意味着,一方面,在股票市场上,油价冲击可能引发投资者进行不同能源股之间的投资转移；另一方面,油价冲击可能会通过需求替代或刺激投资等影响非石油类公司经营活动。显示出在我国虽然石油消费占比远低于煤炭等传统能源,但油价冲击可能会在一定程度对我国能源结构配置产生影响。说明国际原油价格冲击对我国能源市场而言既是挑战也是机遇。其次,文章将股票收益率扩展至13个大类行业,基于动态条件相关性模型和因子分析等手段,探讨了国际石油价格变化率与不同行业股票收益率之间的相关关系特征以及行业之间的异同性。动态相关系数分析结果向我们展示出了不同行业股票收益率与油价变化相关系数在大小、波动性以及阶段性等上的差异：系数最大和波动最大的分别是采掘业和交通运输业股票,而信息技术和传播文化业股票与油价相关关系显著区别于其他行业。对股市冲击特征的分析表明,我国股市的收益变化主要与引起不同行业协同变化的冲击相关。加入油价变化率后的因子分析结果显示,在2008年金融危机之后,存在明显的与我国多个行业股票收益率高度相关的油价公因子,说明油价对一些行业股票的冲击能够通过行业联动扩展至其他行业。在油价冲击与不同行业股票相关性特征的分析基础之上,我们进一步采用多因子回归模型,检验了油价变化对各行业股票的冲击幅度,并考察了油价冲击可能存在的非对称性。研究结果表明,国际石油价格变化对我国采掘业、制造业、交通运输仓储业、信息技术业和批发零售业、房地产行业、社会服务业以及综合类等8个行业股票收益的影响系数显著为正,并对采掘业股票的冲击幅度最大；而对农林牧渔业、电力热力和水等生产及加工业、建筑业、金融保险业和传播文化业股票收益不存在显著影响。这意味着,在受到国际油价下跌冲击时,我国股票市场投资者可以通过构造投资组合规避该风险。非对称性分析结果表明,在油价变化与我国行业股市的短期关系中,基本不存在任何的不对称效应,油价上升和下跌对我国各行业股票投资组合收益的影响没有显著差别。再者,为了理解油价变化为何对中国行业股市存在不同于发达国家的同向冲击效应,结合已有研究中所发现的国际石油价格冲击与中国宏观经济增长之间的正相关关系,文章进一步引入GDP等4个宏观经济变量,分别分析了油价冲击与经济增长、经济增长与我国股市发展的长期关系,以及三者之间的短期互动关系,从而对宏观经济变量在油价冲击影响我国股市收益率中的作用进行探究。在国际原油价格与我国GDP的长期关系探讨中,我们分别在传统协整检验和引入非对称性后的协整模型框架下进行了检验,发现两者之间存在不对称的协整关系,显著区别与传统协整检验结果。分析结果也证实了GDP与我国股市发展之间的长期均衡关系,但在这一关系中存在着结构性变化——1999年末结果性断点发生之后,我国股市与GDP的长期关系有所下降。通过油价变化、股市收益率和宏观经济变量之间的短期互动关系分析,我们发现油价冲击首先会影响通货膨胀率和经济增长速度,并通过这两者将其波动传导至股票市场。对油价冲击来源进行分类后的短期关系分析结果表明,石油需求冲击对我国股市收益率的直接负向影响效应大于石油供给冲击,而石油供给冲击对我国GDP增速的同向影响幅度显著大于石油需求冲击,但总体上石油供给与需求冲击对我国宏观经济变量的影响差异性较小。从投资者行为角度来看,居民消费与股市之间存在着一定的联系。考虑到随着我国居民收入的不断增加、以及居民消费习惯与消费模型的改变,导致近年来我国居民家庭能源消费不断增加这一事实,结合国际石油价格变化情况来看,可能存在石油价格冲击影响居民消费的情况。因而文章还对是否存在油价变化对股市影响的居民消费途径进行了检验。研究发现油价变化与居民人均总消费变化率之间存在着负相关关系；对消费品类型和油价上涨与下跌冲击进行区分后,实证结果显示,油价上涨对食品服装、交通通信和居住类等不同类别分项消费的影响程度与方向不同,体现出在受到国际油价上涨冲击时,居民会对消费结构进行重新配置的特性。并且油价冲击还对数消费-总财富比率存在显著影响。进一步关于消费变化与股市收益率、对数消费-总财富比率与股市收益率关系的分析结果显示,我国居民消费与股市收益之间存在一定“此消彼长”的负相关关系,表明我国股市投资者具有投机特征,属于非理性的风险追逐者。因此在一定程度上可以认为,油价上涨冲击导致居民消费变化与调整,是油价变化影响我国股市收益率的潜在途径之一。并且由于油价对消费、以及消费对股票收益的影响都是负向,这也能够导致油价对股票的最终影响呈现正相关关系。与该领域内的已有研究相比,本文在探讨国际原油价格对中国股市的冲击效应中,在研究手段上,更注重对动态关系与非对称问题的分析。文章结合我国经济周期、股市发展中的政策调整等多方面实际情况,基于动态条件相关系数模型和结构性断点检验等,充分考虑了变量系统之间时变性与结构的稳定性；并在长期与短期关系分析中,都考虑了油价冲击对我国经济与股市影响中可能存在的不对称问题,尤其是在长期引入了非对称协整模型。在研究视角上,本文不仅注重从股市行业层面出发,分析油价冲击对能源相关类股票和所有行业股票的影响,还对宏观经济变量与居民消费在油价冲击影响我国股市收益中的作用,分别进行了检验,得到一些很有实践指导意义的结论。特别是对居民消费作用途径的分析,在以往的研究中还很少被关注。更多还原

【Abstract】 As an important kind of energy, oil is much more efficient and cleaner than the traditional energy, such as coal. It is more and more crucial in the process of Chinese economic development. However, because of the limited supply of oil, China became an oil importer in1993, and now the proportion of imported oil to domestic total oil consumption is close to60%. According to the BP statistical review of world energy, the oil importing proportion will grows to80%in2030, which is too high to keep China’s energy system safety. At the same time, on one hand, world crude oil prices increased from25.51dollar per barrel in January2000to107.87in December2011, and the volatility of oil prices changes is much bigger than before; On the other hand, the oil pricing mechanism in China is becoming more and more market-driven. Both the scholar and government begin to pay more attention to the impact of crude oil price shocks on China’s macro-economy. On the contrary, there is only a few papers have investigated the relationship of oil price shocks and Chinese stock market. Taking this situation and the importance of stock market to the development of economy into consideration, in this paper, we exam the effect of oil price shocks on stock returns through industry levels. We also test whether and how do the macro-economic variables and household consumption play a role in it.Firstly, we analyze the relationship between world oil price shocks and China’s energy related stocks return by adopting time varying conditional correlation and asset pricing models. Our empirical results demonstrated that international oil price changes are correlated with energy related stock returns in China, but in a time varying way. There is a clear sharp increase in the conditional correlation after mid2008, which corresponds to the onset of the financial tsunami. A higher return in energy related stocks might reflect that the market requires compensation for increased risks; therefore, we test the hypothesis that oil price changes are one type of systematic risk. A general market model and also the Fama-French three factor model are borrowed from the finance literature to investigate this hypothesis. When a structural break is incorporated, sub-sample analysis shows that oil prices are important in the period following the financial crisis. These results persisted for three sub-indices, and it was further shown that new-energy stocks are more resilient to oil price shocks than other energy sub-indices.Secondly, we use the Dynamic Conditional Correlation GARCH model and Factor Analysis to discuss how international crude oil prices shock is related to Chinese industrial stock returns. Our results show that the correlations between oil price and Chinese stock market are time varying and different cross sectors. Furthermore, we demonstrate that common factors explain most of the variation in stock returns and collaborative change across different stock sectors, and it is an important channel that world oil price influences Chinese stock market.What’s more, we examine the relationship between international oil shocks and sectoral stock returns in the Chinese stock market by using factors regression model. Our empirical results show that International oil shocks seem to have a strong effect on sectoral stock returns in China, as eight out of thirteen of these sectors have a significant and positive sensitivity to the shocks. These sensitivities vary across sectors. The mining industry is the most sensitive followed by general industry and manufacturing industry. The financial industry has the only negative coefficient, and it is not significant. There is no clear evidence of asymmetries in China’s sectoral stock-oil relationship.Thirdly, taking the facts that the oil shocks impact Chinese stock return differently from the developed markets and that several papers find there exists positive relationship between oil shocks and China’s economic development into consideration, we explore whether and how the macro-economic variables function in the relationship of oil price shocks and stock returns. In the long run, we find that crude oil prices and China’s GDP have asymmetric relationship; the cointegration coefficient is more significant and larger when oil prices increase. And the macro-economy and stock return is also cointegrated after controlling the breakpoint. In the short run, the impulse response function shows that oil shocks have effect on inflation and GDP growth rate, and the later two variables have significant positive impact on stock returns. Since household consumption has some connection with stock market in the respect of investor behavior, we test the influence of oil price shocks have upon residential consumption in China lastly. The empirical results show that oil shocks do have negative effect on average aggregate consumption. Furthermore, we demonstrate that the most immediate and direct effect passes through transport consumption, as might be expected, but also that significant effects pass through other sectors with less immediacy, including’food and clothes’,’medicine’and general’living expenditure’. Given the results, particularly observed asymmetries with respect to rises in international oil prices, it seems that residents will adjust the structure of their consume expenditure. Besides, we find that oil price shocks have influence on the trend deviation of consumption-income and wealth (i.e. log consumption-aggregate wealth ratio). In the second step, we investigate whether consumption can predict the stock return. It is fond that household consumption changes have negative effect on the return of Shanghai composite index, while the log consumption-aggregate wealth ratio has significant impact on equal-weighted ’A’share portfolio return.To sum up, we find significant impact of crude oil price shocks on Chinese stock market. We also find that both the macro-economic variables and household consumption paly important roles.更多还原