【作者】 严鸿雁；

【导师】 苑泽明；

【作者基本信息】 天津财经大学 ， 会计学， 2013， 博士

【摘要】 动态资本结构理论认为,企业内外环境的变化使得企业最优资本结构的选择成为一个动态过程,交易成本的存在使得企业的实际资本结构偏离其最优资本结构,因此企业需要不断地调整其资本结构以达到最优状态。国内外研究都观察到高新技术企业的平均负债率水平明显低于非高新技术企业,高新技术企业呈现出明显的“低杠杆”特征。但是在进一步经验分析高新技术企业这种“低杠杆”特征的影响因素时却得到相互矛盾的结论。究其原因,是因为资本结构理论并不是用来解释实际可观测的资本结构之间的差异,而是来解释不同公司间最优资本结构之间的差异,因此局限于静态研究框架,无法解释观察到的这种“低杠杆”现象。高新技术上市公司“低杠杆”特征的本源,是由于最优资本结构较低?还是因为未能达到最优状态?哪些因素导致资本结构无法达到最优状态?怎样才能实现资本结构的优化?这些问题都有待理论分析和实证检验。根据高新技术企业的判别方法,将中国上市公司划分为“高新组”和“非高新组”,通过对资本结构特征的统计分析,发现与非高新技术上市公司相比,中国高新技术上市公司确有很明显的“低杠杆”财务特征。研究同时发现中国高新技术上市公司资本结构在年度、行业以及生命周期方面存在显著的差异,因此,有必要在分析资本结构决策时考虑宏观经济、制度、行业以及生命周期等因素的影响。在资本结构影响因素理论分析的基础上,基于高新技术企业“高风险”、“高收益”、“无形资产和技术创新的特殊地位”的特征,提出“高新技术企业目标负债率低于非高新技术企业”；“高新技术企业与非高新技术企业资本结构优化速度不存在差异”的研究假设。由于最优资本结构是一个不可观测的量,为防止结论的片面性,本文利用中国高新技术上市公司1999至2010年的非平衡面板数据,采用混合OLS,固定效应和Tobit回归多种方法进行估计,将模型的拟合值设定为公司的最优资本结构,同时考虑行业年度中位数负债率作为公司的最优资本结构替代指标。通过Permutation Tests组合检验的方法比较“高新组”和“非高新组”各个最优资本结构的差异,“高新组”比“非高新组”目标负债率低的假设得到了验证。进一步构建准动态模型,分别用两阶段和一阶段估计方法估计优化速度,分组检验,通过自体抽样(Bootstrap)方法比较两组优化速度,检验结果表明,“高新组”和“非高新组”优化速度不存在显著差异。在动态研究框架内,上述两个理论假设得到检验,可以从理论上解释高新技术企业低杠杆的现象：即高新技术企业特征决定了其最优资本结构较低,因此,在总体动态优化速度不存在差异的情况下,高新技术企业的实际负债率较非高新技术企业低。资本结构动态优化的实质是一个动态调整与控制过程,首先需要辨识资本结构动态优化的影响因素。按照维格与威廉姆森的交易成本因素分析,结合资本结构优化行为的具体情况,把影响企业资本结构优化的因素简化为以下四个主要特征变量：调整数量、公司特征、资本结构优化的决策效率和外部市场因素。通过对中国高新技术上市公司的资本结构优化速度影响因素进行实证检验,得到以下结论：第一、偏离最优负债率的程度与优化速度显著负相关；企业规模、盈利能力、成长性、资金缺口与优化速度显著正相关；公司内部治理效率和外部环境与优化速度显著正相关。第二、为了进一步研究高新技术上市公司资本结构优化程度,将最优负债率与实际负债率之间的比值定义为“最优比率”。研究表明,我国上市公司负债不足,最优比率的均值为1.146。第三、基于动态研究视角发现,高新技术企业实际观察到的负债率自导入期至衰退期,随着成长性的减退,逐渐降低,但是最优负债率则出现随着成长性降低,目标负债率增加的趋势。这种背离通过分析不同生命周期资本结构优化速度影响因素的差异得到了解释。最后提出实现资本结构的动态优化,首先要关注资本结构的动态最优状况,资本结构动态优化能否实现的核心则在于优化速度。进而从降低交易成本的角度提出实现资本结构优化的内部控制机制和外部约束机制。更多还原

【Abstract】 Dynamic capital structure theory believes that the change of external and internal environment of the enterprise makes the choice of optimal capital structure of enterprise becoming a dynamic process. The existence of trading costs makes the actual capital structure to deviate from its optimal capital structure. Therefore, enterprise needs to constantly adjust its capital structure in the changing environment so as to achieve the optimal status. The researchers observed that the average debt ratio level of high-tech enterprises is obviously lower than that of non-high-tech enterprises. High-tech enterprises show obvious "low financial leverage" characteristic. However, when making further empirical researches on the influential factors of this kind of "low financial leverage" characteristic of high-tech enterprises, contradictory conclusion is obtained. This is because the capital structure theory is not used to explain the actual observable differences between capital structures, but used to explain the differences between the optimal capital structures of different enterprises. It is limited to explain this kind of phenomenon under such static research frame. What is the source of "low lever" characteristic of high-tech enterprises? Is it because of the low optimal capital structure of high-tech enterprises or the adjustment to the optimal status? Which kinds of factors lead to the deviation from the optimal status? How to realize optimal capital structure? Theoretical analysis and empirical tests need to be made for all these questions.Firstly, the method of distinguishing high-tech enterprises is defined in this paper.The listed companies in China are divided into "high-tech group" and "non-high-tech group". It is discovered that Chinese listed high-tech enterprises do have obvious "low financial leverage" financial characteristic comparing with Chinese non-high-tech listed enterprises through the statistical analysis. It is also found out in the research at the same time that the capital structures of Chinese high-tech listed enterprises have remarkable differences in such aspects as year, industry and life cycle. Therefore, it is necessary to take the influence of such factors as macro economy, system, industry and life cycle into consideration when analyzing capital structure decision-making.The research hypothesis of "high-tech enterprises’ target debt ratio is lower than non-high-tech enterprises" is proposed based on the characteristics of "high risk","high yield" and "special position of intangible assets and technical innovation" of high-tech enterprises on the basis of the theoretical analysis on the influential factors of capital structure. The research hypothesis of "there is no differences in the capital structure optimal speeds of high-tech enterprises and non-high-tech enterprises" is proposed based on dynamic capital structure theory in this paper. As the optimal capital structure is an unobservable variable, and in order to prevent the one-sidedness of the conclusion, the non-balance panel data of Chinese high-tech listed enterprises from1999-2010is utilized and many methods such as OLS, fixed effect and Tobit return are adopted in this paper to make estimation. The fitted value of such models was set as the enterprise’s optimal capital structure. Meanwhile, the industrial annual median debt ratio is considered as the proxies of enterprise’s optimal capital structure. Through the comparison on the differences among the optimal capital structures of "high-tech group" and "non-high-tech group" with the method of Permutation Tests combination tests, the research hypothesis of "high-tech enterprises’target debt ratio is lower than non-high-tech enterprises" is verified. Standard dynamic model is further built in this paper. The two-stage and one-stage estimation methods are respectively adopted to estimate the optimization speed. And the two groups’ optimization speeds are compared through group tests and Bootstrap method. The test results show that there is no remarkable difference between the optimization speed of "high-tech group" and "non-high-tech group". In the dynamic research frame, the above mentioned two theoretical hypotheses are verified. So high-tech enterprises’ low lever phenomenon can be explained theoretically:namely the high-tech enterprises’characteristics determined its comparatively low optimal capital structure. Therefore, under the circumstance that the general dynamic optimal speeds have no differences, the actual debt ratio of high-tech enterprises is lower than that of non-high-tech enterprises.The essence of dynamic optimization of capital structure is a dynamic adjustment and control process. It needs to identify the influential factors to the dynamic optimization of capital structure. The factors affecting enterprises’capital structure optimization are simplified into four major characteristic variables:adjustment amount, company characteristic, decision-making efficiency of capital structure optimization and external market factor according to the analysis on trading costs factors and the specific situations of the capital structure optimization behaviors. Through the empirical test on the influential factors to the capital structure optimization speed of Chinese high-tech listed enterprises, the following conclusions are obtained:first, optimal debt ratio deviation degree is in remarkable negative correlation with optimization speed; enterprise scale, profit ability, growth, financial deficit are in positive correlation with optimization speed. Enterprise’s internal management efficiency and external environment are in positive correlation with optimization speed. Second, in order to further research the optimization degree of the capital structure of high-tech listed enterprises, the ratio between optimal debt ratio and actual debt ratio is defined as "optimal ratio". It is showed in the research that Chinese listed enterprises lack of debts, so the optimal ratio’s mean value is1.146. Third, it is discovered based on the dynamic research view that the actually observed debt ratio of high-tech enterprises has been decreasing with the slow-down of growth from the star-up stage to degenerating stage, but the optimal debt ratio is in the trend of that the target debt ratio increases with the decrease of growth. This kind of deviation is explained through the analysis on the differences of the influential factors to the capital structure optimization speeds in different life cycles.In the end, a conclusion is proposed to realize the dynamic optimization of capital structure, the dynamic optimal status of capital structure should be firstly concerned. The key to the realization of capital structure’s dynamic optimization lies in the control of optimization speed. And then the internal incentive system, supervision mechanism and external control, and restraint mechanism for realizing the optimization of capital structure are proposed from the angle of reducing trading costs.更多还原