资本配置效率研究

【作者】 刘建中；

【导师】 王东京；

【作者基本信息】 中共中央党校 ， 政治经济学， 2010， 博士

【摘要】 经济发展时刻面临着资源约束,如何妥善利用资源是经济学研究的核心问题。一般情况下,可以用社会总福利状态来考察资源的利用效率。社会总资源可以分成资本以及非资本社会资源。其中,资本是为了增值而存在的社会资源。如果所有社会资源都是资本,社会总福利不会达到最优状态。这是因为,社会总福利状态依赖于社会中每个人的福利状态,而社会中个人的福利又依赖于社会消费品总量、社会财富分配、法制环境、自然环境、科学技术条件、国家安全、教育水平、社会风尚等等。以上这些问题,有些可以通过资源的资本形式得以解决,有些必须通过资源的非资本形式来解决。在不发生冲突的情况下,提高资本使用效率,以及提高非资本社会资源使用效率,都会提高社会总福利状态。当两者发生冲突时,资源如何分配,依赖于何者的边际社会福利产出较高。本文集中研究资本使用效率中的资本配置效率。资本配置指的是,资本在各种使用方式之间的分配。将资本从边际收益低的使用方式配置到边际收益高的方式,会提高单位资本的产出,这就是资本的优化配置,或者说是资本配置效率的提高。可见提高资本配置效率可以提高资本利用效率。资本配置效率损失,是指某种资本配置状态与最优资本配置状态之间的差异。考察资本配置效率,首先要确定什么状态是资本配置的最优状态。为了回答这个问题,笔者提出最适帕累托最优概念。通过其定义,将不难看出,资本配置的最适帕累托最优状态就是资本配置的最优状态。配置成本是本文提出的一个概念,这个概念对于本文的论述至关重要。配置成本可以分成静态配置成本、动态配置成本两大类。配置成本又可以分成预期外信息成本、配置风险成本、以及常规配置成本。预期外信息成本来自于预期之外的投资收益率波动;配置风险成本来自于投资者风险非中性;常规配置成本来源于具体配置过程中发生的状态改变。常规配置成本可以分成常规静态配置成本以及常规动态配置成本。动态配置成本全部都是常规动态配置成本。它来源于配置过程中与配置速度相关的状态改变。常规配置成本还可以分解成进入成本、过渡成本、以及退出成本。笔者证明了,当外部性为零,并且配置成本为零时,市场自发的力量可以引导资本配置从当前状态进入最优状态。借助科斯定理,进一步证明了,当产权清晰,并且配置成本为零时,市场自发的力量可以引导资本配置从当前状态进入最优状态。因此,为了提高资本配置效率,我们需要清晰界定产权,以及降低配置成本。笔者还证明了,不管产权是否清晰,当配置成本为零时,市场自发的力量会使各种投资方式的收益率趋于相等。研究资本配置效率损失时,笔者把总体效率损失进行了分解,这样有利于把握各种损失的来源。以下是本文的主要研究结论。假定产权清晰,并且配置成本为零时的资本配置效率为U1;假定配置成本为零时的资本配置状态为U2;假定现实资本配置状态为U3。U1与U2之间的配置效率损失来源于产权不清晰。可以证明,U2代表的资本配置状态是均衡态。这个均衡态的特征是,此时各种投资方式在每个时点上的投资收益率都相等。U2与U3之间的配置效率损失来自于配置成本。假定预期外信息成本为零、配置风险成本为零时,资本配置状态为U21;假定预期外信息成本为零时,资本配置效率为U22;假定常规动态配置成本为零时,资本配置效率为U23。U21是资本配置的一种均衡态,其特征是各投资方式的周期利润率相等。U2与U21之间的效率损失来源于常规配置成本。U22也是均衡态,特征是各投资方式的风险折减周期利润率相等。U21与U22之间的效率损失来源于配置风险成本。U23同样是均衡态,在U23状态下,各领域的风险折减周期利润率差异小于常规静态配置成本。U22与U23之间的效率损失来自于常规静态配置成本。如果不把瞬时均衡看作一种均衡状态,那么U3不一定处于均衡态。U3具有向U23收敛的趋势,收敛过程中导致的配置效率损失来自于动态配置成本。最后,笔者对我国工业领域进行实证研究,考察工业领域的配置效率以及配置成本。文章采用权益净利率作为投资收益率指标,用变异系数方法测度各领域的投资收益率差异,并且通过考察新增投资与权益净利率之间的关系,解释配置效率变化的原因。在研究过程中,笔者定义了一些概念,希望这些定义能成为此类研究的基础。这些概念包括:配置、配置成本、最适帕累托最优、潜在配置剩余。笔者引入了一些新的分析方法:财富增长曲线法、关联性检验方法。并从一些新角度去思考问题:比如从状态改变的角度去思考成本。研究中,笔者还将弗里德曼的永久收入假说扩展到永久效用假说,并且放松了科斯定理成立的条件。更多还原

【Abstract】 Economic development is always constrained by limited resources. How to use resources efficiently is the core issue of economic study. Generally, we can evaluate the efficiency by the state of social welfare. All the resources in a society can be regarded as being of two types: capital and non-capital resources. Capital is that part of resources which are devoted to acquiring more resources. If all of our resources are employed in the form of capital, the optimum state of social welfare can’t be achieved. Agreed by many, the state of social welfare is depending on the welfare state of every person involved, and the state of an individual is depending on numerous parameters such as aggregate consumption, distribution, law, natural environment, science and technology, state security, education, ethnics. Some of them can be improved by using resources as capital, and some can’t. If there is no trade-off, improving the efficiency of capital and noncapital both can enhance the state of social welfare. However, if there is a trade-off, how to use resources is determined by their marginal products of social welfare.This thesis is intended to explore an important aspect of the efficiency of capital: allocation efficiency of capital. Capital allocation is the distribution of capital among different channels. If we relocate capital from channels of low marginal product to high marginal one, the average return of a certain amount of capital will be increased. This is so called optimization of the usage of capital or the improvement of allocation efficiency of capital. It is easy to see the improvement of allocation efficiency is conducive to the improvement of efficiency of capital.The loss of capital allocation efficiency is referring to the difference between the state of current state and the optimal state of capital allocation. To continue the study of allocation efficiency, the first task is to determine the optimal state of capital allocation. Responding to this question, the author proposes a concept: the best Pareto optimum. From its definition, it is easy to see that the best Pareto state of capital allocation is the optimal state of capital allocation.Allocation cost is an essential concept of this dissertation, which is proposed by the author. This concept is crucial to the discussion of this work. Allocation costs can be divided into static and dynamic allocation costs. Allocation costs also can be divided into unexpected fluctuation costs, allocation risk costs, and regular allocation costs. Regular allocation costs are composed of static and dynamic regular allocation costs. Dynamic allocation costs are all of dynamic regular allocation costs. Regular allocation costs have three continuants: entry costs, transition costs, and exit costs.The author verified that when a society is free of externality and allocation cost, the forces of market will drive capital allocation to the optimal state. By applying Coase theorem, the author further proved that if property rights are unambiguously defined, and allocation cost equals zero, the forces of market will drive capital allocation to the optimal state. Hence, for the purpose of augmenting the efficiency of capital allocation, we ought to clearly define property right and reduce allocation cost.The author also proved that as long as allocation cost is zero, the market forces tend to equate the return rates of every investment. When studying the efficiency loss of allocation, the author breaks the aggregate loss into several parts. By this means, we can understand the causes of these losses. First, if all the return rates are the same, the loss surely comes from the ambiguity of property right. Other losses that come from allocation costs can be further broken down. The main research results are as following:Suppose that when property right is fully clear and allocation cost is null, the allocation efficiency is U1; suppose that when allocation is null, the allocation efficiency is U2; suppose that the current allocation efficiency is U3. The loss between U1 and U2 stems from the unclearness of property right. It can be proved that U2 is an equilibrium state of capital allocation. The feature of this state is that the return rates of any investment are equal in every minute. The loss between U2 and U3 evolves from the existence of allocation costs.Suppose that when unexpected fluctuation cost is null, allocation risk cost is null, the capital allocation state is U21; suppose that when unexpected fluctuation cost is null, the state is U22; suppose that when regular dynamic allocation cost is null, the state is U23. U21 represents an equilibrium state, in which the periodical return rates of any investment are the same. The difference between U2 and U21 comes from the existence of regular allocation costs. U22 also stands for an equilibrium state, in which the risk debated periodical return rate of a certain investment equals any other investment. The difference between U21 and U22 is dictated by allocation risk costs. U23 is still an equilibrium state, in which the discrepancy of risk debated periodical return rates between any investments must be less than regular static allocation cost. The difference between U22 and U23 is determined by regular static allocation cost. If temporary equilibrium is not counted as an equilibrium state, U3 is not necessarily in an equilibrium state. U3 tends to converge to U23. The efficiency loss during this process is traceable to dynamic allocation costs.Finally, the author carries out an empirical study on industry to examine the allocation efficiency and allocation costs. This work uses rate of return on equity (ROE) as an indicator of return rate of an investment, and uses coefficient of variation to measure the differences among investments. The variation of allocation efficiency is explained through the relationship between new investment and ROE.In this dissertation, the author defines several concepts, which is hoped to be the foundations for further research. These concepts are: allocation, allocation cost, best Pareto optimum, potential allocation surplus. He introduces a few analyzing methods such as connection examination method and wealth accrue curve method. The author analyzes certain issues using new perspectives, for example he analyzes costs from the angle of state transformations. In this study, the author extends the permanent income hypothesis and Coase theorem.更多还原