【作者】 李清彬；

【导师】 陈宗胜；

【作者基本信息】 南开大学 ， 政治经济学， 2012， 博士

【摘要】 本文系统研究了有关中国居民再分配的三个问题：倾向决定、政策形成和效应测算。这是收入再分配问题整体链条上的三个核心节点，本文的实质内容也依此分为三部分展开。第一部分集中探讨再分配倾向的决定。这一部分从再分配存在的应然性和必然性这一起点问题展开，梳理了“应然性”上的争论，从演化角度初步论证了再分配活动的“必然性”。基于既有文献的研究成果，本文提炼了影响再分配倾向的四大动机，分别是：经济利益动机、风险规避动机、公平信念动机和声誉理想动机，再分配倾向的诸多影响因素通过单独的或综合的动机来发挥作用。这些因素主要包括：身份特征、社会经济地位、教育水平、流动性预期、就业状况、流动性经历、对公平的认知、文化和社会规范、理想信仰及其他可能的影响因素。在四大动机框架下，采用“中国综合社会调查”的数据对中国的居民再分配倾向的影响因素做了经验研究，分析发现，中国居民的再分配倾向因身份特征而有所差异，并受到根植于四大动机的诸因素的影响。第二部分是再分配政策的形成、现状和作用机理。首先，该部分以个人所得税的最新改革为例，认为其政策制定过程遵循了“议程设置-公众意见-最终出台”的三步法，而再分配政策的形成过程逐渐向“参与式治理”模式转变，但未免“众口难调”的弊病；政策修正的步调存在政府的“渐进”和公众的“激进”相对应的特点，尚缺乏制度来保障对民意的高回应度，仍待进一步改革深化。第二，对我国政府的再分配项目做出界定，描述各主要项目的基本制度规则和规模特征。政府再分配项目主要包括：社会保障制度体系、个人所得税和均等化公共服务项目。目前的政府再分配基本框架逐步完善，制度体系日益健全，各项目的规模都呈上升趋势，但占比呈现更复杂的特征。第三，针对几种类型的再分配项目，探究了其影响收入分配的作用机理。认为，横向再分配项目实质在于从“高收入”转向“低收入”，实现正向调节；纵向再分配实质是平抑生命周期内的“波动”，但仍可以横向再分配的方式反映出来。通过使用基尼系数的推导，我们发现，再分配项目缩小差距的条件是较宽松的。数值模拟的结果显示：按比例支出且有上限的项目起逆向调节作用，其他项目（累进性支出、救济和福利性质收益、均等化服务收益）均为正向调节作用。第三部分为我国政府再分配政策的效应测算。这一部分专门总结再分配效应的测算方法，细致讨论测算过程中的各种选择问题，认为使用微观数据而非宏观分组汇总数据、采用“顺序分解法”更加合意，且应十分注意政策前后收入分布的构建和各种细节的处理。在此基础上，使用区分城镇和农村两套体系的典型地区微观数据以及公开的宏观汇总分组数据做出测算，整体上显示，再分配整体上是起到正向调节的作用，即起到了缩小收入差距的作用。城镇的测算结果显示，城镇整体再分配效应显著正向，较大幅度地拉低了收入差距。社保支出为负向，个税和社会性收益起到了正向的再分配作用，且社会性收益的再分配效应占绝对主导地位。细分项目的结果显示，养老金或退休金项目正向再分配效应占绝对主导地位，其他项目作用较小。显著正向再分配效应的项目主要有：养老金或退休金、个人所得税、社会救济收入；负向效应的项目是个人交纳的养老基金和医疗基金；个人交纳的失业基金、其他社会保障支出和失业保险金等项目作用微弱。以山东农村为例测得的农村再分配效应结果显示，再分配效应为正向，但数值较小。细分项目的再分配调节作用方向均为正向，从大到小依次为：粮食直接补贴收入、无偿扶贫或扶持款、离退休金或养老金、报销医疗费、抚恤金、救济金、灾款。对于中国居民再分配的主要政策建议有：第一，增强再分配力度，并应以农村居民为重点；第二，再分配政策制定应将顶层设计和参与治理有效结合；第三，应以基本公共服务为根本，构建“保险+救助”的保障体系；第四，跳出再分配的循环结：注重初次分配的公平性。本研究的创新工作体现在：第一，初步论证再分配的存在具有演化意义上的必然性，从经济社会演化的角度尝试了推理；第二，构建了四大动机框架，从而在根源动力上来理解再分配倾向，更具系统完备性，将倾向决定问题推向更深入的层次；第三，综合政治学和经济学的双重视角，以最新的个税改革政策制定过程为例，提炼了中国再分配政策制定的模式特征；第四，对政府再分配项目的作用机制做了公式推导和数值模拟，推进了对再分配项目实质作用的理解；第五，进行了更准确全面的再分配效应测算。梳理测算方法，并评述了既有研究中的问题，采用更合现实的顺序分解方法，使用天津和山东的微观数据做了细致测算分析。更多还原

【Abstract】 This thesis systematically explores three dominating parts in the logical chain onissues of income redistribution in China, which is presented in three logicallyconnected parts.Part one focuses on the Preferences for Redistribution. It starts with the originalquestion that whether redistribution “should” and “have to” exist. Based on thediscussion that whether redistribution “should” exist, this paper preliminarily derivesthat redistribution “have to” exist in the sense of social evolution. Sequentially, thispaper proposes a four-motivation framework. They are Economic Interest Motivation,Risk Aversion Motivation, Fairness&Faith Motivation and Reputation&IdealMotivation. Various factors take effect through one or more of Motivations. Thefactors mainly include identity characteristics, social and economic status, educationlevel, mobility expectation, employment status, mobility experience, belief in fairness,culture&social norms, ideal&religion, other possible factors. Under theFour-motivation framework, it adopts the data from CGSS to analyze Chinesepeople’s specific features on this issue. It’s found that preferences for redistributionare showing significant differences due to the identity characteristics, and factorsunderlying the four motivations are influencing the preferences as well.Part Two takes insights into the formation of redistribution policies, the basicfacts of China’s governmental redistribution policies and how these policies affect theincome inequality. It first discusses the recent reform on the Personal Income TaxLaw, and finds the policy formulation abides by the three steps: agenda setting, publicdiscussion and final decision. The formation process are signaling the pattern of“participatory governance”, but can’t still meet everyone’s demand; thepolicy-making speed indicates a conflict between “reform gradually” by governmentand “reform radically” by the public. There’s still no institutional guarantee to keepthe response to the public at a high level, which is waiting to be further stepped. Thispaper then turns to summarize the basic facts of governmental redistribution policies,which are defined as three parts, such as social security system, personal income tax system and equalized public services. The institutional redistribution system isperfecting gradually and the scales of redistribution programs are increasing while theratios are showing complicated features. The third subpart sheds lights on themechanism that how various types of redistribution programs take effects on theincome inequality conditions. It’s found that the nature of horizontal redistribution isthe transfer from the rich to the poor, with a positive effect; longitudinal redistributionis reducing the fluctuation in a life circle and can also be attributed to horizontalredistribution in real effect. Through derivation using Gini index, it’s found that thecondition that allows programs to be reducing income inequality is easily met.Results from the numerical simulations show that taxing programs in a fixed rate withan upper limit line will be surely functioning as a negative adjustor, meaningincreasing the income inequality. Other programs, such as progressive tax, relief andwelfare, equalized public services, are standing in the opposite position.Part Three works on the calculation of redistribution effects. It starts withsummarizing on methodologies on the calculation of redistributive effect. This thesisbelieves using micro-data instead of macro-grouped data, employing “sequentialcomposition” method will be more suitable. It also emphasizes the importance ofbuilding pre-post income and the handling of details in the process of calculation.Upon the summarizing preparations, this thesis is doing the calculations using bothurban and rural micro datasets from typical provinces and macro dataset from thepublic source. On the whole, the redistribution policies have been exhibiting positiveeffects along the aimed years. It means, due to the redistribution programs, the extentof income inequality is smaller than it should have been. Results from the urban dataindicate that the whole redistributive effects are significantly positive, narrowing theincome inequality. Redistributive effects of Social Security Expenditures are negative,while those of Personal Income Tax and Social Benefits are positive. It’s found thatredistributive effects of social benefits are dominating the whole effects. As to thespecific programs, Old-age Pension is absolutely the leading part, while others aretrivial. The positive programs are Old-age Pension, Personal Income Tax, SocialRelief, while the negative ones are personal expenditure parts of Pension Fund andMedical Fund. Unemployment Fund, Other Social Security Expenditure and Unemployment Insurance Benefit are insignificant. Results from rural area inShandong province tell that the whole redistributive effects and those of all specificprograms are positive, but with very small amounts. Arrange the programs in size ofredistributive effects from largest to smallest, and the list will be Direct GrainSubsidy Income, Poor-support Fund, Retired Payments or Old-age Pensions,Reimbursement of Medical Expenses, Pension, Relief, Disaster Relief Fund.Policies indications on China’s income redistributions are as follows.1)Strengthen the intensity of redistribution, with special attention onto rural areas.2)Redistribution policy-making process should be a combination of “top-level design”and “participatory governance”.3) Redistribution policies should be based onproviding basic public services and building a security system of “insurance plusrelief”.4) Step out of the trap of redistribution, and pay more attention on the fairnessof primary distribution.The novelty of this thesis can be summarized below. First, it preliminarilyderives the inevitability of redistribution in the sense of social evolution. Second, itbuilds a framework of four motivations thus to understand the preferences forredistribution in a rooted sense, which promotes the issue into a further depth. Third,with a combined perspective from political science and economics, it demonstratesChina’s features on redistribution-type policy-making process, taking the recentrevision on Personal Income Tax Law as an example. Fourth, it finds a way todescribe the mechanism using formula derivation and numerical simulations,profoundly clarify the nature of redistribution programs. Fifth, it calculatesredistributive effects of governmental redistribution policies in China, accurately andcomprehensively. It’s been done using the sequential composition method andmicrodata from Tianjin City and Shandong Province, after a review on methodologiesand existing researches.更多还原