1、中国税收收入能力估测及其应用研究 The Study on the Estimation of Tax Capacity and Its Application in China 梁季摘 要近年来，随着依法治税进程的不断推进，税务机关在确保完成税收收入预算的同时，更加注重提高征管质量，堵塞税收流失漏洞，实现应收尽收。研究税收收入能力将有助于考量我国征管水平，发现征管薄弱环节，为提高税收管理能力提供支持。另外，研究税收收入能力，对改革我国税收预算编制方法，建立科学、合理的税收计划管理体制也具有重要意义。 税收收入能力指标不仅在税收管理领域应用广泛，它在衡量地区间财政收入能力、考察财政转移支付制度
2、是否科学、评价税制结构优劣等方面也大有作为。 尽管国外对税收收入能力的研究由来已久，且积累了较为丰富的研究成果。但由于国内对税收收入能力研究起步较晚，重视程度不够，从而造成相关研究成果较为匮乏。为数不多的文献也大都以学术论文形式散见于各类学术期刊中，更有相当一部分文献是作为其他研究对象的辅助部分而存在的，因此对税收收入能力的研究缺乏系统性。通过阅读各类文献发现，国内对税收收入能力基本理论研究极为欠缺。如税收收入能力的定义尚未统一，税收收入能力、纳税能力以及征税能力混为一谈；税收努力、征收率等指标相互替代使用等。税收收入能力的估测实践缺乏系统性，估测方法较为单一，且大多局限于对有限几个税种的估测
3、。指标的应用也仅限于对征管水平的考量。 随着对税收收入能力的日益重视，其估测及应用的实践探索也逐渐增多，迫切需要对我国税收收入能力估测及应用的研究成果理论化、系统化，用以指导我国税收实践。研究税收收入能力问题有也助于对我国财税工作中一些重大实践问题进行更为综合、全面、定量的分析，对完善税制、加强征管、建立科学转移支付制度具有现实的促进作用。 以上原因促使我选择税收收入能力作为博士论文的研究对象。 本论文的研究目标主要包括三个方面：一是分析与税收收入能力相关的若干基本理论问题；二是系统详尽地介绍国外税收收入能力估测方法，并借鉴这些方法对我国税收收入能力进行定量分析；三是根据估测结果，分析、解释一
4、些财税现象，验证某些结论。 从研究方法上来说，本文采用规范分析与实证分析相并重、定性分析与定量分析相结合、借鉴国外研究成果与立足于本国实际相结合的分析方法。在此基础上，立足于我国具体实践，侧重于使用规范分析与定量分析方法。 本文共有六章。从文章的内容上看，除了第一章为导论外，全文的主体包括四个部分，第 2 章属于第一部分，从总体上考察税收收入能力的基本问题，第 3 章属于第二部分，介绍税收收入能力估测方法之一， “税柄”法；第 4 章属于第三部分，介绍税收收入能力的另外一种估测方法，RTS 法；第 5 章、第 6 章属于第四部分，介绍税收收入能力估测的最后一种方法，分税种法。每一部分都包括各估
5、测方法的详尽介绍、在我国的应用以及估测结果的分析三方面内容。 第 2 章，税收收入能力的一般考察。首先从定义、相关概念、影响因素等方面考察税收收入能力，然后简要地分析了税收收入能力估测一些基本问题，最后对税收收入能力估测方法作一简要地介绍，为下文的展开奠定基础。该章从经济、税制、征管的角度将税收收入能力划分为三个层次，即大口径的税收收入能力、中口径的税收收入能力、小口径的税收收入能力。然后对纳税能力、征税能力、税收努力、征管力度等其他相关概念进行辨析。接着对决定税收收入能力的经济因素、税制因素及征管因素进行定性分析，最后分析了税收收入能力估测的三原则，即客观性原则、重要性原则及以数据定方法原则
6、，并简要介绍税收收入能力估测的基本要求。第 3 章，中国税收收入能力估测“税柄”法。主要介绍了 “税柄”法产生的背景、内容及其应用情况，在此基础上创造性地将其用于估测我国地区税收收入能力，并得出若干结论。 “税柄”法是上个世纪 60 年代由国际货币基金组织经济学家们创建的，旨在评价一国政府的运行绩效，同时为国际援助提供决策依据。 “税柄”法的主要估测思路是，在假设一国税收收入能力与其经济特征之间有着某种规则关系的前提下，在税收收入能力与其关系密切的主要经济指标之间建立线性回归方程，并根据样本国家数据确定回归系数。将回归系数作为相应税基的“平均”有效税率，估算一国税收收入能力的过程，就是把这一系
7、列“平均”有效税率适用于该国相应税基计算该国预期税收比率的过程。接着将该方法用于估测我国 31 个地区 1996-2003 年间的税收收入能力。在此基础上得出结论：（ 1）经济因素成为我国税收比率的决定性因素，二者之间的互动关系在增强，即其决定性有随时间增大的趋势，而税收努力对税收比率的影响越来越小；（2）我国税收收入能力与人均GDP 呈显著的正相关关系，与对外开放度呈负相关关系，与农业占比呈负相关；（3）地区间税收努力指数表现出随时间而趋于相同的倾向。认为地区间税收努力指数趋同的原因在于各级税务部门对依法治税的重视。 第 4 章，中国税收收入能力估测RTS 法。首先详尽地介绍了 RTS 法产
8、生的背景、内涵、构成要素及其应用情况，然后在借鉴国外 RTS 法的基础上，分析构建我国 RTS 法估测体系的重点和难点；最后利用 RTS 法对我国地方主要税种的收入能力进行了估测，并得出若干结论。 RTS 法于上个世纪 60 年代由美国政府间关系咨询委员会创建。美国各州之间税制的差异以及转移支付制度的需要，促使美国政府间关系咨询委员会创建了统一的“代表”性税制，用以估测各州的税收收入能力。RTS 法由五个基本要素组成：（1）收入覆盖范围；（2 ）税源分类；（3）标准税基的定义；（4）标准税率的定义；（5）对标准税基实施标准税率所得收入的汇总计算。其中，确定标准税基是该方法的重点和难点。RTS
9、法自创建以来，在世界各国广泛使用，估测结果也在各种经济模型中作为解释变量而频繁出现，该方法随着税收实践的变化而与时俱进。本文在借鉴美国 RTS 法估测体系基础上，分析了构建我国 RTS 法估测体系的重点和难点，指出：我国构建 RTS 法的重点和难点虽然也在于标准税基的确定，但其产生原因却与美国不同，美国是由于各州税制不一致，而我国税制统一。造成我国标准税基难以确定的主要原因是由于部分税种的标准税基数据难以获取。提出解决问题的两种办法：一是寻找“代用”税基数据，按照“代用”税基与标准税基之间关系，将代用税基转换为标准税基；二是以“代用”税基作为标准税基，确定标准税率。按照第二种思路，估测了我国各
10、地区 1999-2002 年地方主要税种的收入能力，并粗略地比较了地区间财政能力。得出结论为：（1）地区间财政能力差异极大，排在前五位地区的财政能力是排在后五位的 15 倍以上，并且这种差距有扩大趋势；（ 2）税收努力从30至 135不等，且大部分地区税收努力程度有逐年降低的趋势。建议税收努力指数低的地区应着重提高税收努力；而税收努力指数高，但财政能力弱的地区需要中央政府强有力的转移支付支持。 第 5 章，中国税收收入能力估测分税种法（上） 。在回顾前人研究的基础上，分别采用不同方法对转税类中增值税、消费税、营业税逐一进行估测，在估测结果基础上，计算各税种的税收努力及税收流失情况。首先根据增值
11、税与投入产出表之间的联系，构建增值税收入能力估测方法，即直接消耗系数法。在此基础上，估测我国 1999-2003 年国内增值税收入能力（未扣除减免因素） ，它们分别为 7132.98 亿元、8256.94 亿元、8900 亿元、9898.99 亿元和 11219.26 亿元。税收努力程度（扣除减免因素）为67、 68、74、77、79。得出我国国内增值税税收努力逐年提高，征管漏洞逐年减小的结论。以国家税务总局 2003 年消费税税源调查数据为样本，分税目估测我国 1999-2002 年的消费税收入能力，它们分别为 906 亿元、947 亿元、 1086 亿元、1206 亿元，税收努力分别为 9
12、3、91 、86 、87。得出消费税税收努力逐年降低的结论，并简要分析了其中原因。使用 RTS 法，按照第一种解决标准税基问题的办法，估测了我国 1999-2002 年营业税的收入能力，它们分别为 2094 亿元、2376 亿元、2615 亿元和 2873 亿元，税收努力分别为 81、79 、80 和 86，并对结果的准确性进行了说明。 第 6 章，中国税收收入能力估测分税种法（下） 。在回顾前人研究基础上，构建企业所得税收入能力的估测模型，并阐述企业所得税收入能力的估测思路与步骤。分析了影响我国个人所得税收入能力的主要因素，在此基础上提出我国个人所得税收入能力的估测重点，并采用简便方法对个人
13、所得税主要税目的收入能力进行了估测。最后，对其他类税种中收入占比相对较大的股票交易印花税及城市建设维护税的收入能力进行了简单的估测。在进行企业所得税收入能力估测时，首先分析了影响企业所得税收入能力的主要因素，即利润、税前列支项目、优惠政策、经济类型、行业、企业规模等几个因素。在此基础上，提出了使用因素法估测企业所得税收入能力，并构建估测模型和回归方程。企业所得税收入能力估测的步骤主要包括：数据抽样、数据分类、量化收入能力与毛收入之间关系，估测收入能力。最后按照简便方法估测我国 2001-2002 年企业所得税的收入能力分别为 2940亿元和 3447 亿元。在估测个人所得税收入能力时，首先分析
14、了我国个人所得税税源在地域、城乡、行业方面的差异，在此基础上提出个人所得税的估测重点，即以发达地区为重点、以城市为重点、以金融业、IT 业等行业重点、以国有企业、 “三资”企业及个体经济为重点、以工资薪金、利息所得及个体经营收入为重点。然后简要地阐述了分类法估测步骤。最后估测了 2003 年城镇居民工资薪金项目以及利息项目的个人所得税收入能力，估测值分别为776 亿元和 444 亿元。 本文最大特点就是利用公开发表数据，使用传统、经典计量分析方法，构建简单实用估测模型，得出一目了然的有价值结论。 本文的特色主要表现在： 1首次将理论界和实践中使用较混乱的若干概念系统地进行辨析。 2利用 199
15、5-2004 年国家统计局公开发表的经济数据，将分析国家间税收收入能力的模型创造性地应用于对我国地区间税收收入能力的估测上，得出我国地区税收收入能力与各地区人均国内生产呈显著的正相关关系，但与经济结构、对外开放度之间关系不显著的结论，同时还得出我国自 2000 年以来，经济因素对收入能力的决定程度在逐年提高，地区间税收努力程度也越来越接近等结论。 3在借鉴国外 RTS 法的基础上，提出构建我国 RTS 估测体系的重点和难点，并对我国地方主要税种的收入能力进行估测。定量地分析了我国地区间税收收入能力差异情况。估测结果表明：我国地区间财政能力差异极大，需要强有力的中央转移支付制度来缩小地区间公共服
16、务提供能力的差异等结论。 4发展和完善了增值税收入能力的方法直接消耗系数法，并对我国近几年增值税收入能力进行估测，分析了决定增值税收入能力的关键因素。 5首次对消费税的收入能力进行定量分析，并得出近年来消费税税收努力有下降趋势，并简要分析了其中原因。 6首次尝试性地建立企业所得税收入能力模型，并提出估测思路及步骤。 7分析个人所得税收入能力的估测重点及估测方法。 不足之处在于： 1受数据的限制，尚未对我国税收收入能力与其他经济发展阶段相近的国家进行比较，这种比较对研究我国整体宏观税负是十分必要的。 2在使用 IMF 思路对我国地区税收收入能力进行估测时，应尝试地去分析地区税收收入能力与其他经济
17、因素之间的定量关系，以找出决定我国税收收入能力的关键经济因素。 3个人所得税收入能力的估测模型需待进一步完善。 4估测结果的应用范围应需进一步拓展。 AbstractIn recent years, with the deepening of the course in the administration of tax by law, tax revenue bureaus put more emphasis on enhancing the tax administration and collection, stopping the tax evasion, reducing the g
18、ap between the ideal tax system and real tax system than accomplishing tax revenue budget. Studying on tax capacity contributes to evaluating the level of the tax administration and collection, discovering the weakness of the tax administration and collection, supporting for the improvement on the t
19、ax administration also, it is important to reform the method of budgeting the tax revenue, and establish a rational and scientific tax revenue budget system. Tax capacity index is used widely in the tax administration, furthermore, it will play an important part in comparing the fiscal capacity amon
20、g regions, reviewing the rationality of the transfer system, and evaluating the tax system. Although the studying on tax capacity is time-honored in the foreign countries and accumulated fruitful literature, domestic literature is scarce because it is ignored and studied late. The few literature are
21、 scattered in the learning periodicals, and most of them are the supplement of the other literature, so they are unsystematic. After reading the literature, it is found that there is scarce in the domestic literature on the theory study on tax capacity, for example, there is no the commonly accepted
22、 definition about the tax capacity, the terms “tax capacity“, “the ability of the government to collect taxes”and “the taxpayers ability to pay taxes” are mixed used, the term “tax effort” is the alternative of the term “the ration of the tax collection”. The measure method of the tax capacity is si
23、ngle and ordinary, the few types of tax are measured, and the use of the measure results is limited to the tax administration. With the more attention to the tax capacity and the expanding of its use it is urgent to systematize the theory and the measure methods of tax capacity to guide our tax prac
24、tice. Studying on tax capacity is helpful to analyze the practical tax issues comprehensively and quantificationally to promote the improvement of the tax system, enhancement of tax administration and collection, establishment of the transfer system. All these motive me to do the research on such a
25、subject. The primary aim of this dissertation mainly embraces three: the first is to analyze some basic theory issues concerning tax capacity. The second is to introduce some measure methods and apply the methods to measuring tax capacity in China. The third is to analyze, explain some tax practical
26、 phenomena and verify some conclusions based on the measure results. As for analytic methodology, this dissertation is characterized by combination of the positive with normative analysis, mix the qualitative and quantitative analysis and co-existence of the consideration for the domestic tax practi
27、ce and the use for reference about the external study fruits. Also, based upon domestic tax practice, it emphasizes particularly on the normative and quantitative analysis. The dissertation consists of six chapters. As concerning the contents, except the chapter one, the main body of the dissertatio
28、n is divided into four parts. Chapter two belongs to the first part, which investigates the basic issues on tax capacity comprehensively. Chapter three belongs to the second part, which review one of the measure methods of tax capacity, e.g. tax handle method. Chapter four belongs to the third part,
29、 which discusses at large another measure method, RTS. The fourth part comprises chapter five and chapter six, which investigate the measure methods of all types of taxes. Each of the parts consists of the introduction of the measure method, the use of the method in China and the analysis about the
30、results. Chapter two, General introduction on the measure of the tax capacity. As first of the definition, other concerning conceptions, and determinant factors of tax capacity are introduced, and then some basic issues on tax capacity measurement are discussed, at last, the measure methods of tax c
31、apacity is introduced briefly. From the point of the economy, tax system and the tax administration and collection, the tax capacity is defined：large-scopical tax capacity, medium-scopical tax capacity, and small-scopical tax capacity. And then the terms tax capacity; the ability of the government t
32、o collect taxes and the taxpayers ability to pay taxes are discriminated. The economic factors, tax system factors and tax administration and collection factors, which determine the tax capacity, are analyzed qualitatively. Finally ,the three principles of measuring tax capacity is introduced, which
33、 are objectivity principle, significance principle and method-choice by data principle, and the basic demand for measuring the tax capacity is introduced briefly. Chapter three, the measurement of the tax capacity in China: Tax handle method. Firstly, the background, the model and the use of the tax
34、 handle method are reviewed, and then based on the former, the tax handle method is applied to measuring the regions tax capacity in China in a creative way, and some conclusions are drawn. The economists of the IMF, whose aim is not only to evaluate the performance of the public sector activities b
35、ut also help to international aid, found tax handle method in the 1960s. The approach is: according to the assumption that a countrys tax capacity is relevant to its economic and social characteristics, the regressive equation is established between the tax capacity and the economic factors which is
36、 determinant to the tax capacity, and then the average effective rates is computed that are the net regression coefficients for the samples data. The tax capacity is the result that if a country applied to its tax bases a set of average effective rates on those bases. The approach is applied to meas
37、uring the tax capacity of the 31 regions in China for 1996-2003. On the basis of the results, some conclusions are arrived which are: the economic factors are the determinant ones for the tax ratio in China, the trend is more and more significant, also, the tax effort influence on the tax ratio is m
38、ore and more small. The relation with the tax capacity in China and the per capita GDP is positive significantly, and the tax capacity is relevant positively to the size of the foreign trade and negatively to the agricultural share of income, the tax effort indexes among the regions tend to be simil
39、ar over time.Chapter four, the measurement of the tax capacity in China: RTS method. Firstly. The background, definition, elements and use of the RTS are introduced. And then the emphasis and difficulties faced by Chinas establishing the RTS method are discussed. Finally, in the RTS, the tax capacit
40、ies of the key local types of tax are estimated, and some conclusions are drawn. RTS was founded in the 1960s by the ACIR，because the tax system among the states in USA is very different, and for the purpose of the grant-in-aid, the federal government has to establish countrywide standard tax system
41、 to measure the tax capacity for each state. The RTS has five basic elements. These are: (1) the revenue coverage, (2) the classification of revenues into separate sources, (3) the definition of a standard tax base for each revenue source, (4) the definition of standard tax rate for each revenue sou
42、rce, and (5) the estimation of RTS revenues for each state by applying the standard tax rate for each revenue source to the defined tax base of the state for that source and by summing the results for all sources. One of the most important elements is the definition of a standard tax base. Since the
43、 establishment of RTS, it has been used world widely. Given that state and local taxing practices tend to change gradually over time, the RTS must be updated periodically. On the basis of the RTS in USA, the emphasis and difficulty are analyzed by establishing RTS in China, which lie in the definiti
44、on of a standard tax base. However, the reason is different from that of USA，it is because relevant data of reasonably good quality is not available. The problem is solved by two approaches: one is to look for the proxy tax base, and then get the standard tax base by converting the proxy tax base to
45、 the standard tax base according to the relation between them. Another is to look the proxy tax base as the standard tax base, and then determine the tax rate t relevant to the tax base. According to the latter, the tax capacity of the local key type of the tax for the each region is estimated for 1
46、999-2002. On the basis of the results, some conclusions are arrived, these are: (1) the gap for the fiscal capacity among the regions is very large, the amount of the regions which is listed in the first 5 place is over 15 times that of the regions which is listed in the last 5 place, and the gap te
47、nds to large over time, and (2) Tax effort index ranges from 0.3 to 1.35, and most of them tend to fall over time. It is recommended that the tax effort could be raised for the regions whose tax effort are lower and tax capacity are smaller, and the central government should support them for the reg
48、ions whose tax effort are higher and tax capacity are smaller l by grant-in-aid. Chapter five, the measurement of the tax capacity in China: tax-type method(). On the basis of the earlier literature, the tax capacities for the VAT, Sale tax, and the Excise are estimated respectively, and then accord
49、ing to the results the tax evasion is computed. Firstly, underlied by the relation between the VAT and input-output chart, the measure method for the domestic VAT, direct input coefficient method is established, and then using the method, the tax capacities for the domestic VAT is estimated for 1999-2003, these are: 713.2 billion yuan, 825.6 billion yuan, 890 billion yuan, and 1121.9 billion yuan. The tax effort indexes are 0.67, 0.68, 0.74, 0.77, and 0.79. Based on the results, the conculsion that tax evasion for VAT is falling gradually is arrived. For th
