Information about Asia and the Pacific Asia y el Pacífico

1 Overview

Ehtisham Ahmad, Vito Tanzi, and Qiang Gao
Published Date:
September 1995
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details
Ehtisham Ahmad, Gao Qiang and Vito Tanzi* 

The reform of fiscal relations between different levels of government is crucial to the success of China’s current structural transformation to a social market economy. The reform will profoundly affect macroeconomic stability, growth prospects, and the effectiveness of providing public services and interregional equity. In order to review international experience in this regard and to clearly identify the policy options for China in the short to medium term, a conference was held in Shanghai in October 1993. The papers in this volume include those presented at the conference and a few that supplement it. They embrace the full range of fiscal relations between different levels of government—the background to the current policy choices; national budget policies and the legal and institutional role of different levels of government; expenditure assignments; tax assignments and administration; and transfers and grants from higher levels of government.

The conference was organized by the Chinese Ministry of Finance and the Fiscal Affairs Department of the International Monetary Fund, with the cosponsorship of the World Bank and the Shanghai municipal authorities. The conference program was drawn up by the Ministry of Finance and the IMF, and officials from both institutions presented a large number of papers. The World Bank staff also presented a number of papers and acted as discussants. The Shanghai Finance Bureau hosted the meeting, providing excellent conference facilities, interpretation, and translation, as well as a paper on local finances. The conference was attended by officials from the central departments, including the State Tax Bureau, the State Planning Commission, and the Commission for Economic Restructuring, as well as from a number of provinces.

As described in the introductory speeches by the Vice-Minister of Finance, and by Gao Qiang, Director of the Budget Management Department (Chapter 2), the reform of intragovernmental fiscal relations is critical to the stability and sustained growth of the Chinese economy in the years to come. In this context, a number of fiscal reforms will be needed simultaneously. The conference sought to highlight the main avenues and to illustrate these with examples from other countries, intertwining them with specific proposals and discussion of Chinese problems. As will become clear, the evidence from the international context displays a variety of institutions and outcomes that reflect the economic and historical backgrounds of the countries concerned. Yet there are common themes—including the importance of budgetary discipline, a clarity of budget laws and procedures, the importance of appropriate expenditure and revenue assignments, and the crucial role of objectively determined patterns of grants and transfers. Within each of these areas are also many ways of tackling specific issues that arise. Clearly, the solutions to China’s problems must display Chinese characteristics and the preferences of Chinese policymakers. It is important, however, to ensure that the chosen paths, and subpaths, are consistent and display an overall coherence.


The Chinese authorities have foreseen that sustainable growth with equity, in an economy that is increasingly integrated in the world trade and financial system, must be accompanied by widespread reform of public sector policies. This includes modernization of taxation policies and administration, together with new approaches for formulating and implementing public expenditure policies. The authorities also recognize that such reforms must take account of China’s size and diversity, thus changes to intergovernmental financial relations are essential in ensuring that policies to achieve national objectives blend with policies that take account of specific regional and local needs.

A technical assistance team from the Fund visited China in July 1993 and argued that a continuation of contracting arrangements between the central government and the provinces could be a source of macroeconomic instability in the future. A move to establish greater central government control over revenues through a National Tax Administration would be desirable. This would invariably affect the tax assignments. A greater central government share in total revenues would be sustainable only if accompanied by a system of transfers to local governments, based on scientific criteria.

China’s recent experience with fiscal reforms, as well as with some problem areas and policy options for the future, described by Gao Qiang, on behalf of the Ministry of Finance, in Chapter 2. The shortcomings of the system of contracting revenues between different levels of government are emphasized and some policy options to reform the system are highlighted. Gao Qiang explains the difficulties with the current contracting arrangements regarding taxation in China and sets out the main axes for reform that have been adopted by the government to increase revenues and provide the basis for a rational tax system. These involve both tax assignments, with the determination of central and local tax bases, and tax sharing—relating to the major sources of revenue. As pointed out in Chapter 14, this arrangement of revenue-sharing is equivalent to the establishment of non-equalizing general grants. A second major reform, presented by Gao Qiang, is the establishment of a central tax administration (State Administration of Taxation), to minimize the possibilities of local government exemptions and preferences with respect to such taxes. The service would be responsible for the central taxes and the major sources of revenues. An important question is whether “shared taxes,” which are likely to be the most important in terms of revenues, should be administered by the State Administration of Taxation. A related question is whether the corporate income tax on centrally owned enterprises should be allocated to the central government, and whether the corporate income tax on locally owned enterprises should be administered by and allocated to local governments. In most countries, the administration of the corporate income tax is invariably carried out by a single authority.

The differences between different regions of China are summarized in Chapter 3 by Mr. Zhenjun of the State Nationalities Commission. It focuses on the requirements of inter-regional equity and the policy importance of providing access to a minimum standard of public goods and services throughout China.

When the conference took place, the full extent of the planned fiscal reforms had still to be decided. As a result of the discussion at the conference, and a subsequent workshop organized by the Fiscal Affairs Department, the authorities have made a number of modifications in the proposed fiscal arrangements affecting relations between different levels of government. The most important modifications have been with respect to proposed tax assignments and revenue-sharing arrangements. These changes are reflected in the subsequent guidelines from the National People’s Congress, and the decisions of the State Council.

New Taxation Arrangements

As described in Chapter 2, an overriding objective of the reforms is to replace a system of taxation, based essentially on negotiation and contracts, with a uniform application of taxes based on a legal foundation. The introduction of a modern tax system should remove special disincentives and create a “level playing field” so that investment and labor market decisions are not biased. Initially, it was proposed that the main revenue sources, as described above, would be shared taxes, and the proportions to be shared could vary across taxes and across regions; in addition, there would have been purely central and purely local (provincial) taxes. The revised system, introduced as of January 1, 1994, is described later in the volume in Chapter 9 by Xu Shanda and Ma Lin of the State Administration of Taxation (SAT).

The main difference between the announced system and the proposal discussed at the conference is a much sharper delineation in tax assignments; the value-added tax (VAT) is the main revenue source to be shared, with a 75 percent share going to the central government and the rest to the local governments.1 The central taxes and the shared taxes—that is, the VAT—are to be administered by SAT. This arrangement will increase the initial share of central government revenues, in total revenue collections, from under 40 percent to around 60 percent. To protect local government expenditures, as an interim measure, the central government has agreed to provide transfers that would maintain the provinces’s 1993 level of nominal revenue collections. This interim measure is to be replaced by a “scientific” grants mechanism as soon as practicable.

Budget and Legal Framework

The workings of a market-based economy require a mechanism for budgeting and control that differs from that needed in a centrally planned economy. The budget plays a central role in a market economy. It provides the levers for macroeconomic control; whereas in the former centrally planned context it played a passive role in providing the financing for predetermined plan targets. The Chinese case is surveyed in Chapter 4 by Shahid Yusuf and Bert Hofman, based on the extensive work of the World Bank in this area. In Chapter 5, Ehtisham Ahmad, Maurice Kennedy, and Ingrid Klering set out the main issues that arise in establishing and clarifying budget laws. The experience of many countries of the Organization on Economic Cooperation and Development (OECD) is summarized. The authors note that it is important to clarify the relative roles of the legislature and the executive, the scope of government activities, the importance of the Minister of Finance, the relationships between different levels of government, and the mechanisms for achieving macroeconomic control and accountability of officials.

In connection with the overall macroeconomic management of the economy, the importance of the effective control over all types of borrowings by subnational levels of government needs to be stressed, including indirect methods such as the buildup of arrears and IOUs, the burgeoning of unaccounted for extrabudgetary funds, and the setting up of “dummy” organizations to borrow and perform government tasks. It is evident that a proper classification of revenues and expenditures (stressed in both papers) is crucial if the budget is to perform the functions needed in a market-based context. It would, in fact, be quite straightforward to reclassify the Chinese budget into a format consistent with the Government Finance Statistics conventions of the IMF, which are generally being followed elsewhere.

Expenditure Assignments

The assignment of expenditure responsibilities is typically the first step in determining different forms of intergovernmental fiscal relations. There is extensive debate on the degree of centralization or decentralization that may be relevant in a given context. Much of the literature, however, is based on models of the economy that depend on the mobility of factors, particularly labor and capital, which may not be relevant for China. As Gao Qiang illustrates in Chapter 2, the recent experience of China is based on “contracting” arrangements between different levels of government, which appear to have outlived their usefulness; thus, a study of the Chinese context may involve a somewhat different paradigm from that used in other countries. Important lessons may still be drawn from international experience, however; although in assessing this experience, it would be useful to keep in mind the difference between policy, administration, and financing responsibilities. These issues are stressed in Chapter 6 by Ehtisham Ahmad, in a review of international experience with respect to expenditure assignments, illustrating the diversity of experience in this regard and some of the principles that might be relevant for China.

The paper by Ahmad also examines social protection reform in China. A major issue is that a corresponding rationalization of payroll contributions for social security should accompany reform of the tax system. Thus, regions such as Shanghai should not be penalized because of the higher concentration of the aged, relative to the working population, than the average. The main issue for discussion is the case for a high level of pooling of risks for pensions and unemployment insurance, not only to reduce “tax competition” across localities, but also to ensure uniformity of provision for the main contingencies that require state support. Different considerations apply in the case of social assistance, which could continue to be provided effectively by local governments, as is presently the case in China. Interregional comparisons should be handled in the context of horizontal equalization. The issue of China’s present system of grants and transfers is raised in this context. As described in Chapter 3, the current transfers are based on comparisons of per capita incomes of the mid-1980s. The transfer system would need to better account for price changes if horizontal equity is to be appropriately addressed; this issue is taken up later (see Chapters 1416).

The subsequent papers are devoted to different aspects of expenditure assignments in China. A major issue relates to the responsibility for large or “lumpy” investments. The paper by Bert Hofman and Richard Newfarmer (Chapter 7) sets out the principles and suggests applications for the Chinese context. Somewhat different considerations apply in the context of social expenditures, and the case of health care differs from education, or the provision of housing. Zuliu Hu in Chapter 8 assesses the scope for government intervention and that of the private sector. He also examines the issue of which level of government in China should be responsible for the social expenditures and stresses that intrasectoral considerations may be important. The difficulties with the application of the benefit principle, to guide expenditure assignments, are also discussed.

Tax Policy and Administration

At the conference it was pointed out that sharing taxes in varying proportions could reintroduce the “bargaining” between tax collecting and recipient governments that the new system sought to eliminate. The distinctions between tax assignment, relating to control over tax rates and bases, administration, and revenue allocation could allow a nondistortionary system of tax assignment and administration, which would need to be supplemented by revenue-sharing and transfer arrangements. The revised arrangements are described in Chapter 9, by Xu Shanda and Ma Lin, of the State Administration of Taxation, rewritten after the conference to reflect the tax system introduced in China as of January 1, 1994. In principle, the new arrangements would limit the possibility of local government exemptions to central or shared taxes—which had in the past eroded central government control over tax bases and effective rates.

Xu Shanda and Ma Lin outline the thinking behind the proposed taxation reforms, subsequently approved by the State Council. As they explain, the guiding principles were the need to create “a uniform tax code, equitable tax burdens, a simplified tax system; a rational division of powers, straightening out of the apportionment of tax monies between the national government and local governments, and building a taxation system that is consistent with the needs of a socialist market economy.” The paper outlines the reforms proposed for enterprise profits tax, including the unification of the laws applying to state enterprises, private enterprises and other similar organizations, and introduction of measures that bring the treatment of interest and depreciation charges more into line with international practice. Measures to standardize the income tax and value-added taxes are explained, as is the rationale for proposed consumption and business taxes, which may be seen as largely complementing the coverage of the VAT. They also discuss prospective new, or amended, taxes on natural resources, urban construction, property, securities, a possible social security levy, and the readjustment, merger, simplification, and possible abolition of a host of smaller taxes, some of which have an exotic flavor (e.g., a banquet tax and animal trading tax) but in the end may be a “nuisance,” in the sense that they are unpopular, are hard to administer, and do not raise a great amount of revenue. Finally, the authors argue for more thought to be given to possible local government taxes that might be seen as linked to financing activities at that level of government.

The paper by Shi Yaobin of the Ministry of Finance, Chapter 10, concentrates on the background and reasons for enterprise tax reform. The paper complements the earlier presentation by Xu Shanda and Ma Lin, and includes a valuable account of the complexities and difficulties of the former profit-sharing system.

The discussion of the State Administration of Taxation paper stressed the importance of a centralized tax administration, as well as some potential difficulties in its establishment. Ahmad raised the issue of insulating SAT officials from local influences. It was also argued that the assignment of the enterprise income tax should not be based on the ownership of enterprises—this assignment generates duplicate administration functions and would also pose increasing difficulties for administration as patterns of ownership begin to diversify. However, the arrangements implemented as of January 1, 1994, retain the assignment of the enterprise income tax by ownership type—reflecting the otherwise relatively weak own-tax bases of the local governments. As suggested by Vito Tanzi, the tax reforms and tax assignments should be reviewed periodically, and this issue could be taken up again at a later stage.

In his own paper (Chapter 11), Vito Tanzi has a dual task. First, he surveys international experience in tax assignments, blending theoretical concepts with a policy focus and administrative realities. A pragmatic approach that recognizes the importance of stabilization policy and combines with the need for uniform administration had led a number of countries to assign responsibility for setting rates and administration of some of the major taxes to the central government level. While some taxes, such as sales and personal income tax may be operated at the subnational level, the only tax that was widely applied at the local level was the property tax.

Tanzi expresses his strong support of the thrust of the tax policy reforms proposed by the Chinese authorities, while mentioning some concerns in regard to the design aspects of some specific measures. In particular, he suggests a preference for a single rate for the enterprise income tax, rather than two. There is also a possible concern that the design of the income tax, including the relatively high deduction for living allowances, may restrict the yield of this prospectively important tax. He also cautions that the business tax, and similar taxes, should not be allowed to erode the VAT base, which is so vital to future revenue collections.

With regard to another important subject, local government taxes, Georgio Brosio, of Turin University in Italy, provides in Chapter 12 a comprehensive account of local taxation in an international context. His paper shows that, despite the inherent difficulties of defining and administering local taxes, the share of local revenues in industrial countries has risen in recent years. One factor in this rise was the emergence of local business taxes, especially in Germany and France, which had complemented traditional property taxes. He argues that successful local taxes should strictly adhere to the benefit principle—linking revenue burdens to observed local services provided—and should belevied on as broad a base as possible to minimize distortions to the local economy.

During the conference, Yu Zichong of the Shanghai Municipal Tax Administration reviewed practices in local taxation in China and possible reform directions. His independently derived conclusions echoed Brosio’s comments. In particular, he remarked on the fragmented nature of the current tax system, which makes it difficult to administer. He argued that tax bases should be based on the benefit principle, made consistent within jurisdictions, but not necessarily across jurisdictions (since some diversity is desirable), and be broadened via the assignment of taxes, such as the local business tax, the interest income tax, and natural resource taxes to the local level. He saw a need for a division of tax powers between central and local governments—with local governments given greater freedom to move into certain areas, not utilized by the central government, and to set their own tax rates and determine the relevant bases. In this context, he suggested that it would be desirable for personal income tax revenues to be shared between central and local governments. He endorsed the development of separate national and local taxation administrations and urged the development of self-assessment and withholding taxes.

In Chapter 13, Manuelyan Atinc and Bert Hofman deal with the issue of taxation of natural resources. The authors argue that the “economic rent” generated by the development of the resources—defined as the market value of the resource exploited minus all costs necessary for exploitation—should accrue to the government, and that special taxation measures may be needed to achieve that result. The paper canvasses the taxation options available—ranging from a pure resource rent tax and cash flow taxes to royalties, auctioning, and production-sharing arrangements. Atinc and Hofman express a preference for a modified cash flow tax and suggest that China may wish to consider implementing such a tax. They also argue that the immobility of the tax base makes natural resource taxes a potential candidate for local government taxation. At the very least, this tax base might be shared with the central government, in a manner that would allow local governments to cover development of local infrastructure, associated with resource development, and environmental clean up costs after project completion and implementation.

Discussion of the issues raised by these papers was lively. One question was the extent to which the taxation reform would bring China closer to international experience and the relative benefits that it might bring for economic management and accountability by various levels of government. Another controversial issue was the advocacy of cash flow taxation. Tanzi noted that whatever their theoretical merits, such taxes were as yet untried and there was a question therefore whether China would wish to place itself in the vanguard of experimentation with such taxes, especially in view of possible revenue loss in the short to medium term. The Chinese participants also expressed reservations about the applicability of such taxes at this time and, more generally, questioned the view that potentially complex natural resource taxation issues could be adequately handled at the local government level.

Grants and Transfers

The session on Grants and Transfers opened with a selective review of international experience with intergovernment grants and transfers by Anwar Shah (Chapter 14). The paper lists four main objectives of such arrangements: (1) correcting for fiscal gaps—often labeled as vertical imbalances—in the availability of resources to different levels of government; (2) encouraging the use of national minimum standards of service by subnational governments; (3) corrections of potential underprovision of some services at subnational level where the benefits of programs spill over into other jurisdictions; and (4) the presence of differential fiscal capacities between different subnational governments—which may impede a subnational government’s ability to provide a reasonably comparable level of service at standard levels of taxation. He also lists a number of desirable design characteristics of grants and transfers.

Shah’s review of practices in selected developing countries suggests that the actual practices followed often fail to meet the objectives and design principles he outlines. In order to help policymakers avoid mistakes, he sets down a decision framework that links the type of arrangement to be followed to the policy objective set; revenue sharing to deal with fiscal gaps; conditional nonmatching grants to deal with minimum standards; open-ended matching transfers to deal with benefit spillover; and unconditional block grants using equalization criteria to deal with horizontal imbalances in fiscal capacities.

After the conference, a technical assistance team from IMF’s Fiscal Affairs Department focused on institutional and information requirements for the establishment and implementation of a comprehensive system of grants and prepared a report. The report (Chapter 15) Ehtisham Ahmad, Jon Craig, and Dubravko Mihaljek, is included in the volume to complete the policy options that would need to be considered by the authorities. The paper points out that a comprehensive system of grants would include both special purpose and equalization grants. In some countries, there is an independent Grants Commission, as in Australia, whereas in others, such as Canada, the equalization exercise is carried out by the Ministry of Finance. The methods of estimating equalization grants could be restricted to an assessment of revenue capacities of different regions, as in Canada, to one which includes both assessments of revenue capacities as well as differential expenditure needs. The information requirements differ with the method of estimation chosen.

A paper by Jon Craig (Chapter 16) deals solely with the techniques used by governments to address their horizontal equity objectives. It provides an analytical framework for designing grant arrangements, including the quantification of the vertical and horizontal balances within the system of intergovernment finances. The policy efforts to correct one imbalance can often have implications for the resolution of the other. Craig also refers to a number of case studies.

Sequencing of Reform

The conference introduced a full range of issues and options relating to intragovernmental fiscal relations in a country the size of and as complex as China, with references to international experience. It is our hope that the discussions have served to highlight the options for reform and the sequencing of the steps to be taken in the short to medium term.

The devolution of expenditure responsibilities in China is likely to be maintained and could even be strengthened. At the same time, greater central control is likely to be exercised over the revenue base. Typically, once expenditure assignments are determined, the question of revenue assignments can be addressed in a meaningful fashion. Indeed, this principle was reflected in the structure of the conference. There was disagreement, however, about whether further adjustments to the current expenditure assignments were needed prior to the determination of the revenue assignments and of transfers. At a closed session, Ahmad took the view that the changes proposed by the authorities on the revenue side were fundamental to the reform process and should not be delayed to await further refinements in expenditure assignments—especially since this is likely to be a lengthy process involving a changing role for state-owned enterprises. Thus, a transfer system would be developed on the basis of current expenditure assignments, and this could be adjusted as expenditure assignments change over time.

A crucial feature in the reform of intragovernmental fiscal relations in China concerns the establishment of a separate “State Administration of Taxation.” The institution of the administration based on modern tax administration functions, the likely time frame involved, and whether the officials of such a service can actually be insulated from “local” pressures and influence are among the most important questions for the reform.


Ahmad and Tanzi, International Monetary Fund, Gao Qiang, Ministry of Finance, China.


As described in Chapter 2, the revenues from the natural resource tax are to be assigned to the province in which the resource occurs, although the revenues from offshore oil will be assigned to the central government. While the natural resource tax is described as being “shared,” the sharing is effectively an assignment of different taxes from different types of resources to different levels of government. A transfer tax on share transactions will also be shared, but this is unlikely to generate significant revenues.

    Other Resources Citing This Publication