The System of Equalization Transfers in China

The Third Plenum of the 14th Central Committee of the Chinese Communist Party (CCP) in November 1993 decided in principle for a comprehensive reform of central-provincial fiscal relations. Soon after the Plenum, the central government announced that the new fiscal system, known as the tax-assignment system (fenshuizhi), would be implemented nation-wide in 1994. With the aim of providing adequate revenues for government, particularly the central government, by revamping central-provincial revenue-sharing arrangements, the reform is to “[change] the current fiscal contractual responsibility system of local authorities to a tax assignment system …” and to “gradually increase the percentage of fiscal income in the gross national product (GNP) and rationally determine the proportion between central and local fiscal income.”

This study investigates water management systems at fortresses and cities built by the Urartian Kingdom, focusing on questions such as how water was supplied to settlements and where it was stored. The Urartians often preferred to build their settlements close to freshwater sources in the plains and at the intersections of main roads along the rocky ridges and on top of high rocky outcrops in the mountainous terrain. Although Urartian royal inscriptions state that canals and fountains were built to meet the water needs of the newly established settlements, texts do not provide sufficient information on water supply and storage systems at the settlements. Considering the locations where Urartian fortresses and lower city settlements are founded, we would expect to see cisterns and similar water storage facilities at these habitation centers. However, with a few exceptions, no such structure has been documented by archaeological investigations at Urartian sites. Despite the absence of cisterns, on the other hand, multiple storage rooms with sunken pithoi have been unearthed at almost every excavated Urartian settlement. In Urartian studies, storage buildings with pithoi are generally regarded as storage units dedicated to the storage of wine and other liquid commodities. In this study, we suggest that the storage rooms with pithoi were also used for storing drinking water for the communities inhabiting Urartian fortresses and lower towns.

A model of design thinking from the 4E prospect of enactive, embodied, embedded and extended mind. Enactive thinking as visible and tangible. Rittel’s critique as system failure, not wickedness. Being as Making, Knowing, and Believing, ontologically unique ways of engaging the world, each with their own measures. The classical reality of problem-solving enfolded into design thinking’s reality of experience. Enactive design thinking as the creative engagement of extended mind in the entanglement of characterized ecological information and probable actions. Examples from literature, popular culture, and performance. The realization of “designerly” thinking’s educational promise and Schön’s “learning society” in an enactive reality of making.

University International Studies Program Working Paper 03-12 July 2003 The System of Equalization Transfers in China Zhihua Zhang Jorge Martinez-Vazquez Georgia State University Andrew Young School of Policy Studies The System of Equalization Transfers in China Working Paper 03-14 Zhihua Zhang Jorge Martinez-Vazquez July 2003 International Studies Program Andrew Young School of Policy Studies Georgia State University Atlanta, Georgia 30303 United States of America Phone: (404) 651-1144 Fax: (404) 651-3996 Email: ispaysps@gsu.edu Internet: http://isp-aysps.gsu.edu Copyright 2001, the Andrew Young School of Policy Studies, Georgia State University. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means without prior written permission from the copyright owner. International Studies Program Andrew Young School of Policy Studies The Andrew Young School of Policy Studies was established at Georgia State University with the objective of promoting excellence in the design, implementation, and evaluation of public policy. In addition to two academic departments (economics and public administration), the Andrew Young School houses seven leading research centers and policy programs, including the International Studies Program. The mission of the International Studies Program is to provide academic and professional training, applied research, and technical assistance in support of sound public policy and sustainable economic growth in developing and transitional economies. The International Studies Program at the Andrew Young School of Policy Studies is recognized worldwide for its efforts in support of economic and public policy reforms through technical assistance and training around the world. This reputation has been built serving a diverse client base, including the World Bank, the U.S. Agency for International Development (USAID), the United Nations Development Programme (UNDP), finance ministries, government organizations, legislative bodies and private sector institutions. The success of the International Studies Program reflects the breadth and depth of the in-house technical expertise that the International Studies Program can draw upon. The Andrew Young School's faculty are leading experts in economics and public policy and have authored books, published in major academic and technical journals, and have extensive experience in designing and implementing technical assistance and training programs. Andrew Young School faculty have been active in policy reform in over 40countries around the world. Our technical assistance strategy is not to merely provide technical prescriptions for policy reform, but to engage in a collaborative effort with the host government and donor agency to identify and analyze the issues at hand, arrive at policy solutions and implement reforms. The International Studies Program specializes in four broad policy areas: Fiscal policy, including tax reforms, public expenditure reviews, tax administration reform Fiscal decentralization, including fiscal decentralization reforms, design of intergovernmental transfer systems, urban government finance Budgeting and fiscal management, including local government budgeting, performance-based budgeting, capital budgeting, multi-year budgeting Economic analysis and revenue forecasting, including micro-simulation, time series forecasting, For more information about our technical assistance activities and training programs, please visit our website at http://isp-aysps.gsu.edu or contact us by email at ispaysps@gsu.edu. The System of Equalization Transfers in China Zhihua Zhang ∗ and Jorge Martinez-Vazquez∗∗ Introduction In 1994 in its push toward the establishment of a market economy, China introduced a fundamental reform of its tax and intergovernmental fiscal relations systems, which became known as the “tax-sharing system” or TSS reform. The fundamental goal of the TSS reform was to improve the behavior of the "two ratios"----the central government share in total revenues of the public sector, and the share of public revenue in GDP. After more hesitation, the 1994 TSS reform has been successful in achieving those goals. All in all, the TSS can also be credited with facilitating the rapid rate of economic growth the Chinese economy has experienced over the last decade. However, not all the consequences of rapid growth have been positive ones. One of the consequences of the rapid growth of the economy has been increasing regional fiscal disparities, especially between the coastal provinces and the inland provinces. How to achieve a better balance in the territorial distribution of fiscal resources has become a first policy priority for the Chinese government. The 1994 TSS reform concentrated in reforming China’s tax sharing system, but it also introduced an equalization transfer system. The focus of this paper is to describe the many components of the equalization system and analyze its strengths and weaknesses. We ∗ Ministry of Finance of China ∗∗ Andrew Young School of Policy Studies, Georgia State University We are thankful to Li Zhang for very able research assistance 1 2 International Studies Program Working Paper Series ask several fundamental questions: Is the current system of equalization transfer adequate for reducing current horizontal fiscal disparities? Does the system yield the correct incentives for revenue mobilization and expenditure efficiency of sub-national governments 1 ? Even if the equalization system worked adequately across provinces, what is to be done by the central government to ensure equalization at the sub-provincial level? The paper is organized as follows. The first section reviews the current framework for tax sharing. Revenue sharing, after all, is nothing else than a particular form of transfers in practically every country, but in China the links between the tax sharing system and the system of transfers is particularly strong. Section two of the paper describes the current system of equalization transfers. Section three assesses the current system of transfers in light of first principles and international best practices. Section four offers several conclusions and recommendations. 1. The current tax sharing system in China 1.1 A brief overview of recent developments in intergovernmental fiscal relations Since China began opening to the outside world and to reform its economy in the late 1970s, the reform of the fiscal system has taken a lead role in the overall process of reform. In the 1980-1984 period, the government introduced a system with separate revenues and expenditure assignments to the central and local governments. In the 1985-1987 period, revenue assignments were changed to a system in which revenues were classified into own or fixed central revenues, own (or fixed) local revenues and shared revenues between the central and local governments. The base amounts for the revenues and expenditure of the different provinces in this period were determined according to the levels of real terms reached in the year previous to the reform. During this period, when the base amount for expenditures was larger than the own (or fixed) local revenues, then the province was allowed to keep all the fixed revenue. In addition, this type of province was entitled to shared revenues according to the formula: Local government ratio for shared revenues in the province = (base amount for expenditure in the province - base amount for fixed revenue in the province) / base amount for shared revenue in the province [1] When the base amount of expenditure in the province was less than its base amount of local fixed revenue, then the province had to remit to the central government funds equal to the gap between its base amount for expenditure and the 1 By subnational governments we refer to the provinces but also all the sub-provincial governments: prefectures and cities, counties, and townships. The System of Equalization Transfers in China base amount for own or fixed revenue every year. In addition, the entirety of the (central-local) shared revenue base in this province had to be remitted to the central government. When the base amount for expenditure in the province was greater than both the base amounts for its fixed revenue and all shared revenues, then the province could keep the entirety of both revenue sources. Furthermore, the gap between the base amount for expenditure and that for local fixed revenue and shared revenue in the province was covered by the “fixed amount” grants from the central government every year. For minority regions, the “fixed amount” grants could be increased from year to year at annual rate of 10 percent. This system of revenue sharing and transfer determination was very similar to that being practiced in the former Soviet Union with "regulating" revenue sharing rates and gap-filling transfers.2 In the 1988-1993 period, the government implemented what became known as the “fiscal contracting system.” In reality, several modalities of the “fiscal contracting system” were adopted during this period. The most important modalities include the following: A. “Contracted sharing rate with fixed yearly growth rate of revenue” In this system, the central-local revenue sharing rate and the yearly growth rate of local revenues were on the revenue performance of the province over recent years and negotiated by the central government and the province. In the practical implementation, if the real growth rate of revenues was equal to the contracted fixed yearly growth rate for the province, both the central government and the provincial government could share the total revenue collected within the province according to the contracted sharing ratio. If the real growth rate of revenue was greater than the contracted fixed yearly growth rate in the province, then both the central government and the provincial government would share the total revenue by the following formulae, respectively: Central government shared revenue in the province = revenue in the province in previous year ∗ (1+contracted yearly growth rate of the province) ∗ contracted central shared ratio in the province [2] Local government shared revenue in the province = revenue in the province in previous year ∗ (1+contracted yearly growth rate of the province) ∗ contracted local shared ratio in the province [3] 2 See form example Jorge Martinez-Vazquez and Jamie Boex “Fiscal Decentralization in the Russian Federation,” Washington, DC: World Bank, 2000. 3 4 International Studies Program Working Paper Series In addition, the province could keep all the (increased) revenues above the amount calculated by the contracted fixed yearly growth rate. This latter was based on the following formula: Increased revenue over contracted yearly growth rate = increased revenue in the province in current year ∗ (contracted yearly growth rate of the province – real yearly growth rate in current year) [4] If the real growth rate of revenue was less than the contracted yearly growth rate in a province, then both the central government and the provincial government could share the total revenue within the province according to the contracted fixed shared ratio. However, the province had to make up the decreased part of central shared revenue (compared with the contracted yearly growth rate) resulting from the gap between the real growth rate and contracted fixed growth rate of the revenue within the province. In total, ten provinces China adopted this system. B. “Fixed shared rates in total revenue” Under this system, the central government and the provincial government shared the total revenue in the province according to the local government fixed sharing rate. The Sharing rate was determined on the basis of a base amount for total expenditure and a base amount for total revenue, The following formula was used: Local government fixed shard ratio of total revenue in province i = base amount for expenditure in province i / base amount for total revenue in province i [5] Note, of course, that the sharing rate generally varied from province to province. Compared to system A, this system had no requirement of a contracted yearly growth rate. In total, three provinces adopted this system. C. “Fixed sharing rate in total revenue plus incremental fixed sharing rate” In this system, besides sharing total revenue collected in the province on the basis of a fixed shared ratio as in B. above, the province could share the incremental part of revenues at a different sharing ratio. In total, three provinces adopted this system. D. “Contracted remittance with fixed annual growth rate” Under this system, the province should remit to the central government a fixed amount per year plus a variable amount determined by a fixed yearly growth rate contracted by the center and the province. Only two provinces adopted this system. The System of Equalization Transfers in China 5 E. “Fixed contracted remittance” In this system, the province should remit to the central government a fixed amount every year determined by the following formula: Fixed contracted remittance in province i= base amount for revenue in province i - base amount for expenditure in province i. [6] The provincial government kept all remaining revenue collected in the province. In all, three provinces adopted this system. F. “Fixed contracted grants” For all provinces whose base amount for expenditures was larger than the base amount for revenues, they got to keep all base revenues collected in their jurisdictions, In addition, these provinces got a fixed contracted grants from the central government every year which was theoretically decided by the following formula: Fixed contracted grants in province i= base amount for expenditure in province i - base amount for revenue in province i. [7] In total, sixteen provinces adopted the “fixed contracted grants” system. The reforms of the fiscal system in China since 1980 played an effective role in mobilizing revenues and improving budget efficiency at the sub-national level by increasing incentives to collect revenue by local governments as well as propel local economic development. The reforms during the 1980’s follow an incrementalist approach by each new system building on the lessons learned from the previous system. By large these systems achieved many of the goals set by the government at the time. However, because of the rapid pace of the reform in the rest of the economy, each of the approaches had a limited life and the fiscal system had to continuously adapt. In the case of the “fiscal contracting system” the outcomes became increasingly more inconsistent with the fiscal requirements for the establishment of the market economy. The most important limitations of the contracting system are discussed in the following paragraphs: Fiscal resources became too decentralized leading eventually to a vertical imbalance against the central government. The fiscal contracting system made the fiscal distribution between the central and local governments too rigid. Most of the 6 International Studies Program Working Paper Series turnover taxes which usually were the major source of revenue for which collections increased and increased at the fastest rates, were retained by local governments. The share of the central government in the total revenues dropped dramatically from 36 percent in 1986 to 22 percent in 1993. This caused serious fiscal difficulties for the central government. At the same time, and also quite worrisome because it was unexpected or un-programmed, overall tax effort in the country measured by the ratio of total government revenue to GDP decreased dramatically too. Under this fiscal contracting system, tax revenues were divided in accordance with the ownership (by level of government) of enterprises. Shortening revenues by mode of ownership had a long tradition in China with roots in the Soviet Budget tradition. The sharing according to ownership led to a number of following problems: The first problem was that this system made the fiscal interests of government dovetail with those of the enterprises. This encouraged the government to interfere in the operation of the enterprises and in making it difficult to separate functions of enterprises from government. The second problem was that this system did not help enterprises to gear themselves to the market economy. Stated-owned enterprises usually did not have pressure to maximize profit and minimize risk because in their minds, the government would never let them go bankruptcy. The third problem is that this system was detrimental to fair competition for other enterprises in the market since governments dealt with these enterprises very differently. The fourth problem was that the system provided local governments with incentive to pursue their own narrow fiscal interests on the development and often expansion of those enterprises yielding the highest level of taxes, such as distilleries and cigarette factories. This resulted in serious overlapping in construction and development and prevented protectionist policies. More importantly, this system negatively influenced the allocation in the currency of resources and slowed down the adjustment of the industrial structure. Finally, this system did not benefit tax collection. With the development of market economy, investors were becoming diversified and different more complex forms of enterprises were also introduced. These trends made tax enforcement a greater challenge and contributed to make the ownership position criterion obsolete. The contracting system was difficult to administer. As there were a great variety of fiscal contracted systems in use, calculations for the central government were complex. Besides, this system was easily and greatly affected by ad hoc factors. There was overall a significant role for political negotiations between the central government and individual provincial. In all, the system also led to greater regional fiscal disparities. Since fiscal contracted systems were not unique across provinces and were chiefly the results of negotiation between the central government and provincial governments, fairness was hard to achieve. Provinces with the same fiscal position usually were dealt differently. In addition richer more developed provinces in the eastern - coast provinces like Canton, Shanghai and Shandong etc usually were able to have more advantageous The System of Equalization Transfers in China 7 contracts because of the development strategies and the irrelative political power, western provinces. Under this fiscal contracting system, the assignment of revenue between the central and local governments was not clear. As a result, it was common for local governments to try to appropriate central revenues. This problem was compounded by the fact that tax administration remained quite weak and with dual de facto subordination of tax officials to local authorities. As a result, tax administration, especially the rights to grant tax exemptions because excessively decentralized and in fact abused by many local governments. Some local governments, acting in their own interests and often ignoring central rules and regulations, reduced taxes or granted exemptions without proper authorization. Some local governments negotiated tax bills at will with enterprise, and those agreements usually lasted, 3 to 5 years. Competition to step up economic development in the provinces led a to serious declines in revenues. These problems with the “fiscal contracted system” required a profound reform of China’s fiscal system, This came with the reform of 1994. 1.2 Current intergovernmental fiscal relations: the tax sharing reform of 1994 and Posterior Fine Tuning The tax-sharing system (TSS) reforms of 1994 had the overall goal to establish an efficient and reasonable fiscal allocation mechanism.. The Reform was based on several pragmatic principles: • • • the vested regional fiscal interests were to be respected the adjustment in regional fiscal interest would focus on incremental revenues the ratio of central revenues to total revenues and the ratio of total revenues to GDP would increase gradually The central government devised an actual strategy following several lines. First, it was deemed necessary to specify expenditure assignments, even though the TSS reform did not change expenditure assignment too much. This was the first time in the history of intergovernmental fiscal relations in China in which the central government clarify this important item in the central local government relations. Second, all taxes were classified into 3 categories: central taxes, local taxes and shared taxes between the central and local government. This allowed, among other things to create two separate tax administration, the first in charge of collecting central taxes (including shared taxes), and the second responsible for collecting local taxes. Third, a scheme was needed to guarantee the vested interests of local governments. This was accomplished by establishing the tax rebate system, which in fact guarantees the normal operation of local governments by holding them harmless with respect to the 8 International Studies Program Working Paper Series taxes they were receiving before the reform. Last there was a need for an equalization transfer system based on the assessment of fiscal capability and expenditure needs of the local governments. Below we expand on these themes. The assignment of expenditure responsibilities between the central and local governments. :The TSS reform did not change the assignment of expenditure responsibilities that by law and practice existed before 1994, See Table d. In this arrangement, central government is responsible for nation wide services including national security, international affairs and the general operation of the central government. Central government funds are also used for the restructuring of the national economy, the coordinated development of different regions, the implementation of macroeconomic policies, and the development of the undertakings such as universities, hospitals, research institutes, newspapers, publishing houses, theaters etc., directly under the control of the central government. Local governments are responsible for most other public goods and services including several important aspects of social security and welfare. Local governments are also responsible for the development of the local economy and the operation of universities, hospitals, research institutes, newspapers, publishing houses, theaters etc. This assignment of expenditure responsibilities present many problems, including lack of clarity, excessive overlapping, and the problematic assignment of important aspects of local security at the local level. However, this is not the place for an extensive analysis of current expenditure assignments. See, for example World Bank 2002. Revenue assignments between the central and local governments: In deciding the assignment of revenue sources, the government again used several general pragmatic principles. Those tax instruments that can be used in the pursuit of maintaining national objectives, such as the implementation of macroeconomic policy, were assigned as central taxes. Those taxes that could be interpreted as more closely related to economic development were assigned as shared taxes. Those taxes more suitable to be collected and administered by the local governments were assigned as local taxes. The specific revenue assignments are shown in tables 2 through 4. The System of Equalization Transfers in China 9 Table 1: The Assignment of Expenditure Responsibilities Between the Central and Sub-national Governments • Main expenditure responsibilities of the central government • Defense • Operation of the central government • • Foreign affairs Operational expenses for cultural, educational, scientific and public health undertakings at the central level • Key capital construction • Agriculture • Macro-economic control and regional coordination of economic development • Technical renovation and new product development in centrally owned enterprises • Subsidies • Social security • • • Debt Main expenditure responsibilities of subnational governments Operation of local governments Operational expenses on cultural, educational, scientific and public health undertakings at the local level • Local capital construction • Agriculture • Social security • Fund for technical renovation and new product development in locally owned enterprises • Urban maintenance and construction Subsidies 10 International Studies Program Working Paper Series Table 2: Assignment of Revenues at the Central Level • Central revenues • Customs duties • Excise tax • VAT on imports Enterprise income tax (From: rail transportation, state post, 4 state-owned commercial banks and 3 state-owned policy • • • banks, and enterprises of offshore oil and natural gas) Business tax, urban maintenance and construction tax From: the headquarters of banks, and insurance corporations, and the Ministry of Railroad) Profit remittances by all centrally owned enterprises Export rebates of enterprises engaged in foreign trade Table 3: Shared Revenues between the Central and Sub-national Governments • Central: Subnational shares • VAT (75:25) • Personal income tax (50:50 in 2002; 60:40 from 2003) • Stamp tax on security transaction (97:3) • Enterprise income tax excluding special items (50:50 in 2002; 60:40 from 2003) Resources tax (offshore; 100% central and on land, 100% local) The System of Equalization Transfers in China 11 Table 4: Assignment of Revenue at the Subnational Level • • • • Subnational revenues Business tax (other than from the headquarters of banks, and insurance corporations, and the Ministry of Railroad) Urban and township land use tax Urban maintenance and construction tax (other than from the headquarters of banks, and insurance corporations, and the Ministry of Railroad) • Property tax • VAT on land • Agricultural and animal husbandry tax • Contract tax • Profit remittances by all locally owned enterprises • Gift and bequest tax • Reorientation tax on capital construction • Vehicle and vessel utilization tax • Stamp tax • Tax on special produces • Tax on the occupation of arable land • Revenue from compensation for use of state-owned land • Slaughter tax Other revenue The revenue assignments introduced in the TSS reform in tables 2 to 4 reflect the current revenue assignments. By comparison to 1994 when the TSS was introduced, several changes have taken place. Two of these changes deserve special mention. First, the central government sharing rate for the stamp tax on securities transactions has increased from 50 percent in 1994 to 97 percent in 2002. The central government used two main reasons for this change. First, the stamp tax on securities transactions, which is collected in Shanghai and Shenzhen stock markets, while the resources involved come from the rest of the country. Second, Shanghai and Shenzhen, as the richest cities in China, should not benefit from stamp tax revenues beyond what is necessary to compensate for the fact that they provide some services for the two stock markets. We must note that this change in revenue assignments was phased-in over time so to enable those two local governments to better balance their budgets. In 1997, the sharing rate for the central government increased from 50 12 International Studies Program Working Paper Series percent to 88 percent; then to 91 percent in 2000, 94 percent in 2001; and 97 percent in 2002. Second, the sharing of income taxes was amended to provide the central government with a higher tax share, with an implementation date of January 1,2002. The main justification for this reform has been to provide the central governments with additional revenues so to increase the pool of funds for the equalization transfer system. The plans are for the central government to allocate all incremental revenues resulted from this reform to local governments, mainly those in the central and western provinces middle-western regions, via the equalization grant system; It should be noted that a hold harmless provision will be in place so that local vested interests from income taxes are respected in the new regime. The central government regarded 2001 as the base year for the reform. If the base amount for local shared income tax revenue based on the shared ratio after the reform in a province is less than its real income tax revenue in 2001, the gap will be covered by an income tax rebate from the central government to the province. Under the new regime, all revenues from the income tax on enterprises, except for a list of enterprises3, and all revenues from the personal income tax are shared by the central government and local governments4. In 2002 the central government sharing rate and that of local governments has been kept at 50 percent for each. But in 2003,the central government sharing rate will increase to 60 percent while the local governments sharing rate will decrease to 40 percent of the total shared income taxes. The central government has not yet decided whether it will continue to increase its sharing rate after 2003. Reform of revenue assignments at the sub-provincial level: Sub-provincial fiscal relations in China remain highly hierarchical and encompass the relations between 3 These enterprises include rail transportation, state post office, Industrial and Commercial Bank of China, Agriculture Bank of China, Bank of China, Construction Bank of China, State Development Bank, China Bank of Agriculture Development, Import and Export Bank of China, enterprises of offshore oil and natural gas, China Petroleum and Natural Gas Co. limited and China Petroleum Chemical Co. limited 4 Note that previous to the 2002 reform, the enterprise income tax from centrally-owned enterprises, local banks and non-bank financial institutions were assigned exclusively to the central government, as was the personal income tax from interest of deposits. On the other hand, the enterprise income tax from locally owned enterprises, collective enterprises, joint ventures and individual enterprises were exclusively assigned to local governments, as was revenues from the personal income tax other than from interest from bank deposits. The enterprise income tax from enterprises owned (financed) by both the central and local governments was shared by both levels of government, with the sharing rates determined by the ownership shares of each. The System of Equalization Transfers in China 13 provincial governments and prefecture governments, between prefecture governments and county governments and between county governments and township governments5. After the TSS reform in 1994, almost all provincial governments adjusted their sub-provincial fiscal relations. From the start, most provincial governments paralleling the TSS reforms between the central and provincial governments also centralized fiscal revenues to some extent at the sub-provincial level. The degree of centralization actually varied across regions reflecting differences in regional economic development, revenue structures, and the degree of fiscal difficulties the regions faced. The different reforms adopted at the provincial level vis-à-vis prefecture governments can be summarized into three types: A. Tax sharing system Under this system, local taxes are divided between the provincial government and prefecture governments according to provincial taxes, prefecture taxes and provincial-prefecture shared taxes. Through tax sharing, provincial government usually controlled the major sources of local revenue. Together, 14 provinces are implementing this system. B. Tax sharing plus the sharing of the tax rebate: Under this system, the provincial governments do not only centralize part of the shared taxes, but they also share part of the tax rebate from the central government to the prefectures. In all 12 provinces adopted this system. C. Tax sharing plus the sharing of incremental sub-provincial revenue: Under this system, both provincial and prefecture governments had their own fixed taxes, but for the incremental amount of prefecture revenue (above the base amount), the provincial government got a share according to a contracted fixed ratio. Only two provinces adopted this system. Not surprisingly, given the numbers involved, the changes in the fiscal relations between prefecture governments and county governments showed even more complex characteristics. Much less is know about the changes in revenue assignments between the county and township governments. It appears that often counties have 5 However, for 5 regions including Beijing, Tianjin, Shanghai, Hainan and Ningxia, the provincial governments deal directly with county governments because there are no prefecture governments in these regions. There are also some exceptions, where in a few other provinces like Zhejiang, the provincial governments also deal directly with county level governments, even though they have prefecture level governments. 14 International Studies Program Working Paper Series replicated for the townships the changes they experimented vis-à-vis their prefectures. Several types of arrangement have emerged here: • • • The "following type", because the prefecture governments completely followed the system that the provincial governments had introduced vis-à-vis prefecture governments. In other words, the prefecture governments neither further centralized fiscal resources, nor provided further assistance in the course of the adjustments of sub-prefecture relations. The second favoring type, prefecture governments provided assistance to county governments by reducing the amount of remittances county governments were obliged to send up. The “centralization type”, the prefecture governments centralized fiscal resource from county governments by variety of methods. Several studies by the Ministry of Finance of China (reports on sub-provincial fiscal relations written by budget department, MOF, PRC in 2001) found that prefecture governments in less-developed areas and areas with a higher share of agriculture production in GRP practically did not centralize at all from county governments. Several explanations are possible for this. These countries have lower level of economic development, smaller size revenues, simpler revenue structures and a higher share of incremental amounts from VAT and excise taxes being remitted to the central government. On the other hand, these same prefecture governments did not provide much assistance to their subordinate counties. At the opposite side, prefecture governments in more developed regions provided fiscal assistance to counties with weak fiscal capabilities while at the same time centralized certain fiscal resources from counties with strong fiscal capabilities with the goal of reducing the fiscal gap across counties. An emerging trend in more recent years has been for provinces to adjust sub-provincial fiscal relations, favoring prefectures and counties facing fiscal difficulties, in order to further reduce fiscal disparities, and mobilize revenues, prefectures and county governments. Two sorts of scenarios are typically encountered. One, which is more common in less-developed provinces, provincial government and to some extent prefectures have reduced their sharing rates in shared taxes have cut remittance ratios, or have adjusted the amount of remittance and fixed grants, mainly to ensure minimum fiscal needs of county governments. As a result, fiscal capability in many low-income counties has been enhanced. For instance, Sichuan province adjusted sub-provincial fiscal relations in 2000; as a result, the gap of per capita revenue (after equalization) in the richest prefecture to the poorest prefecture decreased from 3.25 folds in 1999 to 2.97 folds in 2000. The other scenario, more commonly observed in more-developed provinces is one of centralizing resources at the higher level of government with the objective of further consolidating budgets and generating resources to implement gradual equalization at the sub-provincial level. The content of the adjustment may include increases in the type of taxes and the sharing rates specified for provincial governments. For instance, since 1996,the The System of Equalization Transfers in China 15 Canton provincial government has assigned the business tax. The income tax from non-state-owned enterprises, the personal income tax, and VAT on land, as provincial - prefecture shared taxes. In addition, taxes from special industries with concentrated tax bases have been reassigned as provincial taxes. 1.3 Some reflections on the current tax sharing system The TSS reform would appear to have introduced more structure in China’s system of intergovernmental fiscal relations and at the same time it has simplified it. Consider the following establishment of the TSS replaced the 6 types of fiscal contracted systems with a single unified system. The clearer assignment of taxes between the central government and sub-national governments has made the boundary lines of fiscal interests much more transparent, and as a result misappropriation of each other’s revenues between the central and sub-national government has significantly decreased. The clearer assignment of revenues has helped to tighten the budget constraints at all government levels. In addition, following the new framework for tax sharing system between the central government and provincial governments, revenue assignments at the sub-provincial level have been reformed and the ability of provincial governments to exert fiscal control has improved significantly. The new TSS also has addressed some perverse incentives in the previous system and in that way it has contributed to a more rationale allocation of resources and sounder development practices by sub-national governments. For example under current tax sharing system, excise taxes are assigned to the central government while business taxes are assigned to sub-national governments. Thus the incentives for local governments for focusing blindly on developing enterprises with higher tax returns such as distilleries and tobacco are gone. Sub-national governments have currently stronger incentives to improve efficiency in the allocation of resources and develop their economies in a more balanced way. Another improvement associated with the TSS reform has been a better control of tax reductions and exemptions. This process has been disciplined and its administration has been relatively centralized. These measures in TSS now prevent local governments from reducing or exempting tax revenue at will and also have dramatically decreased the misappropriation of central tax revenues. In fact, the TSS reform revoked all tax reductions and exemptions granted by provincial governments for the turnover tax in the past. The current practice is that tax reductions and exemptions approved by the central government need to be reported in a separate schedule of the tax return. These measures seem to have led to significant decreases in the amounts of tax reductions and exemptions actually granted. 16 International Studies Program Working Paper Series The separation of tax administrations for central and local governments has reduced (although not completely eliminated) the interference of local government authorities in tax administration and collection of central and shared. A more professional and neutral tax administration has also contributed to improve the institutional environment for a more fair competition among enterprises. . The TSS reform standardized procedures for tax collection and enforcement by the tax administration. How has the TSS reform performed in terms of the objective of increasing the “two ratios”? Overall and even tough it took several years, the TSS reform delivered both objectives. By 2000, government revenues were RMB Yuan 1339 billion, of which, central revenues 698.9 billion, and local revenues 640.6 billion. As shown in Table 5, the ratio of total government revenues to GDP increased from 11.2 percent in 1994 to 15 percent in 2000, reversing a declining trend of this ratio since 1980. The ratio of central revenues to the total government revenues increased from 22 percent in 1993 to 52 percent in 2000, quite radically changing the historical situation in which central expenditure mainly relied on remittances from local governments. The reversal in trends is noticeable in Figure 1. The enhancement of central revenue under the TSS is providing the means to strengthen the equalization transfer system discussed in the next section. An important objective of the TSS reform in China has been to keep the proportion of centrally collected revenue in total revenue at about 60 percent while keeping the share of the central government in total expenditure at about 40 percent, so that the central government can allocate about 20 percentage points of the consolidated government budget to transfers. Table 5: Revenues collected by the central and local governments (100 million Yuan) 1994 1995 1996 1997 1998 1999 2000 5218.1 6242.2 7408.0 8649.6 9875.9 11444.1 13395.2 central 2906.5 3256.6 3661.1 4226.9 4892.0 5849.2 6989.2 Collected by local 2311.6 2985.6 3746.9 4422.6 4983.9 5594.9 6406.1 Central to total (percent) 55.7 52.2 49.4 48.9 49.5 51.1 52.2 Local to total (percent) 44.3 47.8 50.6 51.1 50.5 48.9 47.8 46759.4 58478.1 67884.6 11.2 10.7 10.9 Total revenue • • Collected by GDP Total revenue/ GDP Source: Ministry of Finance 74462.6 78345.2 82067.4 11.6 12.6 13.9 89403.5 15.0 The System of Equalization Transfers in China 17 Figure 1: Performance of the “Two Ratios” 60 55. 7 52. 2 52. 2 49. 4 48. 9 50 49. 5 51. 1 38. 4 40 33. 8 25. 7 30 22. 4 20 24. 5 22 15. 8 12. 6 11. 2 10. 7 10 0 1980 14 15 1999 2000 10. 9 11. 6 12. 6 1985 1990 1993 1994 Rat i o of r ev. t o GDP 1995 1996 1997 1998 Rat i o of cent r al r ev. t o t The 1994 TSS reform represented a significant improvement in the design of revenue assignments in China. However, left several important problems unsolved. Local governments in China still lack a meaningful degree of tax autonomy. With the very limited exceptions local governments lack discretion to change the rates or define the base of local taxes or to decide whether or not to adopt taxes from the list of taxes assigned to them.6 In contrast to the lack of formal revenue autonomy, local governments to acquire or exercise “informal” revenue autonomy often by introducing unauthorized or “illegal” fees. The fees represent the main source of revenues for extra-budgetary funds, which in turn represent a significant part of local budgets. The proliferation of “illegal” fees at the local level became a concern for the central government because of the distortions they cause ad their relative unfairness. Another issue is that the TSS reform left revenue assignments at the sub-provincial level unspecified. This provides provincial governments with budget flexibility but it deprives local governments of revenue autonomy, reduces intergovernmental accountability, cuts revenue predictability of local governments, and reduces incentives for revenue-mobilization. The sharing between the central and local governments of the VAT and CIT, among other taxes, as stated in the TSS reform, left several important issues unresolved. The most important of them has been the unfair distribution or apportionment of revenues between local jurisdictions given that revenues have been paid almost exclusively in the jurisdiction where the headquarters of the enterprise are located. The central government has taken steps 6 Local tax autonomy is restricted to the selection of the rate of the land use tax within maximum and minimum legislated rates, and the choice of introducing (or not) the entertainment tax and the slaughter tax. 18 International Studies Program Working Paper Series recently to improve the apportionment of CIT revenues by using a formula. The solution to the unfair apportionments of VAT revenues has led to other countries in transition, like Russia, to completely centralize the VAT. The TSS reform also kept in place the “ownership” criterion for the sharing of taxes among different levels of government., providing local governments with significant incentives to develop and own enterprises purely involved in private market activities. 2 The Equalization transfer system in China and the tax rebate 2.1. Overview of the transfer system Under current TSS framework, there are three types of grants from the central government to local governments: • • • The tax rebate. General purpose grants, Special purpose grants As shown in Figure 2, in 2000 the total amount of central government grants to local governments was RMB Yuan 405.8 billion, of which, general purpose grants were RMB Yuan 95.2 billion, special purpose grants RMB Yuan 89.9 billion, and the tax rebate RMB Yuan 220.7 billion. It could be argued that the tax rebate is a special form of tax sharing and that as such should be discussed in the previous section. However, because the tax rebate has important implications for equalization transfers, it is more revealing if the tax rebate is discussed here. As for special purpose or earmarked grants, although they are very important in China’s system of intergovernmental fiscal relations, they are beyond the scope of this paper and they will not be discussed any further in this paper. Figure 2: Structure and size of grants from the central to the local (Total: ¥ 405.8 billion in 2000) Earmarked grants 89.9 billion Equalization grants 8.5 billion Equ. grants for minority regions 2.5 billion Grants for general Tax return220.7 purpose grants 83.8 billion increasing wage standards of civil servants and 2.2 The tax rebate billion welfare levels of social benefits Pre-tax-sharing grants 11.4 billion recepients Other grants 21.4 billion The System of Equalization Transfers in China 19 The tax rebate system was not only a key element of the TSS reform in 1994 but since it has become the largest component of the intergovernmental transfer system, the tax rebate has been the focus of attention for all levels of government in China. The tax rebate was the outcome of a compromise in the design of the TSS reform, which grandfathered the vested interests of local governments prior to the reform. The essence of the tax rebate is to send back to the provinces the amount of VAT and excise taxes that otherwise would have gone to the sub-national governments with the institutions prevailing before the TSS reform in accordance with the derivation principle. The goal of the tax rebate was to smooth out opposition or other difficulties with the tax reform and also ensure its smooth implementation. The tax rebate uses 1993 as the base year. The rebate for 1993 was determined by the difference between the base amount for some local revenue items and the sharing of some revenue items between the central government and local governments. After 1994, the amount of tax rebate was initially let to grow annually from its base amount in 1993, with a rate of growth equal to the national average growth rate of VAT and excise taxes in a proportion of 0.3 to 1. That is, if for example if the national average growth rate of VAT and excise taxes is 1 percent, then the tax rebate from central to local governments is let to increase by 0.3 percent7. In order to give local governments incentives to further develop their economies and expand their tax bases, and also to provide support to the central taxation administration in enforcing tax collection, the central government decided in August, 1994 to change the rule on the growth rate of the tax rebate by basing it on the growth rate of VAT and excise taxes in each locality (rather than national averages), but still according to the proportion of 0.3 to 1. Thus the formula for the tax rebate for one province is given by: TRi = TRi-1 ( 1 + Ri ∗ 0.3 ) Where TRi is tax rebate to region i in current year; TRi-1 is tax rebate to region i in previous year; Ri is the growth rate of VAT and excise tax collection in region i in current year; i = 1, 2, 3, *, n (n=36) 7 The TSS had provisions for what to do if VAT and excise revenues decreased in some provinces. But this situation has never happened. 20 International Studies Program Working Paper Series The Calculation of the Base Year Tax Rebate The base year of the tax rebate in 1993 is the net remittance from local to central governments. The net amount of local remittance was actually determined in 1993 according to the base amount of some local revenue items in 1993 and the re-sharing of some revenue items between the central government and local governments according to the TSS reform. This can be explained in terms of a formula: The base of the tax rebate in 1993 = The net amount of local remittance to the central government in province i in 1993 = total local revenue remitted to central government in province i – total central revenue assigned to province i. The total local revenue remitted to the central government included mainly the base amounts of the following items among others: 1. VAT and excise tax from the industry sector; 2. VAT from wholesale and retailing; 3. Unitary industry and business tax (the former was abandoned after the TSS reform; 4. Enterprise income tax from non-banking financial institutions; 5. Remittance from increase in income of cigarette enterprises. Total central government revenue assigned to local governments included the following 7 main items: 1. 70 percent of the product, VAT and excise tax from centrally owned enterprises in the oil, chemical, electric and several sectors before TSS; 2. Central government shared product tax from certain cigarette enterprises, and the product tax from cigarette industry after price increase; 3. Central government shared special revenue from cigarette and liquor industry, 4. Central government shared 50 percent of urban land use tax; 5. Central government shared 30 percent of land occupation tax; 6. Central government shared compensation revenue for state owned land use; 7. Central government owned resource tax. For years after 1993, the tax rebates are determined according to the following formula: Tax rebate for province i in the year t = Tax rebate i, t-1× {1+ 0.3× }, where, R = [VAT (0.75) in 2002 +consumption tax in 2002] / [VAT (0.75) in 2001 +consumption tax in 2001] -1. What is interesting about the dynamic adjustment based on the so-called “1:0.3” coefficient method is that while it left the base stock amounts unchanged, it allowed for the centralization of future increases in VAT and excise tax revenues. This, of course, supported the central governments objective of increasing its share in total revenues. The approach in theory guarantees the annual increase in the central government share in VAT and excise tax revenues, and with time, it will eventually get close to 100% of the increment in excise taxes and close to 75% of the increment The System of Equalization Transfers in China 21 in VAT revenues according to central government. This process is shown for 1994-1997 in Table 6: The Distribution of Incremental Revenue from VAT (75 percent) and Excise Tax Between Local and Central in 1994-97 (RMB billion and percent) Year Total Increment Central amount Local amount Central share Local share 1994 36.3 26.5 9.8 73.0 27.0 1995 26.1 19.4 6.7 74.2 25.8 1996 35.0 26.9 8.1 76.7 23.3 1997 29.9 23.6 6.3 79.0 21.0 2.3 Equalization grants (The general purpose grants) The equalization grant system from in China involves a number of different grants, including: (a) the pre-tax sharing system grants; (b) transitory period grants; (c) equalization grants for minority regions; (d) grants for increasing wages of civil servants and (e) a number of other grants that also have an equalization objective. Although it may seems surprising that grants for increasing wages of civil servants are grouped with equalization grants, the reason is that they are allocated according to equalization criteria .The relative importance of each of these grants is shown in Table 7 using budget data for 2000, Let us look at each one of these in more detail. Table 7: Equalization Grants from the Central to Sub national Governments in 2000 (RMB billion) Pre tax Transito Equalizatio Grants for Other grants Total sharing ry n grants for increasing system period minority wages of civil grants grants regions servants 11.4 8.5 2.5 51.4 21.4 95.2 2.3.1 (a) Pre tax sharing system grants. Pre tax sharing system grants are actually the contracted fixed grants under the “contracted fiscal system” before the TSS reform in 1994. Recall that under the “contracted fiscal system”, whether a province had to transfer resources to the central government or got grants from the central government depended on the gap between its revenues and expenditures measured in the base year. And the gap was approximately the amount of transfers to or from the central government. Once the 22 International Studies Program Working Paper Series amount of the transfer was decided, it usually did not change over a period of several years unless there was an overall adjustment of intergovernmental fiscal relations. Despite the methodological problems surrounding the “Contracted Fiscal System” it is quite likely that the pre - TSS grants were the most (and continue to be) the most equalizing grants in the history of intergovernmental fiscal relations in China, the methodology followed in its determination8 took into account expenditure needs and low fiscal capacity even the incentives were wrong. At the present time there are only about 16 provinces receiving these grants. Most of these provinces are in the Central and west regions. These grants can represent a significant source of revenues for these provinces. The transfers received from the pre-TSS grants by these provinces in 2001 are presented in Table 8. Table 8: Pre-TSS Grants in 2000(RMB billion) Inner Mongolia 18.42 Jilin 1.07 Fujian 5.42 Jiangxi 0.45 Shandong 1.59 Guangxi 6.08 Hainan 1.71 Sichuan 3.43 Guizhou 7.42 Yunnan 6.73 Tibet 27.87 Shannxi 1.2 Gansu 1.26 Qingha 6.56 Ningxia 5.33 Xinjiang 19.37 Total 114.0 8 Even though these grants were based on the gap between actual expenditures and actual revenues (as opposed to expenditure needs and fiscal capacity) we believe that at the time these grants were derived actual (revenue and expenditure) measures were quite closed to capacity measures. Of course the caveat with using actual measures is that the gap may not reflect actual needs adjusted for capacity. The System of Equalization Transfers in China 23 2.3.2 Transitory period grants. The so-called transitory period grants were established in 1995 as a complementary matching measure to the 1994 TSS reform. Two main reasons explain why these grants were denominated “transitory period grants”, First, since these grants were the first formula-based grants in the history of intergovernmental fiscal relations in China, Ministry of Finance officials responsible for their design, who did not have much experience at the time, and felt a little tentative about how long these grant would last. Officials at the Ministry of Finance also worried about whether the results from the allocation of these grants to different provinces would be reasonable. The second reason was that under the budget constraint at that time, it was impossible for the central government to arrange adequate funds for the pool of equalization transfers, while at the same time local governments were expecting much more from the central government. In this situation, if the grants had been directly called equalization grants, quite possibly the central government would have seen political opposition from local governments. By calling the grants “transitory period grants”, it became easier for local officials to accept them from a political and also technical viewpoint. (i) The determination of the total pool of funds for the transitory period grants. The total amount of the grants is determined by the central government, on an ad hoc basis considering budgetary demands every year. As we can see from Table 9, in the first year, the pool of funds for transitory period grants was just RMB yuan 2.1 billion in 1995 and in 2001 overall funding had reached RMB yuan 13.8 billion. Although these levels of funding are clearly far too small for adequately addressing China’s current horizontal fiscal disparities, the transitory period grants continue to play an important role in structuring China’s transfer system. The question is whether these grants will be the vehicle for the implementation of a more ambitious equalization program by China’s government. In the following paragraphs we analyze current structure of two transitory period grants. Table 9: Funding levels for transitory period grants over the past 7 years (RMB billion) Year 1995 1996 1997 1998 1999 2000 2001 Amount 2.1 3.5 5.0 6.0 7.5 8.5 13.8 24 International Studies Program Working Paper Series (ii) The classification of transitory period grants The overall grant actually consists of two parts: the general grants, and the special grants. Thus the formula for the allocation of transitory period grants to a province is written as Ti=Gi+Si Where Ti is the total transitory period grant to province i; Gi is the general grants of region and Si is the special grants of province i (iii) The allocation of general grants (Gi) The amount of the general grant is determined by 3 factors: standard revenue, standard expenditure and the grant coefficient (βi). Regions whose standard revenue is greater than or equal to their standard expenditure do not qualify to get a grant. The formula used in the computation of the general part of the grant is as follows: Gi = (SEi - SRi)×βi Where Gi is the general grant for province i; SEi is standard expenditure of province i; SRi is standard revenue of region i; and βi is the coefficient of grants of region i. (iv) The determination of standard revenue (SRi) Standard revenue is comprised of standard local own revenues, the tax rebate and general purpose grants net of remittances from local governments to the central government. General purpose grants include pre tax sharing system grants, grants for increasing wage standards of civil servants and welfare levels of social benefits recipients (or simply called grants for wages), the equalization grants for minority regions and other general purpose grants. Summarizing, the formula for the computation of the standard revenue is as follows: Standard revenue = standard local shared and own taxes + tax rebate + general purpose grants - remittances to the central government In the above formula, the amounts of tax rebate, general purpose grants and remittances to the central government are real amounts decided upon by the central government. Standard local shared and own taxes is the measure of fiscal capacity of the province and it can be written as: The System of Equalization Transfers in China 25 Standard local shared and own taxes = Standard VAT (25%) + Standard business tax + Standard urban maintenance and construction tax+ Standard resource tax + Standard tax on urban land use + Standard agriculture tax + Standard vehicle and vessel utilization tax + Standard slaughter tax+ Personal income tax + Stand enterprise tax+ Standard property tax + Standard stamp tax+ Other items of revenue. As for other items of revenue, that are not standardized, they are mostly cut off from the standard local shared and own revenue base because they are not comparable across local governments. However, there are some remittances that are directly taken as part of standard local own revenues. It must be noted that the standard revenue measures entering the formula are not actual revenues but they are in fact estimated. The more detailed calculation of standard taxes is continued in Appendix 1. (v) The determination of standard expenditures (SEi). The determination of standard expenditure, the measure of expenditure needs includes several steps: First total local expenditures are classified into 12 items. Of these 12 items, 7 sector expenditure items are calculated using the same methodology based on personal and office expenditures. These 7 items are: government administration; agriculture, forestry and irrigation; culture and sports; education; health care; and “other”. The other 5 items include pension, “urban maintenance and construction, social security and welfare, grain risk fund, and supporting agriculture, each requires a different method to calculate the standard expenditure. The methodologies followed for arriving to these expenditures are summarized in Appendix 2. (vi) Determination of the Coefficient (βi) The coefficient is given by the formula: Basic coefficient = total grants /(standard total expenditure × standard total revenue). In turn, the basic coefficient is adjusted for each region by the following formula: The adjustment ratio of region i = standard revenue of region i / standard expenditure of region i; In general the higher the adjustment ratio, the lower the coefficient. 26 International Studies Program Working Paper Series (vii) The determination of special grants Special grants include three parts. The first is grants favoring minority regions. The second is grants for border counties. Third one is grants for counties where used to be revolutionary and civil war areas. These grants are not independent earmarked grants, instead, they are actually special factors used for modifying transitory period grants. ---- Grants favoring minority regions Minority regions in China (including 8 regions and 8 prefectures in non-minority regions) typically have weaker fiscal resources, which is reflected into lower levels of per capita revenues. These regions, in addition, usually have to pay higher costs for the provision of public goods and services since most of these regions are located in remote areas with unfavorable natural conditions. Using the uniform formula of the transitory transfer system (recall there are general equalization grants), it is difficult to target special assistance to minority regions. This is the basic reason for the expenditure of special grants for minority regions. What this means is that the minority regions, apart from the regular grants determined by uniformed formula discussed above, can also receive special grants. The special grants are determined by the gap between the standard fiscal expenditure and standard revenue and a coefficient. The formula for minority grants is as follows: Grants for minority regions in region i = (standard expenditure of region i - standard revenue of region i ) × coefficient of grants for favoring minority regions in region i The coefficient in the formula is based on the basic coefficient for all minority regions and it is calculated as the ratio between which the total of grants for favoring grants minority regions in the overall pool of transitory period grants and the gap between aggregate standard expenditure and revenue in minority regions. The specific coefficient for minority region i is calculated as the coefficient for all minority regions adjusted by the: Difficulty degree coefficient of region i = standard revenue in region i / standard expenditure in region i × 100%. ----Special grants for border areas Special grants for border areas is targeted to the continental border areas. If a border area is located in provinces with standard expenditure greater than standard revenue, their grants can be calculated by the gap of standard expenditure and revenue of border areas and the related grants coefficient. If the border area is located The System of Equalization Transfers in China 27 in provinces with standard expenditure less then standard revenue, the amount of grants they receive is half the amount of the border areas above for which standard expenditure greater than standard revenue. 2.3.3 Equalization grants for minority regions In order to offer further support to the development of minority regions, the central government established an equalization grant system specifically geared to the minority regions in 2000. As already pointed out above, minority regions consist of 8 provinces and 8 minority prefectures located in non-minority provinces in 2000. (i) The determination of the pool of funds The total amount of the grants is composed of two parts, the first part was RMB 1 billion which is allocated by the central budget in 2000, the beginning year of for these grants. From 2001 onward, the first part is calculated by applying the following formula: TMt = Mt-1(1+g), where TMt is the first part of the total grants in year t, g is the growth rate of the central shared VAT in year t. The second part is 80% of the total amount of the yearly increase in collection for the central government share of the VAT collected in minority regions. (ii) The allocation of the grants The equalization grant for each minority region also includes two parts, one is a yearly incremental return of central VAT, the other is computed by formula. Thus the formula for the allocation of the grants is as follows: EGMi = VRi + FGi, where EGMi is total equalization grants for minority region i; VRi is the direct yearly incremental return of central VAT, and FGi is the part calculated by factors. In addition, VRi = YVi ×50%, where YVi is equal to 80% of the yearly increase of central shared VAT from minority region i. And FGi = (SEi - SRi) ×βi , where SEi is the standard expenditure of region i; SRi the standard revenue of region i, 28 International Studies Program Working Paper Series and βi is the coefficient of grants of region i. Note that the equalization grants for minority regions is the first equalization grants in China which has a fixed or explicitly stated growth mechanism. In 2000, the central government allocated RMB Yuan 2.5 billion, in 2001, the figure rose to 3.3 billion. 2.3.4 Grants for increasing standard of wages of civil servants These grants were designed to support provinces of the center and west regions in the implementation of policies for increasing standard of wages of civil servants mandated by the central government. The main justification was that without fiscal assistance from the central government, many provinces in the central and west regions of China would have faced serious fiscal difficulties complying with the wage mandate. The central government decided to increase wages starting July 1, 1999 by an amount of RMB 120 per capita each month. During 2001, there were four wage increases: On January 1, 2001 at a rate of RMB 100 per capita per month, and on October 1, at a rate of 80 per capita per month. The third increase was to establish a bonus system for civil servants from 2001 onward, which was equivalent to an increase of about one month of wages for each civil servant. The fourth wage increase was to establish a subsidy system for remote areas. More than 700 counties were qualified to “benefit from” this policy. Complementing these four policy decisions, the central government established four types of grants. However, in addition, the central government also established a system of grants specifically designed to help provinces facing fiscal difficulties to pay wages of teachers in elementary and middle school in rural areas. The general formula of the four grants for increasing standard of wages of civil servants is given by: WGi = WDi ∗bi, Where WGi is the wage grants for region i, WDi is the increased expenditure resulting from the mandated wage policies and bi is the coefficient of grants for region i. The variable WDi is calculated as the number of civil servants in each region and the standard of wages as increased by the central government. The coefficient bi is as usual the ratio of basic expenditure (including wages and office expenses) to disposable revenue. Of course, different regions have different coefficients. The coefficient varies from as low as 20% to as high as 80%. Apart from the regular coefficient, minority regions usually can get another 5 percentage points added to their coefficient. Note that these grants for increasing standard of wages of civil servants have a special well-define purpose and as such can be, classified as conditional or specific The System of Equalization Transfers in China 29 grants. However, because these grants are allocated on the basis of equalization principle criteria they are grouped here with other equalization grants. 2.3.5 Sub-provincial equalization grants. Apart from earmarked grants, provincial government also implements several forms of grants destined to sub-provincial units. These include • Pre tax sharing system grants • Transitory period grants • Grants for wage increase • Other grants. Most of these grants at the provincial level follow the structure of central government grants. The important exception is for the case of transitory period grants. It is important to describe this type of grant further because as is the case at the central level, the transitory period grants are also the most promising type of equalization grants. Transitory period grants to sub-provincial units show several important features. First, the total amount of the grants has reached a respectable level. The Ministry of Finance estimated that the amount of grants within the provinces in 1999 was 6.6 billion RMB Yuan, of which provincial governments provided RMB Yuan 6.5 billion, and prefecture governments provided RMB Yuan 0.1 billion. Second, both provincial and prefecture governments in most regions transfer directly to county governments, although as we just mentioned the amounts from prefectures are much smaller. The direct transfers from the provincial governments to the counties gets around the sharp hierarchical system that works in most other fiscal issues in China. Third, the short-run target of transitory period grants in most areas is to ensure normal operation of governments and payment of civil servants on time. However, it is common for these grants to try to establish incentive mechanisms, encouraging recipients of grants to mobilize revenue and control expenditures. 30 International Studies Program Working Paper Series Table 10: Sources of funds for equalization transfers (Transitory Period Transfers) received by sub-provincial governments. (1994-1997) (in 10,000 yuan) From From From Total From Provincial Prefecture County Central Level Level Level Level National 2744001 1055641 1345317 321132 21911 Total Beijing Tianjin 85538 33033 38498 14010 Hebei 41495 29099 2459 9937 Shanxi Inner 224935 111582 84000 28468 885 Mongolia 47000 47000 Liaoning 57521 14301 43220 Jilin 45889 30853 15036 Heilongjing Shanghai 124100 124100 Jiangsu Zhejiang 70156 35050 35106 Anhui 35261 35261 Fujian 59361 41933 9000 8428 Jiangxi 31000 31000 Shangdong 118882 70110 45122 3650 Henna 83903 57494 19119 7290 Hubei 96357 45057 36000 15300 Hunan 193696 193696 Guangdong 118190 54470 52023 5472 6225 Guangxi 6263 6263 Hainan 30416 8666 16655 5095 Chongqing 134457 75216 47082 10524 1635 Sichuan 137434 105399 13582 18453 Guizhou 425213 5165 227630 179252 13166 Yuannan 38537 27659 10878 Tibet 91181 46181 45000 Shaanxi 90769 40993 34523 15253 Gansu 165538 58944 106594 Qinghai 78272 45539 32733 Ningxia 112637 112637 Xinjiang Source: “Tranistory period transfer in China”, BudgetDepartment, MOF, Sep.1999, China Finance and Economy publishing House The System of Equalization Transfers in China 31 Transitory period grants at the sub-provincial level have helped to address intra-provincial regional fiscal disparities and allow. Nonetheless, this system is not well funded and it does not appear to be able to root out the severe fiscal problems that many sub-provincial government unites still face. From all appearances there are still significant imbalances in sub-provincial distribution of fiscal resources. Table 10 presents data on the cumulative amount of equalization transfers (Transitory Period Transfers) received by Sub-provincial governments in the period 1994-1997. 3. Main issues with the current system of equalization transfers Eliminating the current system of equalization grants in China is not an easy task. We thought that a prudent way to proceed was to establish some standards and then see how well China complies with them. 3.1 Same Principles for the Design of Equalization Grants One way to evaluate the performance of a system of equalization grants is to analyze how well the system fits recognized universal principles that should be followed in the design of equalization transfers. There have been many studies on what those principles should be. For example, Martinez-Vazquez and Boex (2002) argue that all equalization transfer system should obey the following principles: • • • • • • • Equalization transfer systems should provide adequate resources to local governments so as to balance national priorities and local autonomy. The system should support an equitable allocation of resources by providing more resources to local governments with lower tax capacities and greater expenditure needs. Equalization transfers should be predictable in a dynamic sense and should be stable over a period of years to promote revenue predictability and overall budget stability. Equalization grants should be simple and transparent. In order to achieve that, it is important to limit their objectives exclusively to the purpose of equalization of fiscal capacity and /or fiscal need. The distribution of the grants should be understandable to all stakeholders, and should avoid political manipulation or negotiation in any of its aspects. The system should avoid negative incentives to revenue mobilization by sub national governments, and should avoid encouraging inefficient expenditure choice. The transfers of funds should be unconditional lump-sum grants for general-purpose financing of sub national governments. During the introduction of the new transfer mechanism or changes in the transfer system sudden large changes in funding for local governments should be avoided. 32 International Studies Program Working Paper Series 3.2 How well does China’s system fit the ideal design principles? Clearly there is no equalization system in the world that would fit all the desirable principles perfectly. But there are systems that get closer and other systems that are far away from the principles. China’s current transfer system is somewhere in between. It scores well in some areas and not so well in others. First, all equalization transfers, including transitory period grants, grants for wage increase and equalization grants for minority regions, take into account lower tax capacity and greater expenditure needs of local governments. In general, only regions with weak fiscal capacity and high expenditure needs qualify to receive equalization grants, and the coefficients in the three different grant schemes are determined by the degree of fiscal difficulty, with regions in greater difficulty getting relatively higher grants. Second, most of the equalization grants are calculated according to formulae based on objective factors and invoking the principle of fairness. Third, most of the equalization transfers are transparent, so that local governments can calculate the amount of grants allocated to them. Fourth, negative incentive for revenue mobilization and inefficient expenditure choices by sub-national governments are largely avoided in the transitory period grants, which are among the largest equalization grants, because the formula for the transitory period grants uses standard revenue measures of tax capacity and standard expenditure measures of fiscal needs. Fifth, most of the equalization grants, including transitory period grants and the grants for minority regions, are unconditional or lump-sum grants to be used for general purpose financing of sub-national governments. Thus, local governments can decide on the use of these funds according to their expenditure priorities independently. However, there are some other areas where current Chinese practice with transfers is inconsistent with desirable design principles. The determination of the total funding for some equalization grants is uncertain. In particular, there is no explicit rule for determining the overall funding for transitory period grants, a critical type of equalization grant. The level of funding is decided by the annual budget based on ad hoc principles. Second, the equalization grant system has had relatively low level of funding and therefore has had a limited impact on fiscal disparities across sub-national governments. The tax rebate of RMB 2,206 billion in 2000 was two and a half times larger than all equalization grants combined that year (Table 11). Where the government has had some discretion so far it has shown a preference toward The System of Equalization Transfers in China 33 earmarked grants over equalization grants. In 2000 the funding for earmarked grants was 80 percent larger than total funding for equalization grants (Table 11). Equalization grants have represented approximately between 4.5 and 6.5 percent of local government revenues from 1994 to 1999. But they jumped to 10.12 percent in 2000 (Table 11). In part due to the inadequacy of the funds dedicated to equalization purposes, sub-national fiscal disparities have continued to expand. Although transfers from the central government after 1994 have significantly improved the fiscal capability of the mid-west region, disposable revenue in the central and western regions has increased much more slowly. The share of these two regions in national income has decreased in recent years. As a result, the gap between disposable revenue in the central and western region and the east coastal region, in per capita terms, has expanded too. For example, income per capita of civil servants of local governments had increased by 104% between 1994 and 2000, but whole for the, the east coastal regions it increased by 128% in the central and west regions just increased by 96% and 72%, respectively. In 2000, income per capita in central region was 43% of that in the east region and the equivalent figure for the western region was 48%. Third, there has been a tendency toward the proliferation of earmarked grants to the detriment of unconditional equalization grants. In recent years, whenever the central government has introduced an important policy toward sub-national governments, it has also invariably established an earmarked grant to match that policy. Although the introduction of conditional grants maybe attractive and evermore effective in achieving policy goals, under the general budget constraint for the central government, the more earmarked and conditional grants, the less funds available for the equalization grants. Furthermore, earmarked grants from the central government usually have required local governments to match a portion of the funds. As a result, the pressure on local budgets has increased, leading to a de facto less discretion over budgets at the local level, but also to some “creative” public accounting. According to an unfinished statistical report of the Ministry of Finance for 1999, the matching requirement for local governments from central government earmarked grants was up to RMB yuan 35.7 billion, while local governments in practice only matched RMB yuan 24.7 billion. In order to get earmarked grants from the central government, some local governments cheated the central governments by “re-matching” (using the same fund to match different earmarked grants). This was particularly common in some regions facing fiscal difficulties. Also in some counties in Hunan province earmarked grants such as that for agriculture are usually used for paying wages of civil servants. These are examples also dramatized well the point that earmarked grants are difficult to monitor and that by particular objectives of the central government behind the earmarked program are generally not so easy to achieve. Table 11 Transfers from the Central Government to Provincial Governments (RMB yuan billion and percent) Item 1994 1995 1996 National Total 2389.02 100.00 2532 .89 100.00 2672.35 100.00 Tax Reabate 1798.9 75.30 1867.26 72.92 1948.64 72.92 Equalization Grants 228.66 9.57 290.90 11.49 234.91 8.79 Pre Tax Sharing 114.06 4.77 115.21 4.55 110.92 4.15 System Grants Transitory Period 20.71 0.82 34.65 1.30 Grants Equalization Grants for Minority Regions Grants for Increasing Wages of Civil Servants Other Equalization 114.60 4.80 154.99 6.55 89.33 3.34 Grants Regular Earmarked 361.37 15.13 343.73 14.79 488.80 18.29 Grants Earmarked Grants From Bonds Total Local Disposable 3680.72 4445.32 5302.99 Revenue Ratio of Equalization Grants to Total Local Disposable Revenue 6.21 6.54 4.43 Source: Ministry of Finance 34 1997 2800.90 100.00 2011.63 71.82 273.37 9.76 111.92 4.00 1998 3285.33 100.00 2082.76 63.40 324.92 9.89 113.28 3.45 1999 3992.28 100.00 2120.56 53.12 511.39 12.81 113.91 2.85 2000 4747.65 100.00 2206.54 46.48 893.37 18.82 119.78 2.52 50.21 60.54 75.29 85.45 1.80 25.53 0.54 1.79 1.84 1.89 108.37 2.71 216.98 4.57 111.23 3.97 151.10 4.60 213.83 5.36 445.63 9.39 515.65 18.42 590.65 17.98 1139.08 28.53 1198.86 25.25 286.99 8.74 221.25 5.54 448.87 9.45 6058.268 6702.39 7643.9715 8819.04 4.51 4.85 6.69 10.12 The System of Equalization Transfers in China 35 Fourth, China’s equalization grants system is overly complex. If we regard each equalization grants for increasing wages of civil servants as a type of equalization grants, up till now, the central government has established 9 equalization grants. By comparison most countries with decentralized systems of public finance have only one equalization grant. In addition, the complexity of the system in China is not only about the number of grants but often for how those grants are structured. There is ample room for simplification and rationalization of all equalization grants, especially because they have similar repeated features. It should be noted also, that some of the equalization grants are conditional. For example, the grants for wage increases, although they have strong equalization features, their main objective is to promote the implementation of wage polices. The conditionality for the use of the funds is obviously inconsistent with unconditionality of equalization grants. Last, the effectiveness of China’s equalization system is likely compromised because of the strong federalist features in the current vertical structure of government. Provinces can largely decide how to arrange their fiscal affairs with cities and prefectures, and these latter with counties, and so on. Following the TSS reform of 1994, sub-national governments restructured their intergovernmental fiscal relations, but most local governments did not relinquish their vested interests, neither did they change their sub-provincial intergovernmental fiscal relations with the changed economic conditions. Often, it would appear that provincial and prefecture level governments do not allocate enough funds within their budgets for equalization purposes. In fact, some of these governments appear to have retained funds from central grants that were destined to counties and townships. In this sense sub-national equalization grants did not play the necessary supporting role in reducing fiscal disparities. Discretion among sub-national governments for how they want to organize their intergovernmental fiscal relations reduces the ability of the central government to implement its own policies and achieve its own objectives, in this case reduce fiscal disparities. Meanwhile fiscal disparities at the sub-provincial level are as serious or more than at the central-provincial level9. 9 Many Local Governments continue to lack enough funding to provide basic services. For example, in Yunnan province in 2000 revenues per capita term at the provincial and prefecture levels were 11 and 7 times respectively higher than that at the county level. Fiscal difficulties have led to lower level government to the accumulation of budgetary arrears. It is estimated that increased arrears of wages for civil servants in 2001 was up to 6.5 billion yuan, among which, Inner Mongolia, Jilin, Liaoning, Shandong, Yunnan and Hubei were the provinces with the most serious accumulation of arrears. See For Example World Bank 2002. 36 International Studies Program Working Paper Series 3.3 Recent reform initiatives Most of the problems with the current grant system listed above have been brewing for quite some time and generally there has been awareness among government authorities of their importance. In fact, the government recently has taken a series of policy initiatives in this respect. The first policy thrust has been to control the total funding going to earmarked grants so as to increase the pool of funds available for equalization grants. More concretely in recent years, the central authorities have examined all the earmarked grants and some of them have been converted into equalization grants. For example, about RMB 4 billion of subsidies for food in urban areas were transferred into the transitory equalization grants in 2001. The same was done with earmarked grants for the development of border regions. The second policy thrust has been to design specific regulations to enhance the effectiveness of central government policies at the sub-national level. It has been quite difficult for the central government to achieve its equalization objectives without the full cooperation and compliance by provincial governments. Thus, one of the priorities facing the central government is to design an enforceable institutional framework which defines the responsibilities of each level of local governments in the area of equalization. There is a draft document for such an institutional framework, “Direction on some questions of improving sub-provincial fiscal relations” which has been approved by the state council. This institutional framework is organized around several main ideas. First, there is realization that without an effective sub-provincial grants system, fiscal disparities at the prefecture and will not be relieved, even when the central government expands the size of the equalization grants. The goal is to have all sub-national governments in their respective environments, adjust their sub-provincial intergovernmental fiscal relations system to, prevent fiscal disparities from expanding, and ensuring the normal operation of county and lower level governments. But, how are these goals to be achieved? The position taken in the document is that given China’s large site and diversified regions, it would be probably impossible to find a unique format capable of accommodating all the different situations at the different levels of government. The mandate from the central government would be that appropriate adjustments in expenditure and revenue structures are carried out at the provincial and prefecture government, so that they have enough fiscal sources to provide the necessary equalization grants to county level governments. It is also being considered that in less-developed areas, county government could collect all revenues and finances all expenditures of townships, that is, the county governments would take over all the fiscal responsibilities of township governments. The third policy thrust is to get additional and also more regular fiscal resources for equalization purposes at the central level through the reform of the sharing of The System of Equalization Transfers in China 37 income tax. As discussed in the first half of this paper, on January 1, 2002, China introduced a new arrangement for income tax-sharing which increases the share of central government in this tax and also earmarks the additional revenues to funding the expansion of the equalization grants. In 2000, the total amount collected in income taxes was RMB yuan 243.6 billion. Of this, the central government received RMB yuan 92 billion, or 38% of the total. The remaining income tax revenues were kept by the sub-national governments. It must be noticed that these local income taxes mainly stayed within the nine provinces in the coastal areas better off. These nine provinces collect about 70% of the total income taxes10 After the reform, the central government has increased its share in income tax revenues from 38% to 50% in 2002,and it will be up to 60% in 2003. It is estimated that during 2002 another 10 billion RMB will be added to the equalization grants pool. Over the longer term, the increase in funding of equalization grants is expected to be more pronounced as the tax base of the income grows and expands and a noticeable difference. 4. The way forward The steps China’s government has taken recently to improve the system of equalization grants are quite positive and should start producing results in a short period of time. However, the growing dimension of fiscal disparities is likely to require wider action from the government. The reform of the system of equalization grants needs to be synchronized with the reform of the other elements of the intergovernmental fiscal relations system. Reforming the system of equalization grants in isolation, of the related problems in other areas of fiscal relations, will reduce the effectiveness of the recent reforms. What follow are some recommendations for a wider reform effort. First, there is a need to have a clearer assignment of expenditure responsibilities among all different government levels. The lack of clear expenditure assignments still leaves the responsibilities of each level of government quite undefined and therefore it is much harder to determine what ought to be an adequate level of funding and the needs for equalization. There is still ambiguity in the assignment of responsibilities in areas such as infrastructure, agriculture, education, culture and healthcare. Practically all levels of governments share these expenditures, but there are no clear guidelines for how to divide these responsibilities among different levels. Furthermore, in practice, some responsibilities obviously belong to the local governments and ought to be funded by the local revenues, but they are often funded by special funds from the central government. A typical case of this situation is the increase of wages for civil servants, 10 They are Beijing, Tianjin, Liaoning, Shangdong, Jiangsu, Zhejing, Fujian, Shanghai, and Canton. 38 International Studies Program Working Paper Series in recent years, where the central government has shared most of the expenditure for local wage increase of civil servants.11 Second, there is a need to review the current assignment of revenues and find safe ways to increase tax autonomy and revenue mobilization at the sub-national level. The problem of vertical imbalances can be adequately addresses through a higher level of revenue autonomy. The increased level of revenue mobilization by local governments can free resources at the central level to take a more aggressive stand toward funding equalization. Third, there is a need to identify a clear comprehensive objective for the equalization system. The objective pursued by equalization grants in China is not always clear; some grants even show the features of earmarked grants, usually based on temporary considerations. To establish a sound equalization system will require a clear comprehensive objective, at least guaranteeing that every local government is able to deliver a minimum level of public goods and services. Last, there is a need to make the current system of equalization grants simpler and more transparent. To this end, it would be desirable, among other things, to consolidate the current various equalization grants into one single grant. In addition, it would be desirable to establish a mechanism for explicitly identifying the total pool of equalization funds. Some options from international practice include setting aside for equalization a portion of a major tax, several taxes, or some percentage of total central government revenues. How high the pool of equalization funds should be or indeed how much the system should equalize will require systematic economic analysis but also clear political choices. The central government needs to decide on the basis of the potential tradeoffs between equalization on one side and economic growth and overall incentives for revenue mobilization on the other. 11 On the other hand, the local governments also share some responsibilities, which ought to be funded by the central government and thus adding to burden of local governments. For example, expenditure for armed army for public security is the responsibility of the central government according to document of the central government, but in practice local governments have to share some of the expenditure. The System of Equalization Transfers in China 39 Appendix 1: Estimation of Standard Revenues for the Transitory Period Grants ---- Standard VAT (25%) The standard VAT (25%) is determined by two components, VAT collected from industry and that from wholesale and retailing industry. Each of them is based on the simple formula: standard tax base × standard tax rate. The values added in industry and the value added in wholesale and retailing are taken as the standard tax bases in each respective formula. The local average effective VAT rate is used as the standard rate for the VAT calculations, but the effective VAT rate for industry is adjusted by a coefficient of enterprise size and structure. These two factors influence collection efficiency greatly. The formula of standard VAT is as follows: Standard VAT(25%) in region i= (value added of industry in region I x local average effective tax rate of industry x coefficient of scale and restructure of industrial enterprises in region i + value added of wholesale and retailing industry in region I x local average effective tax rate of wholesale and retailing) x 0.25. ---- Standard business tax Under current tax regulations, there are 9 main activities that fall under the business tax: transportation, construction, banking and insurance, postal services and communication, culture and sports, recreation, other services, exchange of intangible assets, and selling of real estate. Generally speaking, the standard business tax needs to be calculated item by item according to the list of activities above. However, since the business tax collected from banking and insurance is “regulated”, the figure reported in tax revenues for this item is reasonable, so in this case it is unnecessary to calculate the standard tax. For some other activities subject to the business tax it is not possible to get suitable data for the computation of the standard tax base. In all these cases data from actual revenues are taken as the standard revenue. Eventually, only business taxes collected from construction, food services, real estate trade, and road transportation are standardized, based on the formula: standard business tax base × standard business tax rate. Of course, the standard business tax base differs from industry to industry. For construction the base is income from the construction industry, for food services is the total sale amount, for housing also sale amount, and for road transportation the turnover amount. The tax rate applied is the local average business tax rate for each industry. ---- Standard urban maintenance and construction tax The urban maintenance and construction tax is a surcharge on the actual amounts due on VAT, exercise tax, and business tax. The tax base for the urban maintenance and 40 International Studies Program Working Paper Series construction is based on the real excise tax revenues, standard revenue for VAT and business tax. Tax rates are based on the effective tax rates in each locality, which in turn are determined by the ratio of real revenue from urban maintenance and construction tax to real revenue from VAT, excise taxes and business tax for each locality. Thus the formula for the standard urban maintenance and construction tax is as the following: Standard urban maintenance and construction tax in region i =(standard VAT in region i + standard business tax in region i +real exercise tax revenue in region i)×effective rate of urban maintenance and construction tax in region i. where, Effective rate of urban maintenance and construction tax in region i= real urban maintenance and construction tax in region i /(real VAT in region i + real exercise tax in region i + real business tax in region i)×100% ---- Standard urban and township land use tax The standard urban and township land use tax is determined by the formula of “standard tax base ×standard tax per unit”. The area of occupation of commercial housing is used as the standard tax base. The standard tax per unit is the real local average tax per unit adjusted by GRP per capita in each region. The formula of standard tax per unit is as the follows: Standard tax per unit in region i = local average effective tax per unit×(GRP per capita in region i/ GDP per capita of the state). ---- Standard agriculture tax The standard agriculture tax is determined by the formula “tax base” x (1+ price adjustment coefficient)”. The tax base means agriculture tax revenue in the base year, which is calculated by the following formula: Agriculture tax base in region i = (Agriculture tax revenue in region i in 1976 + Agriculture tax revenue in region i in 1977 + Agriculture tax revenue in region I in 1978) /3. The price adjustment coefficient is the ratio of the price index for grain purchase prices of the current year to those in the base year(1978). The grain include rice, wheat, corn, and soybean. The reason a different approach is used in the calculation of the standard agriculture tax is that agriculture tax policy differs across regions and as a result the rate of the The System of Equalization Transfers in China 41 agriculture tax also varies from region to region. This makes it difficult to determine standard tax revenue by a standard approach. In addition, the data for agriculture tax revenue in 1976-1978 were reliable because under the socialism planning economic system there was a strong reporting system. Note that the agriculture tax chiefly relates to grain purchase prices since grain production is relatively stable. ----Standard vehicle and vessel utilization tax The standard vehicle and vessel utilization tax is also determined on a “standard tax base × standard tax rate” basis. The standard tax base is calculated separately according to the quantities and capacities of passenger vehicles, general trucks, two-wheel motorcycles, three-wheel motorcycles, motor vessels and non-motor vessels in each locality. Tax rates are determined by the average of the upper tax rate and lower tax rate stipulated in the "Clause for Vehicle and Vessel Utilization Regulation". ---- Standard personal income tax Currently, personal income tax is chiefly collected from salaries and income of private industrial and commercial enterprises. So the calculation of standard personal income tax is based on these two taxable items using the formula “standard tax base ×standard tax rate”. In the case of income tax revenues from other sources, the actual tax revenue is regarded as the standard revenue. The calculation of standard personal income tax collected from salaries is based on the following formula: Standard personal income tax from salaries in region i = income tax base of salaries in region i ×income tax rate of salaries in region i The income tax base of salaries is based on per capita taxable salaries net of exemptions and number of employees. While the tax rate is based on local average effective tax rate, adjusted with a regional coefficient. The formula is as follows: Income tax rate of salaries in region i = (real local income tax collected from salaries/ local income tax base of salaries) × coefficient of region i The coefficient of region i is in turn based on the following formula: Coefficient of region i = per capita disposable income for urban residents in region i /local per capita disposable income for urban residents in the entire country. 42 International Studies Program Working Paper Series The calculation of standard personal income tax collected from income of private industrial and commercial units is based on the following formula: Standard personal income tax from income of private industrial and commercial units in region i = standard income tax base of private industrial and commercial unit in region i ×standard income tax rate of private industrial and commercial unit in region i In turn, the standard income tax base is approximated by the turnover amount of private industrial and commercial unit. The standard income tax rate is based on the following formula: Standard income tax rate =real income tax collected from local private industrial and commercial unit / real local turnover amount of private industrial and commercial unit. Apart from the above taxes, the enterprise income tax, the tax on special produces resource tax, slaughter tax, property tax are all also calculated as standard tax revenue based on the formula of *standard base ×standard rate”. The remaining standard revenue items such as stamp tax, VAT on land are based on real collection figures. Appendix 2: Estimation of Standard Expenditures for the Transitory Period Grants ----The determination of standard expenditure for the 7 sectors Government administration, agriculture, forestry and irrigation, culture and sports, education, health and “other”. The standard expenditure in this case includes both personal and office expenditures. Personal expenditure means salaries and bonus. Which is determined by the following formula: Standard personal expenditure of sector j in region i = the standard number of civil servants of sector j in region i × standard per capita salaries of sector j in region i (j=1.2…7 and i =1.2.3…31). ----The determination of standard number of civil servants. The standard number of civil servants plays an important role in calculating standard expenditure. In order to make this number more reasonable, the Ministry of Finance has tried different approaches. Most recently standard numbers of civil The System of Equalization Transfers in China 43 servants for 2001 was identified for different sectors and levels of governments using several methodologies. The standard number of civil servants in: • • • • • • The government administration is determined by agriculture population, no-agriculture population, and the number of government units The education sector is determined by the number of students, area of residential spots, which means area for housing and other facilities of life in different villages and districts, and the number of counties and districts The health care sector is determined by the number of population and the number of counties and districts The culture and sports sector is determined by the numbers of population, counties and districts and the total area of residential spots The agriculture, forestry and irrigation sector is determined by the number of people employed in agriculture, the number of counties, and the area of arable land The other sector by population and the number of counties and districts. Note that there are several basic steps in the calculation of the standard number of civil servants in any of these sectors. Taking the case of government administration, for instance, the first step is to find out the most significant identifying determinants factors in this sector using, for example regression analysis, Agriculture population, Non-agriculture population, and the number of government units are identified as these determinants. Each determinant is given a weight as follows: W no-agriculture population = 0.1940 W agriculture population = 0.3627 W number of government units = 0.4433 The second step is to calculate the unit quota for each factor based on the above weighs and the total actual number of civil servants in government administration. Assuming that each factor’s effect is independent, the result is as the following: I. Unit quota of non-agriculture population in different regions with different population size N few than 300000 = 0.0078 per civil servant / per person of population N 300000 - 600000 = 0.005 per civil servant / per person of population N100000 * 1 million = 0.0046 per civil servant / per person of population N more than 1 million = 0.0032 per civil servant / per person of population 44 International Studies Program Working Paper Series II. Unit quota of agriculture population in different regions with different population size N few than 900000 = 0.0094 per civil servant / per person of population N 900000 * 2.5 million = 0.0028 per civil servant / per person of population N2.5 million * 4 million = 0.0027 per civil servant / per person of population N more than 4 million = 0.0016 per civil servant / per person of population III. Unit quota for government units. Regressive analysis shows the result: N unit quota for government unit= 958 civil servant /per county or district The third step is to sum up the different numbers of civil servants are as calculated by the different factors under the assumption that their influences are independent. For instance, given Baoliang city in Hebei province has 1,560,000 non-agriculture population, 8,940,000 agriculture population, and 25 counties in 2000, then the estimated standard number of civil servants in the government administration in Baoliang should be: 1560000 × 0.0032 + 8940000 × 0.0016 + 25 × 958 = 43246 (persons) ----The determination of standard per capita salaries. Standard per capita salaries are based on the average local per capita salaries at the different government levels province, prefecture and county, respectively. Office expenditure refers to those for fuel and maintenance of vehicles, heating, and all other office expenses. Again different approaches are used to calculate each of those expenditures. The formula for total standard office expenditures is as follows: Standard office expenditure = standard expenditure of fuel and maintenance of vehicle + standard expenditure of heating + standard expenditure of other office expenses ----The determination of the standard expenditure of fuel and maintenance of vehicles. The calculation of these expenditures can be expressed as: Standard fuel and vehicle maintenance expenditure = Standard expenditure for fuel + Standard expenditure for vehicle maintenance The System of Equalization Transfers in China 45 Standard expenditure for fuel is determined on the basis of the standard number of vehicles, the quota of fuel consumption per vehicle per year, unit fuel price and a plateau area adjustment coefficient. This formula is as follows: Standard expenditure for fuel = Standard number of vehicle × fuel consumption per vehicle per year × unit fuel price × plateau area adjustment coefficient, Where, the Standard number of vehicles is determined on the basis of the standard number of civil servants and people-vehicle ratio in various sectors covered by the budget. The index of fuel consumption per vehicle per year is provided by the Ministry of Transportation. The unit fuel price is taken as the price(s) determined by the central government. The plateau area adjustment coefficient is determined on the basis of fuel consumption data furnished by the National Bureau of Standards and average elevation data furnished by the National Bureau of Mapping. Standard expenditure for vehicle maintenance is determined on the basis of standard number of vehicles, single vehicle annual actual maintenance expenditure and the road condition adjustment coefficient. The formula for determining standard expenditure for vehicle maintenance is: Standard expenditure for vehicle maintenance = Standard number of vehicles × single vehicle annual actual maintenance expenditure × Road condition adjustment coefficient, Where, single vehicle annual actual maintenance expenditure is determined in light of the national average expenditure for vehicle maintenance and the road condition adjustment coefficient is determined on the basis of the rate of sub-quality roads and a coefficient of damage done by sub-quality roads to vehicles. The formula for road condition adjustment coefficient is: Road condition adjustment coefficient = Rate of sub-quality roads ×Vehicle damage coefficient, Where the rate of sub-quality roads is the ratio of the length of grade 4 roads and under -grade 4 roads to the total length of roads. The determination of the rate of sub-quality roads in a certain locality can be done through the following formula: Rate of sub-quality roads in a certain locality = (Length of grade 4 roads in the locality + length under grade 4 roads in the locality)/Total length of roads in the locality. 46 International Studies Program Working Paper Series The vehicle damage coefficient is taken to be equal to 0.2, and its value is provided by an automobile research institute of China. ----Determination of standard expenditure for heating Standard expenditure for heating is determined by the number of standard civil servants and annual per capita standard heating expenditure. In order to calculate the latter factor, the entire country is divided into five groups of regions on the basis of the average daily temperature in all localities (based on capital cities in each region) between October and April. Each group of regions shares the same annual per capita standard heating expenditure, which is determined by the real yearly heating expenditure of administrative organs and the real number of civil servants within the group of regions. The formula for the calculation of standard heating expenditure is as follows: Standard expenditure for heating in a certain locality = Standard number of civil servants × standard per capita heating expenditure of the region where the locality is located. ----Standard expenditure for other office expenses Standard expenditure for other office expenses of a particular sector is determined by the standard number of civil servants and actual local per capita office expense in terms of real civil servants. The formula is given by: ----Standard expenditure for urban maintenance and construction Expenditures for urban maintenance and construction mainly cover infrastructure such as roads, urban utilities and residential housing construction. The historical profile of the expenditure for urban maintenance and construction shows approximately 50 percent of the expenditure for urban roads, 25 percent for urban population and 25 percent for urban area factors. Therefore, the so standard expenditure for these purposes is determined on the basis of urban road, population and urban area factors. This can be expressed as the following formula: Standard expenditure for urban maintenance and construction of region i = real local expenditure for urban maintenance and construction (0.5 × urban road distribution coefficient of region i + 0.25 × urban area distribution coefficient of region i + 0.25× population distribution coefficient of region i). The urban road distribution coefficient is derived on the basis of three elements, namely urban population, area of completed urban zones and number of urban vehicles. The weights for those elements are obtained through regression analysis. The following formula is used to derive the urban road distribution coefficient: The System of Equalization Transfers in China 47 Urban road distribution coefficient of region i = weight of urban population × urban population of region i /national urban population +weight of urban vehicles × number of urban vehicles in region i /national number of urban vehicles + weight of area of completed urban zones× area of completed urban zones in region i /national area of completed urban zones. Furthermore, we have that: Urban area distribution coefficient of region i = area of completed urban zones in region i /national area of completed urban zones, and Urban population distribution coefficient of region i = urban population of region i / national urban population. The weights for the different factors entering the urban road distribution coefficient are determined also by regression analysis with the following: 0.1194 for urban population, 0.5317 for vehicles and 0.3489 for completed urban zone. ----Standard expenditure for the grain risk fund Standard expenditure for the grain risk fund is determined by the amount of earmarked grants for the grain risk fund and the required matching ratio by the central government (what is for example a common value for the matching ratio) It can be expressed as the following formula: Standard expenditure for grain risk fund in region i = the amount of earmarked grants for grain risk fund from the central government × required matching ratio by the central government in region i 48 International Studies Program Working Paper Series References Martinez-Vazquez, Jorge and Jameson Boex. 2001. The Design of Equalization Grants: Theory and Applications. Andrew Young School of Policy Studies, Georgia State University. World Bank. 2002. China: National Development and Subnational Finance: A Review of Provincial Expenditures. Washington D.C.

RELATED PAPERS

RELATED TOPICS

Log In

The System of Equalization Transfers in China

The System of Equalization Transfers in China

Related Papers

RELATED PAPERS

RELATED TOPICS