A Study of Productive Expenditure Bias in County-level Finance in China
2012; Taylor & Francis; Volume: 33; Issue: 1 Linguagem: Inglês
10.1080/02529203.2012.650414
ISSN1940-5952
Autores Tópico(s)Taxation and Compliance Studies
ResumoAbstract 中国县级决策者主要对上级负责 , 追求尽可能高的经济增长率 , 而非居民福利最 大化 , 导致其财政决策偏向生产性支出。基于2067个县(市)2001—2005年财政经济 数据的实证检验,为县级财政生产性支出偏向的存在提供了有力的证据。生产性支出 偏向严重地妨碍公共财政体制的建设和向公共服务型政府的转型。解决这一问题的根 本出路在于逐步增强县级政府满足本地居民需求的激励。 关键词: 财政支出结构 财政支出偏向 县级政府 Being responsible mainly to their superiors, decision-makers at the county level in China pursue the maximum growth rate rather than their residents’ welfare. This leads to a preference for productive expenditure in their fiscal decisions. An empirical test based on 2,067 counties (cities) provides robust evidence for the existence of productive expenditure bias at the county level. Such bias causes a serious impediment to the construction of a system of public finance and the transformation to public service-oriented government in China. An appropriate solution to this problem is to gradually enhance incentives for county- level governments to satisfy the demands of local residents. Keywords: fiscal expenditure structurefiscal expenditure biascounty-level government Notes The authors appreciate the support of the National Natural Science Foundation of China (Grant No.: 70773010) and the Key Project of the Chinese Ministry of Education (Grant No.: 309007), as well as the constructive advice of peer-reviewers. Of course the authors bear full responsibility for any errors in this paper. 1 We can often hear about complaints about “mouth-feeding budgets” from county-level or evenhigher level officials, who say that their financial income falls far short of demand and has to be mostly spent on payrolls. The implication is that if allowed, they will do their best to promote “construction,” a syndrome similar to the “investment thirst” of the planned economy. 2 Qiao Baoyun, Fan Jianyong and Feng Xingyuan, “Fiscal Decentralization and Compulsory Primary Education in China.” 3 Ping Xiaoqiao and Bai Jie, “Fiscal Decentralization and Provision of Local Public Goods in China.” 4 Fu Yong and Zhang Yan, “Chinese-style Decentralization and Fiscal Expenditure Composition: The Cost of Rivalry for Growth.” 5 The latest empirical research employing county-level financial data includes the following main articles: H. Yin, “Fiscal Disparities and the Equalization Effects of Fiscal Transfers at the County Level in China,” pp. 115-149; Yuan Fei et al., “Fiscal Transfers and Expansion of Fiscal Dependents in China’s Fiscal Recentralization”; Yin Heng and Zhu Hong, “The County-level Fiscal Gap and Equalization of Fiscal Transfers.” 6 Provincial fiscal expenditure responsibilities cover administrative expenditure at the provincial level; expenditure of provincial-level representative offices; expenditure of provincial public security organs, procuratorates and people’s courts; expenditure on provincial or cross-prefectural water conservancy projects and transportation infrastructure (including local railways, provincial highways and expressways); expenditure on provincial science and technology R&D (including applied research); expenditure on provincial agricultural production and technology popularization; expenditure on provincial cultural events and provincial-level historical and cultural heritage protection; subsidies for provincial universities, vocational schools and high schools; spending on compulsory education (the provincial finance share); expenditure on second-tier health care (provincial prevention and cure of endemic diseases and subsidies for provincial public hospitals); provincially-supported social security funds and outlays for provincial-level affairs commissioned by the central government. The expenditure responsibilities of county-level governments include county (city) level administrative expenditure; expenditure of county (city) level public security organs, procuratorates and people’s courts; spending on water conservancy projects and transportation infrastructure (including highways and expressways); public services such as transportation, gas and tap water; expenditure on urban infrastructure construction and maintenance; spending on applied research and other science and technology R&D; outlays on cultural activities and historical and cultural heritage protection; outlays for county-level vocational schools and high schools; expenditure on compulsory education (the county-level share of finance); primary-level health care (prevention and cure of endemic diseases and subsidies for public and township hospitals); county-supported social security funds; and expenditure on county-level matters commissioned by the central government. 7 For instance, education expenditure averaged 25.26% of total county-level expenditure in 2003. The figure for the provincial level (and below) was 15.7% (see China Statistical Yearbook). 8 The Chinese-style decentralization characterized by “political centralization” and “economic decentralization” has attracted extensive scholarly attention. A large stock of literature explains China’s economic growth from this perspective. See Yingyi Qian and Gérard Roland, “Federalism and the Soft Budget Constraint,” pp. 1143-1162 9 To simplify our analysis, we do not consider for now fiscal transfers from higher-level governments. It is true that as the policy of “the city repays the countryside” is further implemented, the proportion of fiscal transfers in county-level fiscal revenues is increasing. However, the introduction of financial support from higher-level governments into this basic framework will not change the result. 10 The Budget Law of China stipulates that local governments should not have a budgetary deficit. 11 See P.M. Romer, “Increasing Returns and Long-run Growth,” pp. 1002-1037. In our model, this condition comes from introduction of the Production Function (1) with productive public expenditure included and proportional tax rate, so that output is linear to capital and the marginal productivity of capital does not change. 12 Economic performance is the crucial factor in the performance evaluation of lower-level government officials conducted by their superiors. Higher-level governments do relate political promotion to local economic performance, with those who have a relatively outstanding economic performance getting promoted. See Hongbin Li and Li-an Zhou, “Political Turnover and Economic Performance: The Incentive Role of Personnel Control in China,” pp. 1743-1762. Competition for promotion among local officials also greatly stimulates the pursuit of economic growth. See Zhou Li-an, “Governing China’s Local Officials: An Analysis of the Promotion Tournament Model.” 13 The details of the proof are omitted here. 14 Ping Xinqiao and Bai Jie employ a similar approach. They argue that local budgetary expenditure is subject to more constraints and local government officials have more control over off-budget expenditure. They compare the different impacts of on- and off-budget income on the expenditure structure at the provincial level. See Ping Xinqiao and Bai Jie, “Fiscal Decentralization and the Provision of Local Public Goods in China.” However, absent are county-level data on off-budget income and expenditure. 15 Being freely disposable and the above-mentioned expenditure constraints are different concepts. The former is used as opposed to institutionally regulated and specific use of earmarked funds. For example, county-level fiscal revenues are in the charge of and therefore can be freely disposed of by county-level governments. The latter concept refers to governments’ having to prioritize the fiscal funds at their disposal; for instance, their fiscal revenues are first of all subject to the expenditure constraint of “ensuring the normal operation of government.” 16 For example, fiscal transfer for the reform of rural taxes and fees is such a policy: with a view to ensuring the smooth progress of the reform, higher-level financial authorities consider as a whole factors such as increased revenues thanks to raising the agricultural tax rate, reduced income because of abolishing township-level fee collection and the butchery tax, lowering the rate of tax on special agricultural products and adjustment of methods of handling village public accumulation funds. The appropriate subsidies are provided through fiscal transfers to areas where a net income decrease has emerged. Transfer payments are mainly to be used for primary and secondary education costs, village administration, the support of households enjoying the five guarantees, and may not be used for other purposes by county-level governments. 17 The more detailed procedure of “two ups and two downs” is adopted in some places. The budgetary unit reports its suggested budgetary plan for the next year to the supervising budget agency (the first “up”), which, after an overall consideration of its financial resources, informs the budgetary unit of the specific budget cap (the first “down”). The budgetary unit then makes a new plan according to the budget cap and reports it to the supervising agency (the second “up”), which revises and collects budgetary unit plans and reports them to the People’s Congress for approval in January, then returns them to the budgetary units (the second “down”). 18 Incomplete statistics indicate that there are ten-plus regulations at the provincial and central government levels stipulating that a single expenditure item must occupy a certain proportion of fiscal expenditure or must increase by a larger margin than gross expenditure (or income). For instance, the Agriculture Law that came into force on July 2, 1993, stipulates that agricultural expenditure must increase by a larger margin than recurrent income; the Education Law that took effect in January 1, 1994, prescribes that education spending shall reach 4% of China’s GNP; and the State Council’s Decision on Environmental Protection, which entered into force on August 3, 1996, rules that related input shall be 1.5% of GNP. 19 See P. Mauro, “Corruption and the Composition of Government Expenditure,” pp. 263-279; C.A. Shelton, “The Size and Composition of Government Expenditure,” pp. 2230–2260. In accordance with this article and available data, the control variables are defined here as including GDP per capita, proportion of primary industry, proportion of secondary industry, population density, proportion of the agricultural population, proportion of fiscal dependents, proportion of primary school students, proportion of high school students and number of hospital beds per 10,000 population. 20 J.M. Henderson, “Local Government Expenditures: A Social Welfare Analysis,” pp. 156-163. This effect remains a heated topic in the sphere of public finance. See R.P. Inman, “The Flypaper Effect.” 21 Theoretical circles tend to regard this as an irrational phenomenon, because according to the standard theory of the utility maximization of the median voter or representative resident, different income sources should not have different impacts on expenditures. See J. Hines and R. Thaler, “Anomalies: the Flypaper Effect,” pp. 217-226. Some scholars also attempt to understand this phenomenon within the framework of the rational man assumption; for example, J. Roemer and J. Silvestre, “The ‘Flypaper Effect’ Is Not an Anomaly,” pp. 1-17. 22 “Fungibility” is another important topic frequently discussed in the literature on local public finance and international aid; for instance, H. Pack and J.P. Park, “Foreign Aid and the Question of Fungibility,” pp. 258-265. 23 In fact, some scholars have noted that fungibility is lacking in the income sources of local governments, which is another cause of the “flypaper effect.” See J. Hines and R. Thaler, “Anomalies: the Flypaper Effect,” pp. 217-226. In this sense, the model in this paper provides a theoretical explanation of the “flypaper effect” from a new perspective. 24 J. Hamilton, “The Flypaper Effect and the Deadweight Loss from Taxation,” pp. 148-155. 25 B. Knight, “Endogenous Federal Grants and Crowd-out of State Government Spending: Theory and Evidence from the Federal Highway Aid Program,” pp. 71-92. 26 Matz Dahlberg et al., “Using a Discontinuous Grant Rule to Identify the Effect of Grants on Local Taxes and Spending,” pp. 2320-2335. 27 See J.M. Wooldridge, Introductory Econometrics: A Modern Approach. 28 Treasury Department and Budget Department of the Ministry of Finance, Fiscal Statistics of China’s Prefectures and Counties. 29 In the first case, with change of name or administrative subordination but no essential change in jurisdiction, we regard the county-level units as the same as before. In the second case, with no change of name but an essential change in jurisdiction, we treat them as different places, cease using the previous code and give them new codes. In the third case, where the name changes and there is an essential change in jurisdiction, the previous code is eliminated and new codes are given. 30 Some studies do not define in advance the nature of public expenditure; instead, they treat it as one of the research objectives. This expenditure is regarded as productive expenditure if its outcomes are favorable to production and growth. See Shantayanan Devarajan, Danyang Xie and Heng-fu Zou, “Should Public Capital Be Subsidized or Provided?,” pp. 319-331. 31 The proportion of “other expenditures” averaged 34.4% of the total county-level fiscal spending during 2001-2005. The proportion of administrative spending was 21.4% on average. 32 Agricultural productive spending for 2001 and 2002 refers to aid to agricultural production and operating expenses on agriculture, forestry, weather and water conservancy; for the period of 2003-2005 it includes expenditures on agriculture, forestry, weather and water conservancy. 33 Health care expenditure is essentially welfare spending. It is analyzed separately as only data from 2003 to 2005 are available. 34 We also discussed the case without the rigid constraints above, and analyzed the 50%, 40% and 30% samples of the regions with the highest county-level revenues. We found that remains significant in every case, and the coefficient of productive expenditure is a relatively high positive figure while that of welfare expenditure is negative. This marked difference in expenditure structure further indicates the existence of productive expenditure preference. We do not report these findings here due to space constraints. Furthermore, areas are sequenced in the order of their average revenues for the period 2001-2005. 35 See A. Shah, “A Practitioner’s Guide to Intergovernmental Fiscal Transfer”; R. Boadway and A. Shah, Intergovernmental Fiscal Transfer: Principles and Practice. 36 Bardhan and Mookherje believe that the efficiency of fiscal decentralization lies in the local supervision that confronts local governments. see P. Bardhan and D. Mookherjee, “Capture and Governance at Local and National Levels,” pp. 135-139. 37 Xiaobo Zhang et al., “Local Governance and Public Goods Provision in Rural China,” pp. 2857-2871.
Referência(s)