The Effects of Political and Fiscal Incentives on Local Government Behavior: An Analysis of Fiscal Slack in China

ABSTRACT

This study examines the determinants of fiscal slack from the perspective of Chinese local government officials. Given China's rapid economic growth over the past 30-plus years, Chinese local governments reportedly hold huge slack resources that attract public scrutiny. In an effort to improve their fiscal performance, some localities recently established budget stabilization funds, following a top-down initiative. However, it remains unclear to what extent fiscal slack has accumulated and which factors affect slack resource levels of Chinese local governments. By employing a panel dataset (249 prefecture-level cities, 1999–2009), this study finds that political and fiscal factors exert significant influence over local officials’ decisions about slack resource levels. The findings of the study bear implications for establishing fiscal rules and improving the performance of sub-national governments in China and other countries.

ACKNOWLEDGMENT

An earlier version of this article was presented at the Improving Chinese Public Service: Theories and Empirical Evidence seminar at the City University of Hong Kong in April 2013. The seminar was supported by grants from Chiang Ching-kuo Foundation (#CS002-P-12) and City University of Hong Kong (#1860403). I am grateful for the comments from Carolyn Heinrich, Richard Walker, Yijia Jing, and participants of the seminar; the editor of the journal and two blind reviewers; and Yilin Hou for his advice at the early stage of the research. Fangzhi Ye provided excellent research assistance.

Notes

Local governments in China refer to all sub-national levels of government in this study.

In recent years, the international market and academic circles have kept a keen watch on Chinese governments’ debts, particularly local government debts. The debt finance of Chinese local governments is completely devoid of fiscal transparency (Ma 2013). Local governments had not been formally granted the authority to borrow until the recent revision of the Budget Law in 2014. Thus, local government debts have accumulated mainly in the form of contingent liabilities, which are operated off budget (Ma 2013). By establishing local financing vehicles as off-budget entities, local governments have circumvented legal constraints and incurred considerable debts with explicit and implicit government guarantees and obligations for repayment if the capital projects were to fail. Due to the lack of reliable public data, it remains difficult to track local government indebtedness in China (He, Lin, and Tao 2014). Most debt revenues of Chinese local governments have been used for long-term infrastructure investment rather than for operating purposes (Ma 2013).

Since implementing its reform and open-door policy in 1978, China has maintained an average annual GDP growth rate of over 9% (Ding and Knight 2009; Prasad 2009). According to some sources, few countries have matched China's growth rate over the same period of time.

See http://www.china.org.cn/opinion/2012-07/05/content_25826944.htm.

The Chinese government system is comprised of five hierarchical levels: the center, provinces, prefectures, counties, and townships (Chen 2009; Wong 1997). There exists a strict vertical hierarchy among the five levels in fiscal (as well as other) affairs (Shah and Shen 2008). For example, prefectures fall directly under their respective provincial administration (Wong 1997).

The 1994 tax reform gave the central government 75% of the value-added tax and provinces and their localities the remaining 25%. Provinces and their localities were ordered to give 50% of personal and enterprise income tax to the central government in 2002 and 60% in 2003 (Kung, Xu, and Zhou 2013; Shih and Zhang 2007).

Similarly, some studies in the Western literature examine the link between the re-election probabilities of US gubernatorial officials and their choices of economic policy and the career concerns of firm managers (Besley and Case 1995; Murphy and Zimmerman 1993).

The study includes the two variables of a city party secretary rather than a mayor in the models due to some practical considerations. Under China's party-state political system, the concentration of power in the hands of city party secretaries, nicknamed first-in-command (yibashou in Chinese), grants them more control over socioeconomic policies in their jurisdictions than mayors.

For the evaluation of local leaders, work achievement accounts for 60 to 70% of performance, while other aspects, such as political integrity and competence, account for the remainder. Competition between local officials at the same level is an essential part of the personnel control system, as their performance ranking among peers is directly linked to their chances for promotion (Edin 2003).

The hukou, a household registration system in China, divides residents into two categories: urban and rural hukou holders. This system ties residents’ access to benefits and services to their residential status, which puts rural residents in a disadvantageous position. For several newly developed cities in the Pearl River delta, including Shenzhen, Zhuhai, Zhongshan, and Dongguan, all of their permanent residents constitute the urban hukou holders. Thus, their values for the urbanization level variable are 100%.

The absence of debt-level variable may generate an omitted variable bias problem. For this analysis, the inclusion of prefecture- and year-fixed effects in estimation at least partly removes the effect of omitted variable bias.

The four municipalities directly under the central government are excluded from this analysis. Prefectural cities in regions mainly for ethnic minority groups, including Xinjiang, Tibet, Qinhai, Ningxia, and Yunnan in western China, are also excluded because they operate quite differently from other cities, and there are missing data for some major variables.

Data on the two political variables are garnered from two websites: http://baike.baidu.com/ and http://www.hotelaah.com/liren/index.html. Cross-checking with other data sources was carried out to ensure accuracy and reliability.

The exchange rate between the US dollar and CNY has been approximately 6.20 in recent years.

For a more detailed discussion of the method, please see Heinrich (2002).

When population squared is included in the model, the result is not statistically significant and does not change the sign of the coefficient on population.

Additional information

Notes on contributors

Wen Wang

Wen Wang ([email protected]) is an Assistant Professor of the School of Public and Environmental Affairs at Indiana University–Purdue University Indianapolis. His research interests include capital budgeting and finance, education policy and finance, and intergovernmental fiscal relations.