https://orcid.org/0000-0002-1585-5023
![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
Interregional inequality of public services is accompanied by their combined supply by government and market. This study investigates whether the spatial allocation of regional investment influences inequality and explores the differentiated source of such influence, distinguishing between government investment and private investment in China. Results suggest that the interregional inequality of public services is spatialised by the distorted spatial allocation of investment, which manifests itself as capital market segmentation, and it is aggravated by government investment but relieved by private investment. Such effects derive from differences in local protectionism, interregional competition and capital flow between government investment and private investment.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the editor and anonymous reviewers for their insightful comments and constructive suggestions.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the author(s).
Correction Statement
This article has been corrected with minor changes. These changes do not impact the academic content of the article.
Notes
1. The report is called the Evaluation of Basic Public Service Capacity in Chinese Cities (2017), which takes the capital cities of provinces, municipalities and cities with independent planning in China as samples to evaluate the levels of basic public services (including nine dimensions: public transport, public safety, housing, education, social security and employment, public health, the urban environment, culture and sports, and public office service). This report points to a phenomenon of increasingly worsened disparities in intercity and interregional public service development in China. In particular, developed provinces tend to improve while underdeveloped provinces fall further behind.
2. The standard yardstick competition follows a classic principal-agent model in explaining the political and administrative dynamics under interjurisdictional competition (Besley & Case, Citation1995). Yardstick competition is categorised as either ‘from the bottom’ or ‘from the top’. Under the ‘from the bottom’ pattern, voters (principals) in democratic countries compare the performance of politicians (agents) in local governments to their counterparts in other jurisdictions through voting, and local governments compete with their peers by using peers’ information with regard to the tax rate and fiscal planning as yardsticks (Besley & Case, Citation1995; Caldeira, Citation2012). Under the ‘from the top’ pattern, governments at higher levels (principals) evaluate the performance of governments at lower levels (agents). Notably, China exemplifies the ‘from the top’ pattern paired with highly centralised vertical authority. Specifically, China is a unitary country with the central government setting uniform statutory tax rates across provinces. Thus, the yardstick competition is not from the bottom driven by voting populations; rather, it is from the top with the central government as principal. The next section will present an introduction to the institutional background of China.
3. Details are listed section A1 of Appendix A in the online supplemental data.
4. The indicator dimensions are similar to those in current studies (e.g., Li et al., Citation2017; Li et al., Citation2023). Additionally, they are consistent with the guidance of the 13th Five-Year Plan for the Equalisation of Basic Public Services from the China State Council. For the specific indicator selection strategies, taking the area of public culture as an example, in the thinking of indicator selection, the services provided by libraries and cultural facilities are the focus public culture resources. Therefore, library circulation and the population of television programs are the chosen indicators of the results in public culture (Li et al., Citation2023). Similarly, the ideas behind the indicator selection regarding other public service dimensions are consistent.
5. Details are listed in section A2 of Appendix A. The public service indicators and the weights are presented in section B1 of Appendix B.
6. Here, the definition and setting of government investment are unique and depend on the Chinese context, as supplemented in section A3 of Appendix A.
7. Details are shown in section A4 of Appendix A.
8. We cannot completely rule out the possibility that changes in the interregional inequality of public services are persistent over time; thus, we control for lagged values in the regressions. Furthermore, the core dependent variable CMS may be endogenous because of simultaneous errors, the omission of relevant explanatory variables, and measurement errors. The Urb variable is also endogenous, as the size of the provincial population simultaneously determines the urbanisation level and regional distribution of public services. The system GMM technique has the advantage of dealing with the endogeneity of all the explanatory variables by using their lagged values as instruments. In addition, there is no prior information on whether we should control for the lagged dependent variable in the regressions. Therefore, we also adopt the FE estimation method based on the static panel model, which can not only lay the foundation for subsequently introducing the SEM but also provide additional robustness evidence across estimation methods.
9. We do not add the control variables to the IPS model but add them to the CMS model for three reasons. First, we are primarily interested in how capital market segmentation arises and whether it can be explained by government investment and private investment. Second, the second equation is intended only to present the indicative link between capital market segmentation and the interregional inequality of public services. Third, this setting can reduce the number of endogenous variables of the two equations. Furthermore, Economic development is excluded from Equations (2) and (3) for the correlation between the investment and GDP indicators.
10. All the control variables along with the investment variables are used as instrument variables in the SEM under 3SLS estimations, to isolate the simultaneous effects between the endogenous CMS variable and the IPS variable.
11. The presence of serial correlation will render the lags of the endogenous variable invalid. Therefore, the Arellano and Bond (Citation1991) test for serial correlation in the first-difference equation residuals can validate the instruments in the dynamic GMM model. Specifically, autocorrelation of the first-order in the differenced errors (AR(1)) is acceptable, but there must not be any second-order serial autocorrelation in the errors (AR(2)).
12. Settings of the SAR model to examine the existence of spatial interaction in investment among provinces, and the specification of weight matrix in spatial econometrics are elaborated in section E1 of Appendix E. Moreover, given that the spatial lag variables are constructed based on economic similarity among provinces, WGov and WPri may be endogenous regressors. Following Halleck Vega and Elhorst (Citation2015), alternative spatial weight matrixes with more exogenous geographical attributes are incorporated into the SEM for robustness consideration. The analyses are detailed in section E2 of Appendix E.
13. The inflection point values are calculated by the first derivative CMS variable of Gov and Pri variable, respectively.
14. The inflection point value is compared with the mathematical characteristics of closeness centrality variables, and the quartile distribution of the four variables is shown in section B3 of Appendix B.