Broadband infrastructure and local government fiscal risk: evidence from China

ABSTRACT

Network infrastructure is a cornerstone of Internet applications and digital economy. In this paper, we primarily evaluate the effect of broadband infrastructure on local government fiscal risk by using the enforcement of the ‘Broadband China’ Strategy (BCS) as an exogenous shock. After a simple theoretical analysis by establishing a three-department endogenous growth model, we conduct a staggered Difference-in-Differences (DID) estimation and use a panel datum of China’s 285 prefecture-level cities from 2003 to 2019, finding that the BCS policy apparently reduces fiscal risks of the pilot cities compared to the untreated ones. This impact is achieved through two channels: the enlargement of data creation and the enhancement of data use efficiency. Moreover, we discover that this policy contributes more to cities with mature IT industry, deepened information consumption, strong innovation capabilities and flourishing cross-border e-commerce. In addition, it is demonstrated that cities with severe financial burden always benefit more from the BCS policy; meanwhile, the policy effects will spill over to surrounding districts. Finally, we find that the construction of broadband infrastructure greatly improves residents’ satisfaction with local government. Overall, this paper provides constructive insights into broadband infrastructure for diminishing local government fiscal risks in developing countries.

KEYWORDS:

• Broadband infrastructure
• the “Broadband China” Strategy policy
• local government fiscal risk
• data creation
• data use efficiency

JEL CLASSIFICATION:

• H54
• H62

Acknowledgments

All the authors contributed equally. Qianbin Feng thanks the National Natural Science Foundation of China (#101302-N11502) and Zhejiang Provincial Soft Science Research Key Programme (2022C25035) for financial support. Lingxi Jiang thanks the Zhejiang Provincial Soft Science Research Key Programme (2022C25035) for financial support. We thank the advice provided by Prof. Ying Ge (Zhejiang University), Lexin Zhao (Zhejiang University), the editor and two anonymous referees.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability statement

The data that support the findings of this study are available from CNRDS and CSMAR database. Restrictions apply to the availability of these data, which were used under license for this study. Data are available at https://www.cnrds.com/Home/Login, https://cn.gtadata.com/ with the permission of CNRDS and CSMAR database and https://authorservices.taylorandfrancis.com/data-sharing/share-your-data/data-availability-statements/.

Notes

¹ Data sourced from ‘China Broadband Development White Paper (2022)’ and ‘China Digital Economy Development Report (2022)’, both proposed by China Academy of Information and Communications Technology (CAICT)...

² The core ideology of placebo test is to conduct estimation by fabricating treatment groups or treatment time points. Our two tactics are in correspondence with these two methods respectively...

³ In order to improve the identification power of this test, we repeat the above random process 1,000 times..

⁴ Same as the falsification method above, in all these five columns, we replace BCS with the false BCS dummies, and re-estimate Equation (16)(16) $Fisris{k}_{it}=\alpha +\beta BC{S}_{it}+\lambda X_{it}+Z_{jt}+{\nu }_i+{\mu }_t+{\epsilon }_{it}$(16) ...

⁵ We conduct this indicator-substitution robustness test for the reason that we have discussed it is the inefficient and unsustainable allocation of financial resources rather than fiscal deficit itself that would necessarily induce fiscal risk..

⁶ Compared to PSM, EB’s most prominent advantage is to enable researchers to simultaneously control for multidimensional balance of the treatment and control groups’ sample covariates, thereby a more exact matching of the two groups can be achieved.

⁷ We consider these four policies because they are large-scale, authoritative and representative, and their policy effects on Chinese economy have been corroborated by multiple literature (Zhang, Hong, and Zhu 2022; Gao and Yuan 2022)...

⁸ As a supplement, we have also discussed the endogenous problem, with methods and results shown in Appendix Table A4.

⁹ Take channel 1 as an example, BCS_Internet_1 actually represents $BC{S}_{i,t}\times \mathrm{\Delta }Interne{t}_{i,t+1}>0$, while BCS_Internet_2 denotes $BC{S}_{i,t}\times \mathrm{\Delta }Interne{t}_{i,t+1}<0$. Specifically, $\mathrm{\Delta }Interne{t}_{i,t+1}>0(<0)$ is a dummy variable equal to one if Internet users increase (decrease) in period t + 1, else zero...

¹⁰ Transaction cost in this paper is calculated by the selling expenses of listed companies as a percentage of operating revenue, referring to Liu, Li, and Xue (2009).

¹¹ We apply the Peking University Digital Financial Inclusion Index of China (PKU_DFIIC) to measure the digitization level of China economy in terms of digital finance applications (Guo et al. 2020). The website of PKU_DFIIC database is https://idf.pku.edu.cn/yjcg/zsbg/index.htm...

¹² The website of database is http://www.isss.pku.edu.cn/cfps/. The CFPS, which was launched in 2010 by the Institute of Social Science Surveys (ISSS) with Peking University, employs a multistage probability sampling procedure. It is nationally representative, covering 162 counties or districts (counties hereafter) across 25 mainland provinces that represent 95% of the total population in China.

¹³ For CFPS only provide the accurate province code and the city code is fictitious rather than standard, we cannot merge it to our database. Therefore, we only evaluate the effect of broadband infrastructure on the residents’ evaluation of local government..

Additional information

Funding

The work was supported by the National Natural Science Foundation of China [#101302-N11502]; Zhejiang Provincial Soft Science Research Key Programme [2022C25035]