Municipal merger and debt issuance in South African municipalities

ABSTRACT

Existing research reveals opportunistic free-riding behaviours by municipalities facing mergers; these behaviours include over-issuance of debt before mergers because their burden will be shared by newly constructed municipalities. Nevertheless, the literature presents few solutions to this problem. We therefore investigate whether South African municipalities that faced mergers engaged in free-riding behaviours before the mergers. In South Africa, the upper government of municipalities implemented the monitoring of borrowings before mergers. Using a difference-in-differences analysis and quarterly data of South African municipalities, we find that South African municipalities decreased the amount of borrowings before their mergers. This result contradicts previous research, indicating that proper policy for municipal mergers may prevent the fiscal common pool problem caused by free-riding behaviours. To the best of our knowledge, our study is the first to demonstrate a reduction in borrowings before municipal mergers.

KEYWORDS:

• Municipal merger
• common pool problem
• free ride
• DID
• fiscal monitoring

JEL CLASSIFICATION:

• H83
• H72
• H77

Acknowledgments

This research is based on the project by the Japan International Cooperation Agency. We are very thankful to Hirofumi Kawakita for the support. We also thank Clement Mulamba, Katsuyoshi Nakazawa, Nobuo Akai for their comments, the National Treasury of South Africa for provision of data and research environment, JSPS KAKENHI Grant Number JP18J12301, and the CBI Program of Osaka University for financial support. Appendix of this paper is available at https://bit.ly/SA_DID. The authors are responsible for all remaining errors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Blom-Hansen (2010) also studies this.

² Since district governments have few works, this study focuses on metropolitan and local areas.

³ The financial year of South African municipalities starts from July. We omit districts in the analysis.

⁴ The MDB itself can also submit the request.

⁵ 16 criteria listed in the Municipal Demarcation Act are in appendix. For the endogeneity, we investigate the propensity of merger using logit regression, although obvious propensity cannot be observed. See appendix.

⁶ Although electoral commission was supposed to be involved in the process of demarcation change, according to Municipal Demarcation Board (2016), the commission did not oppose to the decision.

⁷ See appendix for detail.

⁸ See appendix for explanation of placebo test.

⁹ Since the National Treasury considers the audit outcomes of ‘Unqualified – With findings,’ ‘Unqualified – No findings,’ and ‘Outstanding’ to be clean outcomes, Clean Audit dummy takes 1 if the outcome was one of these three and 0 otherwise.

¹⁰ The only yearly data available are population, GVA and Clean Audit. We use linear prediction to divide population and GVA data and substitute the current year’s value for Clean Audit data to construct quarterly data.

¹¹ Debt reduction is observed only in the last quarter of FY2015/16 in our result. This is because it took a time to stop contract to borrow money even though the monitoring began from January 2016.

¹² In the robustness checks in appendix, using the yearly and half-yearly data, the coefficients of the interaction terms are not significantly different from 0. This result is reasonable because debt reduction is observed only in the last quarter of FY2015/16. Moreover, this result is also different from the existing research since over-borrowing is not observed. We also conduct some more robust checks in appendix. See appendix.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science [JP18J12301].