Examining the relationship between municipal consolidation and cost reduction: an instrumental variable approach

ABSTRACT

This study explores the cost-reduction effects of local government consolidation, using data from Japanese municipalities. We use municipal panel data from Japan for the years 2000, 2005 and 2010, a data set including numerous consolidation cases. Reverse causality between consolidation decisions and size of municipal expenditure is a concern in the voluntary consolidation of local governments. An instrumental variable (IV) methodology is employed to deal with possible endogeneity associated with consolidation decisions. The main finding is that municipal current expenditure per capita increases immediately after consolidation, but then gradually declines. Because consolidation seems to increase per capita expenditure in the absence of the effect of this long-term declining trend, incorporating this trend into estimation is crucial in cost-efficiency analysis of consolidation. It is also shown that conventional panel data regressions yield biased consolidation effects, suggesting the superiority of the IV approach.

KEYWORDS:

• Expenditure
• instrumental variable approach
• local government
• municipal consolidation

JEL CLASSIFICATION:

• H11
• H72
• H77

Acknowledgements

I would like to thank Takero Doi, Junya Hamaaki, Chung Mo Koo, Toru Takemoto, Hiroki Tanaka, Satoshi Watanabe, Izumi Yokoyama and the participants at the 69th meeting of the Japanese Institute of Public Finance, the 2013 Japanese Economic Association Spring Meeting and the Hitotsubashi University and Kyushu University seminars for their helpful comments. This work was supported by the Japan Society for the Promotion of Science under JPSP Grant-in-Aid for Young Scientists (B) 22730256 and by Kyushu University under a special research fund.

Disclosure statement

No potential conflict of interest was reported by the author.

Supplemental data

Supplemental data for this article can be accessed here.

Notes

¹ Municipalities comprise cities, towns and villages, and form the basic units of local government, while prefectures are broader regional governments, each encompassing multiple municipalities. Regarding Japanese municipal consolidation, recent studies attempt to explore the existence of common pool problem in terms of local bond issues (Hirota and Yunoue 2017; Nakazawa 2016).

² ‘Inhabitant tax’ is levied on both individuals and corporations.

³ See, for example, CLAIR (2009), MIC 2012, 41) and Yokomichi (2007).

⁴ This was conditional on the submission of a consolidation application by 31 March 2005.

⁵ However, municipal consolidations are not compulsory under Japanese law and, in fact, are more voluntary now than previously (Yokomichi 2007).

⁶ Instead of overall expenditure, current expenditure – defined as total expenditure minus debt payments and investment – is employed as a dependent variable because the debt expenses and investment are affected by past municipal decisions.

⁷ Mochida (2006) introduces the Japanese unconditional grants in detail. Reschovsky (2007, 414–416) briefly explains the mechanism of the correction (modification) coefficients.

⁸ Outliers, which outweigh two standard errors of LEXP, are excluded from the regressions. Also, municipalities that consolidated more than twice in the period are excluded.

⁹ For a robustness check, the sample including all the years 2000–2010 is also used.

¹⁰ In this respect, it is well known that in Japan more fiscally healthy municipalities were reluctant to consolidate even during a surge in municipal consolidations.

¹¹ The marked increase in 2009 probably reflected the expansion of subsidies from the national government aimed to stimulate the economy following the 2008 worldwide recession.

¹² The estimates of FOREIGNER probably reflect the fact that the larger the proportion of foreigners, the higher per capita spending on public welfare and schools for foreign students.

¹³ Consolidated municipalities would raise their public expenses to cover costs of consolidation-related projects such as integration of computer and compensation systems and reallocation of public facilities (MIC 2006d).

¹⁴ In fact, many merged municipalities were required to reduce the number of seats contested in the four yearly municipal elections to cut costs (Yokomichi 2007).

¹⁵ See, for example, RIIRE (Research Institute of Industry and Regional Economy) (2000). RIIRE (Sangyo Chiiki Kenkyu-jo in Japanese) was founded and is supported by Nikkei, a major newspaper company in Japan. RIIRE is mainly engaged in research on the Japanese economy, business and regional affairs, including local governments and local economy.

¹⁶ The index presented by RIIRE is a weighted sum of these specific public output indices. The current study does not calculate such a combination because the weights used in the calculation are set based on discretion and may not adequately represent true public service levels.

¹⁷ The estimates of the public services are presented in the Electric Supplementary Material (ESM). Consistent with the inference that larger public output generally increases costs, the coefficients of ageing society and education are positive and statistically significant.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science under JPSP Grant-in-Aid for Young Scientists (B) 22730256 and by Kyushu University under a special research fund.