Free riding, empire building, and cost management prior to and post municipal enterprise mergers in Japan

ABSTRACT

Opportunistic overspending tendencies induce inflexible cost adjustments and encourage bad cost stickiness. To examine whether this perspective applies to public sector organizations (PSOs) and mergers, we test the asymmetric cost behaviors, (1) free riding and (2) empire building, associated with managerial incentives in merged Japanese municipal enterprises. Consequently, we find evidence that prior to mergers, administrators’ opportunistic spending originating from free-riding behavior affects cost stickiness; conversely, an analysis of the post-merger sample shows that the empire-building incentive might partly mitigate sticky costs. Our results corroborate the contradictory role of managerial discretion in resource adjustment decisions between pre- and post-mergers.

KEYWORDS:

• Municipal enterprise
• merger
• asymmetric cost behavior
• free riding
• empire building

Acknowledgments

We thank Jong-Hag Choi, Jong-Seo Choi, Yu-Lin Chen, and the anonymous referees and workshop participants at the 2019 Korean Accounting Association Annual Conference and the 3^rd Asian Accounting Associations (AAAs) Conference in Gyeongju, Republic of Korea for their insightful comments and guidance.

Disclosure statement

No potential conflicts of interest are reported by the authors.

Notes

1. Mergers in PSOs are defined as consolidation processes in public organizations and are often synonymous with both amalgamation and annexation (e.g., Tavares 2018).

2. Mergers in PSOs have occurred in many diverse places, such as Asian countries, European countries, Oceanian countries, the US, and Canada (e.g., Holzer et al. 2009; Tavares 2018).

3. PSO mergers might be expected to produce larger territorial units, with potential impacts on economic efficiency, managerial effectiveness, and democratic outcomes (e.g., Holzer et al. 2009; Tavares 2018).

4. Regarding the PSO’s cost inefficiency in association with mergers, some researchers have indicated both political effects (e.g., Weingast, Shepsle, and Johnsen 1981) and fundamental bureaucratic influences (e.g., McGuire 1981); however, it remains unclear whether these influences significantly affect PSOs’ cost management in conjunction with mergers.

5. This theory has been illustrated in depth by many scholars (e.g., Tullock 1959) and discussed in relation to the merger of municipalities (e.g., Jordahl and Liang 2010; Blom-Hansen 2010; Hansen 2014; Saarimaa and Tukiainen 2015; Nakazawa 2016).

6. The ‘law of 1/n’ is explained as follows. As the number of groups increase, any one group absorbs a smaller share of a project’s costs since projects are paid for out of a ‘common pool.’ Each group then internalizes all of the marginal benefits from its project but incurs only 1/n-th of its marginal costs. There are free-riding incentives that increase n, and the same is true for the project size, total common pool spending, and inefficiency (e.g., Weingast, Shepsle, and Johnsen 1981).

7. Blom-Hansen (2010) expressed opportunistic spending before amalgamation as ‘last-minute spending’ before ‘closing time’ in his research.

8. The empire building motive has been illustrated in various contexts, such as private rent-seeking or achieving an opportunistic purpose, simultaneously putting an organization at higher risk and lowering the optimal return for shareholders, which is regarded as a fundamental agency problem (e.g., Williamson 1964).

9. Cohen, Karatzimas, and Naoum (2017) argued that understanding how costs ‘behave’ in PSOs could help to render PSOs more effective through the better allocation and management of costs.

10. The accounting principle of MEs is based on the accrual basis, not on the cash basis, adopted by many public organizations, and it uses almost the same accounting method as the method used by commercial enterprises.

11. Empirical studies of asymmetric cost behavior have excluded PSOs, such as utilities or other regulated industries, since they have argued that public services adopt a different accounting system (Shust and Weiss 2014) and that cost behavior analysis models only apply to competitive business fields because these markets reduce measurement errors due to a potential pricing effect (Weiss 2010). However, the same studies tested the cost behavior among local governments and revealed evidence of the asymmetric cost behavior among them that arises from the bureaucratic nature of the organization (e.g., Bradbury and Scott 2018; Cohen, Karatzimas, and Naoum 2017).

12. Rational resource management represents desirable managerial behavior, which contributes to creating value, whereas wasteful overspending due to administrators’ opportunism is interpreted as a sign of value-destroying overspending (e.g., Banker and Byzalov 2014).

13. The concept of public interest can be defined as both the importance of services (i.e., that are necessary and convenient for everyday life) and the roles and responsibilities of governments (e.g., Pesch 2006).

14. Saussier and Klien (2014) split MEs from the viewpoints of decision-making rights, organi-zational control, and property rights.

15. In some small municipalities, the mayor might double as the administrator instead of appointing an independent administrator to avoid the burden of additional payroll or in the cases of small business-scale MEs.

16. For details, refer to the Japanese government report ‘Evaluation, verification and analysis of “the municipal merger of the Heisei area,”’ published in 2008.

17. Sticky costs (anti-sticky costs) are defined as when costs decrease less (more) when sales activity falls than they increase when sales activity rises equally (Anderson, Banker, and Janakiraman 2003; Weiss 2010).

18. Operating income is taken from the profit and loss statement. The change in working capital is the difference of accounts receivable plus inventory less the difference in accounts payable. Capital expenditure equals the difference of tangible assets plus depreciation; both are taken from balance sheet statements.

19. We statistically verify whether the FCF variable is significantly different between merged and non-merged MEs. The t-test result shows the significant differences between them (t-value = 3.99; p-value < 0.01).

20. Chen, Lu, and Sougiannis (2012) found that the corporate governance variable drives down cost stickiness, and they concluded that corporate governance is an effective remedy for wasteful overspending behavior.

21. Deis and Giroux (1992) argued that the number of auditors determines the audit quality and originates good corporate governance for PSOs.

22. The number of auditors is stipulated by law, with four in each the 20 ordinance-designated cities and two in other cities, towns and villages. However, the fixed number of auditors can be increased by each municipality’s ordinance.

23. Cohen, Karatzimas, and Naoum (2017) and Lee, Pittman, and Saffar (2020) empirically found that sticky cost behavior is stronger during election years, and they argued that the political uncertainty surrounding elections can cause asymmetric cost responses to activity changes (i.e., cost stickiness).

24. The source of Japanese GDP is based on the following URL: https://www.esri.cao.go.jp/en/sna/data/sokuhou/files/2019/qe194_2/gdemenuea.html

25. This paper analyzes the following steps and then lists the optimal results in the results tables. First, we carry out the pooled ordinary least squares; we next estimate with the fixed effects model, which controls both the cross-section and year; then, we test the results with the F-test. As a second step, we estimate with the random effects model and test the results with Hausman test.

26. Based on the ‘law of 1/n’ approach, the Free-rider estimation appears only a few times before MEs are merged. As a result, the value of Free-rider reveals mainly a small impact in the descriptive statistics.

27. In fact, the average debt of the post-merged MEs was approximately double the average debt of the pre-merged MEs according to the Municipal Enterprise Yearbook.

28. Saarimaa and Tukiainen (2015) used revenue instead of population as a proxy for free riders.

29. Considering the high debt problems after the mergers, we also calculate free cash flow to equity (FCFE) as an alternative measure of FCF. Following the FCFE model specified by Damodaran (2006), FCFE is defined as net income plus depreciation minus capital expenditures minus the change in working capital and plus net changes in the long-term debt issued. Then, FCFE is scaled by total tangible assets. The definition is referred to in the Appendix.

30. Cohen, Karatzimas, and Naoum (2017) argued that the population, i.e., the inhabitants of a municipality, influences the cost management of PSOs based on the OECD 2008 report, and they added the population to their model by only employing a standalone variable. They did not test population intensity in the interactions with sticky costs. We predict that the population with the right to choose a politician would restrain sticky costs concerning mergers through rational decisions.

31. Cohen, Karatzimas, and Naoum (2017) employed the ABJ model with debt intensity to estimate the local Greek government cost behavior.

32. Before splitting the sample, we carry out statistical difference tests (i.e., t-test) for key variables such as operating costs (t = 2.42; p < 0.05), operating revenue (t = −1.58; p > 0.1), free-rider (t = 8.63; p < 0.01), and empire-building (t = −1.33; p > 0.1) between (1) Bureaucrats and (2) Politicians. However, we could not find significant differences for all variables. When comparing the coefficients, the same equation should be the preferred approach. This additional subsample analysis leaves room for improvement in analytical methods.

33. Jacobsen (2006) reports the local authority politicians and bureaucrats in 30 Norwegian municipalities and where differences in spending preferences were empirically investigated; then, as a result, his paper shows that bureaucrats seem to be less expansive than their political counterparts.

34. The literature on managerial incentives has shown that managers are likely to cut off slack resource costs when they must meet a particular performance benchmark (Dierynck, Landsman, and Renders 2012; Kama and Weiss 2013).

Additional information

Funding

This work was supported by the Japan Accounting Association under grant of the Japan foundation [English proofreading].