Competition for territories under the switching cost

ABSTRACT

This study considers an effective policy to mitigate market powers held by retail liquefied petroleum gas distributors in Japan. A bilateral spatial competition model is constructed to analyse the effect of switching costs of consumers on retail competition in designated areas where stores provide free delivery service at uniform prices. Model analysis shows that, at equilibrium, stores have exclusive delivery areas which are their natural territories given endogenously. If stores commit to their territories, promoting competition with the adjacent market would be effective. Whereas, if stores do not commit, policies to lower the switching cost are effective.

KEYWORDS:

• Free delivery service
• switching cost
• undercut proof equilibrium
• retail territory

JEL:

• D43
• L13
• L81

ACKNOWLEDGEMENTS

An earlier version of this study was presented at the 12th International Conference on Applied Business and Economics at the University of Cyprus, Cyprus, European Association for Research in Industrial Organization 2013 annual conference at Evora, Portugal, and other seminars. I am thankful for the helpful comments by participants. I would like to express my sincere gratitude to the anonymous reviewers for their insightful and constructive comments, which have significantly contributed to the improvement of this manuscript. Additionally, I am thankful for the professional language editing services provided by Editage, which greatly enhanced the clarity and readability of my work.

AUTHOR CONTRIBUTION STATEMENT

The author confirms sole responsibility for the conception and preparation of this manuscript, including the design, analysis, and writing.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author(s).

ETHICS STATEMENT

This research did not involve human participants, animals or clinical trials, and therefore did not require ethical approval.

Notes

1 Note that at those second stages, both stores may earn positive profits for all possible phases in competition even if a store has no exclusive DA and the price is unstable, because its rival store need not lower its price as low as to bring zero profits. If the rival store has its exclusive DA, the rival store always has an option to set a high price and secure a positive profit.

2 Byford (2015) showed that the UPE is the core of a cooperative game, although the game analysed here does not satisfy the condition required.

3 Since Smithies (1941), ‘territories’ has been sometimes used to describe the set of locations of consumers who prefer to buy from a store in spatial competition models. Here, the exclusive supply areas are those the rival stores would not intrude on even when consumers want to buy from rival stores. In more recent literature, ‘territory’ refers to a type of vertical constraint that is used by an upstream firm to mitigate competition among downstream firms. See Mathewson and Winter (1984), Rey and Tirole (1986) and Rey and Stiglitz (1995) among other notable papers on this subject.

4 ‘Liquefied Petroleum Gas Guide 2017’, LP Gas Center, http://www.lpgc.or.jp/corporate/information/guide.pdf

5 See Ministry of Economy, Trade and Industry (2013). By the author’s calculation on the data from ‘The Oil Information Center, The Institute of Energy Economics’, while the standard error and the mean of the average prices in 47 prefectures is JPY 1090 and 13,099 per 20 squares metres use of LP gas, the mean of the ratios of the difference between the maximum and the minimum to the mean is 0.498 for 47 prefectures. The maximum of the ratios is 1.02.

6 ‘Survey Report on Consumers Behavior of Petroleum Gas’, Nippon Consultant Group Inc., https://www.kanagawalpg.or.jp/images/201509_shouhishajittai.pdf

7 ‘Consumer Consciousness Survey on LP Gas’, National Federation of Regional Women’s Organization, http://www.chifuren.gr.jp/kikanshi/news-bk/380/newsback-380_5.html

8 ‘Survey Report on Trade Practice of Retail LP gas in 1999’, Japan Fair Trade Commission, http://www.jftc.go.jp/info/nenpou/h10/02070005.html

9 Regarding restrictions on customer movement, the Annual Report of Japan Fair Trade Commission, 1998, admits widespread practice of free piping, stating that ‘If customer movement is restricted due to the practice of free piping, it may cause problems under the Antimonopoly Law’. https://www.jftc.go.jp/info/nenpou/h10/02070005.html

10 See, for example, Kido (2002). In 2018, The Fair Trade Commission issued a cease and desist order against a regional LP gas trade association in Kanagawa Prefecture that attempted to prevent new entrants from entering the market to gain customers from incumbent businesses. https://www.jftc.go.jp/houdou/pressrelease/h30/mar/180309_1.html

11 A survey report in 2014 shows 46.4% of consumers did not notice that they have a choice of LP gas provider. https://www.kanagawalpg.or.jp/images/201509_shouhishajittai.pdf

12 The reason for assuming a reservation price here is that the model is designed to be applicable to the Japanese LP retail market. The Japanese retail energy market is particularly competitive, especially among energy sources: Kerosene and electricity are considered to compete with the LP retail market, and consumers are clearly aware of the competitive price levels for each. However, city gas has a limited service area and is not competitive with LP gas.

13 These assumptions are made to simplify the model. In the presence of a constant production cost, it would be necessary to incorporate that cost into the equilibrium price. Since we focused on the Japanese LP gas retail market, we are not considering any product differentiation apart from consumer location.

14 Although it is easy to start with a more basic assumption, for the purpose of saving space, we assume the convexity of the DAs.

15 The proofs of the propositions and lemmata discussed in this study are presented in Appendix C in the online supplemental data. Readers interested in the detailed derivations and mathematical justifications can refer to the corresponding sections in the appendix.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science Grant Number JP23530314.