Introduction
Japanese convenience-store (CVS) chain retailers have grown by establishing store networks. In fact, Seven-Eleven, Lawson and Family Mart continuingly have opened about 1000 new stores per year. The reason for the rapid growth of their store-networks is that a key aspect of a chain retailer’s marketing strategy is the number of stores its needs to reach its customers (Srinivasan et al. 2013). In particular, CVS chain retailers seek to open new stores and obtain spatial dominance in a particular geographical area, which is called “area-dominance strategy,” so they can save on logistical costs, increase consumer proximity and loyalty, and prevent rival from opening new stores in the area (Ogawa 2004, Tamura 2014, Nishida 2014). Thus, a retailer’s decision of how to expand store-network in a given regional market is important to improve its sales. However, little attention has been paid to this problem in Japanese academic research. This study attempts to explain the influence of entry of rivals on a focal retailer’s store-network in regional markets of Japanese CVS industry. Especially, the author focuses on the regional competition between a focal retailer who is the first entrant and rivals who are late entrants in the region. First, we review prior research, and then propose hypotheses about the influence of entry of rivals, the degree of dominance of a focal retailer, and entry of rivals in multiple regions, on the number of the focal retailer stores. This is followed by an empirical analysis with panel data. Last, we discuss some implications and direction for future research.
Literature Review and Proposed Hypothesis
Entry of rival stores
Prior research suggests that the existence of rival chain stores in the same market decreases the store-revenue of the focal retailer (Erickson et al. 2013, Nishida 2014).When rival retailers open the large number of new stores in a regional market, the focal retailer may be taken away their business of existing stores and latent new stores, so the focal retailer will be forced to close existing stores and slow down the pace of opening new stores. Then, we propose following hypothesis:
H1: In a regional market, the higher the number of net increase of rival stores is, the lower the net increase of focal retailer stores is.
Dominance of the focal retailer
According to prior research, CVS chain retailers benefit from area-dominance (opening own stores aggressively in a given region), because it enables retailers to reduce their distribution and promotion cost, increase consumer loyalty and their power against manufactures (Ogawa 2004, Tamura 2014, Nishida 2014). If the focal retailer has already established a high density store-network, and had strong relationships with its customers and manufactures in the regional market, they will be less likely to suffer from entry of rival chain stores, and they will be able to continue expanding their store-network. These arguments lead to:
H2: In a regional market, the higher the degree of dominance of the focal retailer is, the smaller the negative effect of entry of rival stores is.
Entry of rival stores in multi-market
Though H1 and H2 do not consider multi-market competition among chain retailers, this macro-level competition may have a great impact on their competitive action in a given region (Chen 1996). When a rival entries to multiple regions simultaneously where the focal retailer has already operated, the focal retailer will delay its decision making and competitive responses, so the impact of entry of rival will be larger (Poter 1980, Ferrir 2001, Boyd and Bresser 2008). Therefore, we propose following hypotheses:
H3a: The higher the number of regions which rival entry is, the bigger the negative effect of entry of rival chain stores is.
H3b: The higher the number of net increase of rival stores across the regions is, the bigger the negative effect of entry of rival chain stores is.
Methodology
To test the proposed hypotheses (see FIGURE 1), we collected panel data from the Census of CVS Market, which includes the number of stores of Japanese CVS chain retailers in each prefecture. Due to the restriction of data availability, we treated prefectures as the unit of regional market, and focused on the cases that Lawson was the first entrant, and Seven-Eleven and Family-Mart ware the later entrants in prefectures. Accordingly, the sample was limited in space to 17 prefectures, and limited in time to the period from the year that Seven-Eleven or Family-Mart opened their stores for the first time to 2014.
Results
We tested the hypotheses using panel date analysis by fixed effects model. The estimated results are shown in FIGURE 2. Regarding our hypotheses H1, involving the negative effect of rival entry on the focal retailer’s store-network, is not supported. However, the interaction of “the dominance of the focal retailer” with “rival entry” and the interaction of “rival entry in multiple regions” with “rival entry” are significant, although their signs of coefficients differ depending on whether the rival is seven-eleven or family-mart. Thus, hypotheses H2 and H3b are supported in part.
Implication and Future Research
Our findings have several important implications. First, our empirical results suggest that the effect of rival entry on a focal retailer’s store-network depends on (1) rival’s position in CVS industry, (2) the focal retailer’s dominance, and (3) rival’s multiple entry. Second, when a rival has a superior position than the focal retailer, dominance advantage of the focal retailer increases the negative effect of rival entry, which is contrary to our expectation. This implies that enhancement of the density of own store-network will cause cannibalization, so each store of a focal retailer may be highly vulnerable to entry by a rival who has superior competitive position (i.e. Seven-Eleven). Finally, multiple entry by a rival in superior position reinforces the negative effect of their entry on expansion of the focal retailer’s store-network in the regional market. Though this study was a rare attempt to explain regional competition among Japanese CVS chain retailers empirically, it did not include the prefectures that Seven-Eleven and Family-Mart were the first entrants. This may limit generality of the empirical results, hence it is valuable to take this problem into consideration in future research.

• Minami Nishikawa(Chuo University, Japan)