Different Uses of the Plural Form within the Franchised Networks:
Evidence from the US and French Marketsa Multi-Countries Empirical Study
Rozenn PERRIGOT1, Gérard CLIQUET2
1 Rennes International School of Business of Rennes (ESC Group), CREM UMR CNRS 6211
2, rue Robert d’Arbrissel, CS 76522, 35065 Rennes Cedex, France
2 Institute of Management of Rennes (IAE), University of Rennes 1, CREM UMR CNRS 6211
11, rue Jean Macé, CS 70803, 35708 Rennes Cedex, France
Abstract. Plural form has received an important flow of research works since Bradach’s model ( in 1998). Most of these works were developed in North American, Australian, or European countries. Their purpose was mainly to explain the degree of plural form within the franchised networks through economic, environmental, organizational, or marketing variables. In the present paper, we do not focus on plural form in one particular market but we compare the uses of the plural form within the US and French networks. The main result of the present paper stands in the fact that the rate of company-owned units is significantly higher in France (36.09%) than in the United States (9.45%). This finding can be analyzed through the differences in the territory area, and also managerial and strategic differences in the way retail and service networkschains are run in both countries. Differences in the use of plural form are also explored according to the retailing or services orientation of the networks. SomeOther determinants of the use of plural form are finally highlightedalso explored.
1 Introduction
In franchising research, plural form is a recent concept first announced by Bradach and Eccles (1989), then highlighted by several research works (Lafontaine and Kaufmann, 1994; Lafontaine and Shaw, 1999), and clearly defined by the Bradach model (1998) through the main challenges this concept is supposed to support. This model is actually based on meeting four challenges related to: 1) spatial expansion, 2) brand protection, 3) reaction against competition, and 4) service and/or retail concept evolution.
Many research works tried to focus on various aspects and/or applications of this concept of plural form and of this model. It was first defined through an exploratory research in US restaurants (Bradach, 1997). Other articles enhanced advantages and drawbacks of plural form networks in hotel industry, bakery sector or retail cosmetics in France (Cliquet, 2000), or compared plural forms or tapered integration (Bradach and Eccles, 1989) with other theoretical approaches such as signalling theory and resource based theory (Dant and Kaufmann, 2003), property right and transaction cost theories (Windsperger, 2004a; 2004b), or the theory of incentives and the agency theory (Chaudey and Fadario, 2004). Some articles focused on particular elements of the model such as innovation (Lewin-Solomons, 1999; Cliquet and Nguyen, 2004), the organizational learning process (Sorensen and Sørensen, 2001), or the royalty rate (Pénard, Raynaud, and Saussier, 2003). Ehrmann and Spranger (2004) examined cost reduction, quality enhancement, growth stimulation, and optimized risk control.
These research works were developed in various countries (Australia, Austria, France, Germany, Spain, and for the most important part in the United States). But as far as we know, there is no attempt to compare the situation across countries. This paper aims at comparing the degree of plural form, or plurality, measured by the rate of company-owned units between France and the United States and France. Do US and French franchisors use company arrangements within their networks in a same manner? Or are there any differences in this use of plural form within the US and French networks? and to test some variables likely to explain the differencesDifferences according to the sector (retailing vs. services) are also explored and some determinants of the plural form are exposed.
The paper is organized as follows. In the next section, we present the main approaches used to explore the plural form within the franchised networks. Then, in section 3, the research design, and more specifically data, variables and methodology, are explained. Section 4 depicts the main results of the comparative analysis of the US and French networks. Finally, the results are discussed in section 5.
2 Literature ReviewMain approaches of the plural form concept
The plural form concept has been studied in detail for about four years, albeit often in an exploratory manner. Four important models can be distinguished:
- econometric models drawn from economic research,
- channel management models from accounting data,
- models of rupture in the franchise process,
- models founded on spatial considerations drawn from management sciences, or management and marketing, or even geography.
The first models rely on regression analyses and econometric techniques using databases. These are purely statistical studies which aim at proving the hypotheses established from theories such as the agency theory (Shane, 1998a), or founded on the study of certain concepts such as the importance of the brand name (Lafontaine et Shaw, 1999), or the elements of the marketing mix such as price (Lafontaine, 1998) or advertising (Michael, 1999). The dependant variable is generally the proportion of franchised and company-owned units. They are the global models of store network study. The second category of models has been initiated by Kaufmann, Gordon and Owers (2000) and relies on the notions of accounting and economic value. The third manner of modeling the plurality of forms uses the study of ruptures in the franchise process (Frazer, 2001). The recent nature of these two last categories does not allow in-depth development as only one model has been developed in each. Finally, the fourth kind of models studies particular aspects of the management of store networks: innovation (Sorenson and Sǿrenson, 2001) and location (Ghosh and Craig, 1991).
This section is split into two paragraphs: the first one describes the econometric models and the second one deals with the three other kinds of models.
2.1 Econometric models
Most of the econometric works linked to plural form have been carried out in the United States so far. Many have lead to non significant connections. This is the case in works based on networks belonging to 10 sectors of activity over 10 years, which show that only franchisee sales by establishment explain the percentage of company-owned units (Wade, O’Hara and Musgrave, 1990). This tends to indicate that it would be in the interest of the franchisor to internalize and therefore increase the percentage of company-owned units because his/her return rate would be inferior to that of his/her franchisees. This may explain the behavior of certain franchisors who own the large establishments and let the smaller ones to the franchisees. Research works on the level of salaries tend to confirm this tendency insofar as the salaries of employees in company-owned units are higher and increase more rapidly than their counterparts in franchises (Krueger, 1991). More recent studies have shown that the proportion of franchises in retail trade are positively connected to size and geographic expansion and negatively associated with the rate of growth and the level of investment (it will not be surprising that neither age nor royalty rates are significant) (Alon, 2001).
In fact, modeling studies connected to the relative proportion of franchisee activity compared to that of company-owned started in the mid-1980s. Then, based on a rather simplistic first model (O’Hara and Thomas, 1986), Thomas, O’Hara and Musgrave (1990) develop a model with a dependent variable as a ratio between franchisor sales by unit and the total sales by unit of the franchise. The explanatory variables are the degree of vertical integration, measuring the relationship between the number of company-owned units and the total number of units, the rate of conversion, or the relationship between passing into franchise and the total number of ownership changes, the rate of repurchase, or the relationship between the number of units repurchases by the operator and the total number of ownership changes, with a dummy variable measuring the difference of investment, equal to 1 if the median investment of the franchised units is superior to that of the company-owned units, equal to 0 if not. The main result of the model, carried out in 10 sectors of activity over 10 years, is that, when there are too many company-owned units, losses are noticed. This invalidates the life-cycle argument according to which one repurchases the franchises in the maturity stage (Oxenfeldt and Kelly, 1968-69). The authors also test a vertical integration model with a dependent variable measured by the degree of vertical integration, measuring the relationship between the number of company-owned units and the total number of units. The explanatory variables are: intra-system sales by franchised units (total of franchisor sales to franchisees divided by the number of franchised units), sales by franchised units, and the same dummy variable as above. Only the franchised unit sales are significant and positive concerning the degree of vertical integration. This means that the success of the franchisees encourages operators to own units.
However, in the end, these models are less than explanatory in a more managerial arena. It is therefore necessary to examine other models. Three categories of models will be studied in more detail. The first also deals with the proportion of franchises/company-owned units but with more variables. The second concerns the agency theory and the third studies the connection between certain elements of the marketing mix.
Brown (1998), Lafontaine (1992) and Lafontaine and Shaw (1999) have published works of an econometric nature concerning the plurality of store networks. The first article is based on the transaction cost theory. The two last ones deal with the impact of different environmental and organizational variables on the proportion of franchises/company-owned units within plural form networks. Using multi-sector data, Lafontaine shows that the proportion of franchises rises with the geographic dispersion, the rate of growth and the age of the network. She also explores, in the same article, the determinants of the rate of repurchase in the franchise contracts. She notes that the econometric estimations better explain the proportion of franchised units than the terms of the franchise contracts. However, this proportion decreases with the average sales and the capital invested per store.
Using the transaction cost theory, Brown (1998) shows that the firm leans toward a long-term equilibrium between the proportions of franchised units and company-owned units. In fact, it uses a more efficient system of internal promotion in order to motivate employees and leaves it up to the franchisees to motivate their employees who would be inevitably disadvantaged with contracts founded on performance. These kinds of contracts would implicate high monitoring costs for the operator.
Lafontaine and Shaw (1999) use a sample of more than 4,800 North American store networks from 1980 to 1997. For each network, it is possible to know the creation date, its number of years of franchising experience and its domain of activity. Furthermore, for each year, the authors know the number of company-owned units versus the number of franchises, the rate of repurchasing and the up-front fees. Using these data, it is noticed that after eight years of franchise experience, the networks tend to stabilize themselves: after declining in the first few years, the proportion of company-owned units maintains an average between 10 and 20%. This stability is found regardless of the sector, the networksize or the rate of growth. However, large differences exist from one firm to another one or from one sector to another one in respect to the target of managerial control. The target company-owned proportion in the restaurant industry is around 20%, which is far higher than that of construction and maintenance services (5%) or of car repair (10%). Moreover, networks offering services have a higher proportion of franchises than sales networks. Lafontaine and Shaw also show that the rate of company-owned units increases with the worth or quality of the brand name.
Studies by Spanish researchers are currently underway in order to shed light on the existing relationships between the rate of franchised units, networksize, brand recognition and sector popularity (López, Gonzàlez-Busto, 2001). It emerges from this research that, when faced with the simultaneous development of brand recognition and sector popularity, the number of franchisees increases, albeit to the detriment of the rate of company-owned units. This imbalance tends to undermine concept control and therefore risks affecting the image of the brand. This was already pointed out in a previous exploratory research (Cliquet, 2000). The necessity of counterbalancing this disproportion of franchised units imposes the opening of new company-owned units, which in turn increases the degree of integration, in order to preserve the uniformity of the network (Bradach, 1998). Furthermore, a virtuous cycle is created between the degree of integration of the networkchain, the brand recognition, networksales, franchisor profits and the number of company-owned units (López, Gonzàlez-Busto, 2001). The opening of company-owned units has a positive effect on brand recognition and concept uniformity which in turn affects sales. These relationships are studied using data from 5,000 American and Canadian networkschains, established from 1980 to 1997, taken from the works of Lafontaine and Shaw (1999) and compared with data collected in Spain. This research attempts to model the rate of company-owned units (or franchised units) while taking the first two of Bradach’s (1998) challenges, adding new units and maintaining uniformity across units, into account. The pattern studied can be summarized as follows:
Speed of the franchise development of the network=> notoriety of the brand => attraction of new franchisees => loss of image of the brand => opening of company-owned units maintaining conformity across units => improvement of brand image => attraction of new franchisees => rapid development => etc…
This reasoning applies to developing networks. The introduction of the two other challenges, local responsiveness and system-wide adaptation, would require more mature networks to take at least two other supplementary concepts into account: the incentives that a franchise system can provide, concerning responsiveness, and a veritable brand management for the adaptation. Actually, the responsiveness can also concern the development phase (as well as the maturity phase), and only the challenge associated with system-wide adaptation more closely concerns the mature networks. While incentives favor the maintaining of a rather high rate of franchised units, the restrictions of an acceptable growth of return require the operator to develop company-owned units, either by opening new stores or repurchasing franchised ones. The necessity of a brand management imposes the plurality of status, for, while the company-owned units favor uniformity, the franchises allow for better local responsiveness and often allow a company to take advantage of numerous and simultaneous opportunities. The thinking developed from the works of López and Gonzàlez-Busto (2001) also leads up to the idea of a stabilization of franchise/company-owned proportion.