The effects of new franchisor partnering strategies on franchise system size.

1. Introduction

Organization theory and strategic management research have long held that a firm's ability to partner with other firms helps to increase the size of its operations quickly (Eisenhardt and Schoonhoven 1996, Stuart et al. 1999). The value of partnering has led to a wealth of research that examines how young firms attract other firms to work with them (Bruderl et al. 1992, Eisenhardt and Schoonhoven 1996, Powell et al. 1996).

Much of the research on the efforts of firms to attract partners has focused on their use of reputation, social ties to prominent actors, or the exploitation of observable assets to attract others (Stuart et al. 1999). However, as Bhide (2000) has observed, most young firms lack all three of these things and must attract partners through strategic actions. This paper discusses strategic actions by firms to attract partners, focusing on the empirical setting of business format franchising.

Prior research has shown that the size of a franchise system depends, at least in part, on the franchisors' ability to attract franchisees to work with them (Dant and Kaufmann 2003, Lafontaine and Kaufmann 1994). Franchisors can adopt specific strategies to facilitate the attraction of franchisees and expand system size (Gallini and Lutz 1992). In this article, we argue that the key strategies involve pricing policy--which we define as decisions involving royalties on sales to end users, up-front franchise fees paid by franchisees to franchisors, and initial investment made by the franchisees--and strategic control--which we define as decisions about ownership of outlets and the approach to financing those investments--because research has shown that these strategic decisions are central to securing channel partners, such as distributors, dealers, wholesalers, and retailers (Anderson and Weitz 1992).

Prior research in franchising has looked at determinants of pricing policy decisions (Agrawal and Lal 1995, Kaufmann and Dant 2001, Lal 1990), drivers of strategic control decisions (Dant and Kaufmann 2003, Dant et al. 1996, Kalnins 2004, Lafontaine and Kaufmann 1994, Lafontaine and Shaw 2005), determinants of survival of franchisors (Shane 1996, Shane and Foo 1999), and the influence of ownership mix on franchisor performance (Sorenson and Sorenson 2001). However, analysis of the influence of these strategic actions on franchise system size is limited. A notable exception is Srinivasan (2004), who examines how dual channels (owned and franchised outlets) affect a franchisor's sales. However, Srinivasan's (2004) variable of interest is sales per outlet, and her focus is dual channels, while our focal variable is the number of outlets in the franchise system and its relationship with franchisor's strategic actions.

One reason there is not much empirical research on the drivers of franchise system size is that research needs to carefully correct for three important issues: (1) the problems of sample selection due to franchise system failure, (2) the effects of unobserved variables on franchisor performance, and (3) the endogeneity of strategic decisions. First, many young franchisors fail, making it important to correct for such failure in empirical studies (Shane and Foo 1999). Second, much prior empirical research in franchising suffers from the confounding of the effects of unobserved variables (Barney 1991) with those of strategic actions. Thus, any evidence of the effects of strategic actions on performance in prior studies may be an artifact of these unobserved characteristics (Lafontaine and Shaw 1999). Finally, much previous empirical research in franchising does not treat decisions, such as those on pricing policies and strategic control, as endogenous, potentially biasing estimates of their effects on firm performance. Thus, rigorous empirical research is necessary to have an accurate understanding of the relationship between strategic decisions and franchise system size.

In this paper, we develop specific hypotheses about the effect of franchisors' strategic actions on franchise system size. We test the hypotheses using panel data on 1,292 business format franchise systems from 152 industries established in the United States between 1979 and 1996. We estimate a model of franchise system size that accounts for the endogeneity of royalty rate, franchise fees, and ownership; controls for unobserved firm heterogeneity; and controls for selection effects due to possible system failure. The results show that franchisors that grow large lower their royalty rates as the systems age, have low up-front franchise fees and raise them over time, own a small proportion of outlets and lower that percentage over time, make initial franchisee investment low, and finance franchisees.

Some of these results are counterintuitive. For example, the relationships we uncover between franchise system size and royalty rate and between system size and franchise fees contradict those proposed by Lafontaine (1993) and Lafontaine and Shaw (1999). Our results suggest that franchisors that become large lower their royalty rates as their systems age and charge low up-front franchise fees and raise them over time.

This paper proceeds as follows. The next section discusses the setting of the study. The third section provides a conceptual framework and presents the specific hypotheses that are tested. The fourth section presents the data, and the fifth section covers the model. Section 6 describes the results. The final section discusses the findings and outlines the implications, limitations, and conclusions for this study.

2. The Setting: Business Format Franchising

Franchising is an economically important form of entrepreneurship. The U.S. Commerce Department estimates that there are more than 500,000 franchised outlets in over 2,500 franchise systems, accounting for 13.5% of the U.S. gross domestic product and 35% of retail sales, and employing eight million workers (Lafontaine and Shaw 1999). Business format franchising exists in a variety of industries, from the Internet to banking, but is most common in eating and drinking establishments, business services, and retail (Lafontaine 1992, Shane 1996).

A business format franchise is a network of legally independent organizations that jointly exploit a common asset--the franchisor's plan for the provision of a product or service to end customers. Under a business format franchise arrangement, the franchisee obtains the right to use the franchisor's brand name and business plan in return for paying a royalty and franchise fees and agreeing to oversight by the franchisor (Shane and Foo 1999).

Franchise system size is of paramount concern to franchisors, especially young franchisors (Shane 1996). Not only is this a key managerial issue in marketing channels in general (Anderson and Weitz 1992), but it is also central to franchising as a business strategy (Shane 1996). First, as brand name is important in many franchised businesses and there are economies in advertising and promotion, the per unit cost of promoting the brand name is lower for larger systems. Therefore, many franchisors grow their systems to build their brand names more efficiently (Shane 1996). Second, many franchised businesses have a high fixed cost of development relative to the marginal cost of additional applications. As a result, a large system is important to reducing the average cost of opportunity exploitation and making the business more profitable. Third, larger franchise systems often have greater bargaining power than smaller systems. As a result, a large system provides franchisors with the ability to obtain lower cost inputs, thereby improving profitability (Gallini and Lutz 1992, Lafontaine 1993). Fourth, many of the new venture opportunities exploited by franchisors are unproven. To minimize the cost of bearing the uncertainty of new venture opportunities, franchisors often start their systems on a small scale and expand if they discover that demand exists and that they have the capabilities to meet that demand (Caves 1998). As a result, franchisors are often established below minimum efficient scale and need to become large to survive (Geroski 1995).

Franchised businesses involve high up-front fixed costs, exploit uncertain business concepts, and take advantage of size-based bargaining power, all of which encourage the adoption of strategies to expand system size (Shane 1996). Therefore, understanding which strategies are associated with large franchise systems is an important issue for management scholars to address.

3. Conceptual Development and Hypotheses

We propose a conceptual framework in which we identify the determinants of franchise system size. Although a number of variables potentially influence franchise system size, we focus on those determinants that involve strategic actions by the franchisor and treat other determinants as control variables. There are two broad categories of these strategic determinants of franchise system size: pricing policy and strategic control. Under pricing policy, the key strategic decisions of the franchisors include those of royalty rate, up-front fixed fees, and franchisee initial investment. A franchisee's initial investment includes capital expenditures on items such as real estate, training, equipment, and consulting fees. Under strategic control, the strategic franchisor decisions include proportion of ownership of the outlets and the decision to finance the franchisees.

The franchisor's decision variables also include franchisee training and advertising, but we do not focus on these variables for two major reasons. First, these variables can be viewed as recoverable by the franchisor as part of franchisees fixed costs (up-front fees and initial investment). Second, we do not have data on these variables to model their effects in our subsequent empirical analysis. But because we control for unobserved heterogeneity, omission of these variables does not pose a significant problem in our estimation of the effects of our focal interest. We develop hypotheses relating franchise system size to franchisor strategic decisions primarily based on a theory of signaling by the franchisor to the franchisee of the attractiveness of the franchise. The conceptual framework appears in Figure 1.

The size of a franchise system depends on the ability of the franchisor to attract franchisees after the franchise system is established. This process requires strategic actions on the part of the franchisor to signal the attractiveness of the franchise, because uncertainty and information asymmetry make it difficult for potential franchisees to discern which franchise systems are worth investing in. When they first begin to franchise, franchisors cannot easily demonstrate the value of their assets because the major asset that they offer to franchisees--the business format--is intangible and therefore hard to value prior to purchase (Gallini and Lutz 1992, Lafontaine 1993). For example, it is difficult to know in advance whether the recipes of a new fast food chain will prove to be popular with consumers.

[FIGURE 1 OMITTED]

In fact, many business formats offered by new franchisors are not valuable. Research has shown that most of the franchise systems established every year die quickly (Shane 1996, Shane and Foo 1999). This pattern makes potential franchisees skeptical of the value of business formats offered by franchisors when they first begin to franchise. Moreover, these franchisors cannot use their reputations as franchisors to attract franchisees, because they have not yet developed these reputations (Gallini and …