Franchising as a plural system: A risk-based explanation

Abstract

Empirical studies show that most franchise systems consist of both franchisee-owned and franchisor-owned units. We contribute a new theory that explains why such a mixture exists, using a model that focuses on the franchisor's optimal risk allocation. The costs of risk and controlling franchised units explain the varying fraction of franchisee-owned to total selling units, and the incentive to franchise decreases with an increasing fraction of franchisee-owned to total selling units, as well as with decreasing costs of control. Our explanation for these plural systems is consistent with the ownership redirection hypothesis.

Introduction

When buying at an outlet, most customers do not know whether it is a franchisee- or franchisor-owned branch. According to franchisors and franchisees, this situation marks a success, because it indicates that the franchisor has successfully transferred its brand name to the franchisee. A naïve assumption might suggest a system containing uniform outlets uses either a pure franchise system or a pure company-owned system, depending on which form proves superior in the particular conditions. However, franchise systems usually contain some company-owned outlets; that is, franchise systems generally are plural systems rather than pure franchise systems. Bradach (1997) reports that of the 100 largest U.S. restaurant chains, 74 employ plural systems, 22 are systems with exclusively system leader-owned units, and only 4 represent pure franchise systems.

This article contributes a new explanation for the existence of plural systems that specifically considers the fraction of franchisee-owned outlets versus the fraction directly operated by the franchisor. Several existing theoretical and empirical studies address the ownership patterns of franchise systems, mostly by examining changes in ownership patterns as franchise systems mature (Dant and Kaufmann, 2003, Dant et al., 1992, Dant et al., 1996, Hunt, 1973, Lafontaine, 1993, Lafontaine and Kaufmann, 1994, Lafontaine and Shaw, 1999) on the basis of various theoretical backgrounds and concepts, such as the resource acquisition (Caves and Murphy, 1976, Norton, 1988, Oxenfeldt and Kelly, 1968) and signaling (Gallini and Lutz, 1992, Lafontaine, 1993) theories of franchising.

However, no previous studies attempt to determine the percentage of franchisees in the system by taking into account the franchisor's risk considerations. We examine how franchisor risk and related costs develop when outlets convert from franchisee- to franchisor-owned units or vice versa and base our findings primarily on the franchisor's risk assessments, which have been largely ignored in the literature thus far,¹ even though risk assessments play significant roles in explaining unit-level decisions. According to agency theory, the franchisor's goal conflict consists of stimulating a selling incentive versus allocating risk efficiently, which prompts fixed transfer payments and sales shares (Lafontaine, 1992, Lal, 1990). In this sense, the typical assumption that franchisees are more risk averse than franchisors appears plausible because of the common relative proportions of size by the two parties.

We focus on the risk reduction the franchisor can achieve when it substitutes a franchisor-owned with a franchisee-owned unit, which it does as long as its costs of control remain sufficiently low. This conceptualization may seem counterintuitive at first, because risk is shifted to the more risk-averse franchisee, which should increase the cost of risk for the whole system. However, from solely the franchisor's perspective, increasing the percentage of franchisee-owned units decreases its risk and thereby reduces the costs associated with financing, in that the franchisor substitutes part of the variable income it earns from its franchisor-owned units to fixed income it obtains from franchisee-owned units. If the franchisor minimizes the costs associated with risk taking and control of the system, the preferred system structure will include both franchisor- and franchisee-owned units, and the fraction of franchisee-owned units should depend on the costs the franchisor incurs to take risks and control the system.

We organize the remainder of this article as follows: In the next section, we provide a brief overview of extant literature that attempts to explain the existence of plural systems. Then, we examine whether it is possible to make a definitive statement about the optimal fraction of franchisee-owned units, or the optimal ratio of franchisee-owned to the total number of outlets. Subsequently, we develop a new approach that explains the combinations of franchisee- and franchisor-owned outlets in one system. Finally, we clarify why such risk transfer within the framework of a franchising system affects the fraction of franchisee-owned units and argue that it strongly supports the introduction of a plural system.