فرانچایزینگ چند واحدی بین المللی : توضیح نظریه نمایندگی

In spite of its considerable popularity, multi-unit franchising has been considered an anomaly from an agency theory perspective. This paper addresses this anomaly by attempting a comprehensive agency theoretic explanation of international multi-unit franchising. Although past agency theoretic examinations have mostly focused on single-unit franchising, a closer examination of the international context, which is characterized by significant geographic and cultural distance between franchisors and franchisees, suggests that multi-unit franchising may be more appropriate. This paper examines several agency problems inherent in multi-unit franchising. These include: bonding, adverse selection, information flow, shirking, inefficient risk-bearing, free-riding, and quasi-rent appropriation. Consideration of each of these problems using agency theory perspective leads to the suggestion that multi-unit franchising might address agency problems better than single-unit franchising in the international context.

Agency theory has been a popular perspective to investigate issues in franchising (Bates, 1998 and Shane, 1996a). Agency theory arguments of franchising, however, focus on single-unit franchising (SUF): they are “uniformly couched in terms of the local management of single-unit operations” (Kaufmann & Dant, 1996: 346). This preoccupation with single unit franchising seems surprising, given that multi-unit franchising (MUF) is ubiquitous, especially in location-based industries where franchising is most popular. For example, using Bureau of the Census data, Bates (1998:114) noted that “among the recently started restaurant franchise units, 84% were units of multi-establishment corporations”. Kaufmann and Dant (1996: 346) note several other examples leading to the conclusion: “It is clear that the typical location-based franchising system (of which the fast-food franchises are the prime and modal example) is populated with multi-unit franchisees”. In spite of the widespread prevalence of MUF, it “remains a curious anomaly” (Kaufmann & Dant, 1996: 346), without a clear theoretical explanation. Prior research, by and large, has focused exclusively on single-unit franchising on the assumption that agency problems are resolved by having owner operators. The moment they become multi-unit systems, individual units are managed by employees, thereby reintroducing the agency problem. The objective of this paper is to explain this anomaly, using an agency theoretic perspective. We do this in the specific context of international franchising, where MUF appears to be particularly dominant. That is, we examine the question: Why do international franchisors prefer MUF over SUF? By examining the rationale for MUF from an agency theoretic perspective, we complement the work of Kaufmann and Dant (1996) which suggests that MUF, though an anomaly from an agency theoretic perspective, has merits over SUF as an entry mode from a capital acquisition perspective. Given the success of agency theory to explain various aspects of the principal-agent relationships that characterize the interactions between a franchisor and a franchisee, we believe that this theoretical perspective should be able to explain the preference international franchisors have demonstrated for MUF. Eisenhardt (1989: 59) succinctly expressed the wide scope of agency theory when she wrote: “[T]he domain of agency theory is relationships that mirror the basic agency structure of a principal and an agent who are engaged in cooperative behavior, but have differing goals and differing attitudes toward risk”. This scope is no exaggeration, as suggested by a large volume of both conceptual and empirical work in diverse areas (see Sharma, 1997, for some examples). Therefore, the status of MUF as an “anomaly” seems attributable more to the lack of a focused, in-depth theoretical attempt to explain MUF than due to any inadequacy of agency theory itself. Multi-unit franchising is different from single-unit franchising in that franchisees own, operate, or control more than one outlet (Kaufmann & Dant, 1996). Even within MUF, there are many different types. Incremental or sequential MUF occurs when a single-unit franchisee within a successful system is awarded additional units based upon the performance of existing units. Master franchising, on the other hand, involves the franchisor granting a franchisee rights to multiple units from the outset (Kaufmann, 1992). Area development agreements are a special type of master franchising which requires the master franchisee to open and operate multiple units within a geographical area according to a prespecified schedule. Individual outlets, in turn, are run by employee managers hired by the master franchisee (Kaufmann & Dant, 1996). In this paper we attempt to take an important, long overdue first step towards developing an agency-based explanation for MUF. Our specific focus is on international franchising, because (a) international franchising is growing much faster than domestic franchising (Shane, 1996b) and (b) this context offers a unique setting for the comparative evaluation of the magnitude of agency problems associated with MUF and SUF. In the domestic context, we see instances of both single-unit and multi-unit franchises. We assume that the choice, therefore, is essentially driven by industry-specific factors. In the international context, on the other hand, the preference appears to be predominantly for multi-unit systems. Therefore, we assume that internationalization introduces a set of management considerations that are unique to foreign operations. Examining the rationale for MUF in the international context would therefore permit extensions and useful insights for the future development of a more generalizable theory of multi unit franchising. For example, just as capabilities developed to address agency problems in the domestic context have proven valuable while expanding internationally (Shane, 1996b), we believe that capabilities and insights developed through international franchising may also lead to improved management of franchising operations at home. The rest of the paper is organized as follows. First, a review of agency theory perspective of franchising is presented. Then, we examine international multi-unit franchising from this perspective. In that section, we discuss the following agency problems: bonding, adverse selection, information flow, shirking, inefficient risk-bearing, free-riding, and quasi-rent appropriation. Finally, implications for research and practice are discussed.

In this paper, we have developed arguments suggesting why IMUF would better address agency problems than ISUF. These arguments have several implications for both theoretical and empirical research in the future. Some of these promising directions for further research are presented in the following paragraphs. First, in this paper, we have treated MUF as homogeneous. There are, however, several types of MUF. Recently, researchers have been paying more attention to the different types of MUF (Bradach, 1995, Kaufmann and Kim, 1995 and Kaufmann and Dant, 1996). These types include sequential multi-unit franchising, master franchising, and sub-franchising. An examination of these alternative forms of multi-unit franchising from a more comprehensive theoretical framework such as the one attempted in this paper should complement the extant explanations for the prevalence of multiple types of MUF. In particular, examining why any one type better addresses some of the agency problems than other types might aid practitioners in making intelligent choices that best address the most crucial agency problems in a given context. For instance, when a franchisor has to decide about opening a new unit in a foreign location where it already has a franchise, it might be argued that sequential multi-unit franchising addresses the adverse selection problem better than does master franchising (Kaufmann & Dant, 1996). This is because the former provides system-specific information about the capabilities and behaviors of a franchisee, whereas in the latter, franchisors have to rely on external sources of information, mostly provided by an aspiring franchisee. External information can be costly and relatively less reliable than can be obtained through a close working relationship with a franchisee. The information provided by franchise applicants, on the other hand, may be less reliable due to incentives to misrepresent abilities and traits. Thus, in overseas markets where opportunism is a more manifest aspect of local culture (see Dant & Nasr, 1998, for some examples), international franchisors might prefer sequential multi-unit franchising to master franchising. On the other hand, where speed is more critical and opportunism is relatively incompatible with local culture, master franchising may be more suitable. In situations where speed, as well as opportunism are critical issues, franchisors might install a system based on sub-franchising. This might involve a trade-off between sharing the returns with one more layer of agents and the advantages of having agents that might be geographically and culturally less removed from the foreign market than might be the franchisor. Second, the preference for one type of franchising over another may also be driven by the differences in industries. For example, the need to protect brand name capital is likely to be more strongly felt in those industries (e.g., fast food restaurant, hotel chains) that provide services that are difficult to protect against imitation through legal means such as patents (Fladmoe-Lindquist & Jacque, 1995). This suggests that the agency problems requiring monitoring would be differentially critical across industries. In turn, the relative advantage of IMUF over ISUF would also vary from industry to industry. Industries may also differ in terms of the location of specific information within the system. For example, some industries (e.g., computer software) are characterized by a strong service component (McNaughton, 1996). Computer software industry, an example of knowledge intensive high technology sector, requires substantial customization, which, in turn requires investments in R&D. These investments may be specific to a firm’s products, but may be appropriable over similar customers of the firm in a country. A single franchisee serving a small market in a foreign country may have less incentives to invest in such knowledge based specific investments compared to a master franchisee serving a large number of similar customers spread over a wide region in a foreign market. Thus, by permitting centralized specific informational investments that can be appropriated over a large number of customers, IMUF may be more attractive than ISUF in some industries. Other industries might be more labor intensive rather than knowledge intensive. Fast food restaurant industry would be one such example. In labor intensive industries, monitoring problems are more critical due to increased incentives to shirk and shade quality. This suggests that there may be an interaction among agency problems, particularly in a given context. The possibility of such interactions suggests that exploring more complex explanations of IMUF would aid practicing franchisors, and therefore, could be a fruitful future research area. Third, although our focus in this paper has been the international dimension of franchising, an even more productive avenue for future research may be the simultaneous consideration of both the international dimension and industry specific factors within the context of multi-unit franchising. While empirical studies in franchising (e.g., Shane, 1996a and Shane, 1998) tend to include a large number of industry dummy variables to control for any possible industry effects, a clear theoretical rationale for hypothesizing such effects is still not available. Developing such a theoretical rationale may be fruitful. For example, the problem of adverse selection can vary across industries. A larger pool of candidates for franchising licenses may be available for a very visible and popular industry such as fast food service than for another not so visible and popular (e.g., house cleaning industry). A larger pool would require a more careful screening on the part of an international franchisor. In turn, the franchisor may have to incur higher screening costs to minimize the probability of adverse selection. Extending the arguments presented earlier in the paper suggesting that adverse selection problems would be less in IMUF than in ISUF, it follows that IMUF will be more prevalent in highly visible and popular industries than in less visible and popular industries. Another agency problem we studied is that of post-contractual information asymmetry. This problem may also vary across industries. In some industries (e.g. remodeling industry), more detailed, qualitative information concerning buyers’ unique needs and decision processes may have to be exchanged between the franchisee and franchisor to develop products and pricing strategies than in other industries (e.g. fast food) where the information is not very “sticky”. Sticky information is that information which is not readily packaged for dissemination (Dant & Nasr, 1998). Again, extending the arguments presented in the paper suggesting that IMUF is less susceptible to distortion or retardation of information than ISUF, it follows that IMUF will be more prevalent in industries characterized by higher stickiness of information concerning the buyer behavior. Stickiness may also be a function of the country environment. In some countries, such information may be readily available while in other countries collection and dissemination of such information may present considerable challenges. Fourth, although we have discussed each of the agency problems separately, it is likely that they are highly interrelated. An example would be the interrelationship between monitoring and residual claimancy via adverse selection. Residual claimancy provides higher return than fixed employment wages only when the agent is qualified (Shane, 1996a). When the agent is not qualified, the reverse may be true. Unqualified individuals may either not be aware of this complex relationship, and therefore, may misrepresent their abilities to franchisors, or may honestly overestimate their qualifications for franchising. In any event, pre-contractual information asymmetry might lead to adverse selection, and unqualified individuals may become franchisees. Subsequently, realizing that their residual returns are not any higher than employment wages, such unqualified franchisees then might regress to opportunistic behavior due to external attribution (most likely, to franchisor), survival instinct, or a combination of both. In turn, franchisors may have to incur higher monitoring costs to curb such opportunism. Scenarios such as the one described above are better addressed by MUF than SUF, especially in the international context. Following the arguments presented in the body of the paper, there would be more returns to pre-contract information search for both franchisor and franchisee in IMUF than in ISUF, so that the chances of adverse selection, and resulting monitoring problems, would be reduced. Fifth, future research can attempt a more nuanced explanation of IMUF based on cultural differences across nations. One of the crucial aspects of agency theory is the assumption that agents would act opportunistically. However, the extent of opportunism may vary across cultures. For example, Dant and Nasr (1998: 15) suggest that in Lebanon, there was anecdotal evidence suggesting that franchisees sometimes attempted to bribe the auditing market research firms in order to provide franchisors with rosy, inaccurate market information (Nasr, 1995). Clearly, a very high level of agency problems addressed in our paper would exist in a foreign market like Lebanon. On the other hand, Churchill Coffee Company USA, a coffee shop chain based in Springfield, Missouri has recently granted a license to a Japanese company, JEL, for opening up to 300 coffee shops. Based on their experiences, the company feels that trust can be easily taken for granted in business deals in Japan. Often, contracts are either non-existent or merely outline the agreed terms. Therefore, Churchill feels that there is very little need for monitoring the conduct of a Japanese franchisee. This suggests that agency explanations would have a limited role to play in a culture low in opportunism. Thus, we would expect differences in the extent to which IMUF is used across international markets. Sixth, we have exclusively focused on agency theory and as a result, our arguments have not predicted ISUF. However, the existence of international single-unit franchising can be explained using other theoretical perspectives. For example, literature on power and conflicts in distribution channels suggests that an explicitly-defined power structure exists in franchise systems (Gaski, 1984). Hunt and Nevin (1974) established empirically that coercive (e.g., ability to punish) as well as noncoercive (e.g., ability to provide high quality assistance) sources are important drivers of a franchisor’s power over a franchisee, though franchisees are more satisfied when the latter is used. Also, administrative coordination produces superior operational efficiency (Etgar, 1976) suggesting that power has a positive effect on channel performance (Gaski, 1984). However, powerful channel members can gain more favorable terms than can less powerful bargainers (Walker, 1972). We expect that more power over a franchisor will be enjoyed by a multi-unit franchisee than by a single-unit franchisee. Therefore, a multi-unit franchisee would be able to extract significantly higher concessions from a franchisor than would a single-unit franchisee. This suggests that some franchisors may be willing to risk some channel satisfaction by using single-unit franchisee in order to avoid sharing power with multi-unit franchisees, who might demand higher quality of service or sharing of information, both of which may be very costly in an international context. In addition, international franchisors using single-unit franchisees may safeguard against a high level of franchisee dissatisfaction by exercising coercive power only rarely but keeping it in reserve to ensure channel coordination at a lower cost. As Gaski (1984: 22) notes: “For instance, the imposition of hard sanctions upon channel members (exercised coercive sources of power) seems certain to cause dissatisfaction and conflict, while the dormant presence of the potential to invoke such sanctions (unexercised corrective sources) could conceivably be regarded by franchisees or dealers as benevolent restraint. Likewise, the granting of beneficial assistance (exercised noncoercive sources) should be favorably received, but withholding of such benefits (unexercised noncoercive sources) may not be.” Thus, when issues of power distribution are considered to be more salient than agency concerns, a franchisor may decide to resort to ISUF. Norton, 1988a) suggested that the growth and the size of firms are constrained by limitations to monitoring. Given that monitoring is a key agency problem, our theory can be tested by comparing growth rates and sizes of franchising systems that predominantly use IMUF with growth rates and sizes of those franchising systems that predominantly use ISUF. The growing prominence of franchising in the world economy has led to two major changes in the organization of firms. The first is a shift away from vertical integration to vertical control. The second is the increasing use of hybrid organizational forms to overcome traditional limits to growth. Each of these interrelated trends deserve our attention. Vertical integration was traditionally one of the growth modes preferred by firms (Chandler, 1962). In recent years, however, firms have mostly moved away from vertical integration, mainly due to the increased bureaucratic costs associated with the management of a vertically integrated firm. Instead, the trend has been towards “quasi-vertical integration”. These are close and long-term contractual relationships that provide a firm with most of the benefits of vertical control without the managerial problems or investment needs associated with vertical integration (Carney & Gedajlovic, 1991). Franchising systems, both multi-unit and single-unit, are examples of quasi-vertical integration. Multi-unit franchising seems to be a better means of attaining vertical control in large international franchising systems than ISUF because of the reduction in agency problems. The increasing preference for hybrid organizational forms such as joint ventures, alliances, and networks is another clear trend. Hybrid organizational forms, as discussed earlier, seem to be able to better replicate market mechanisms, while continuing to retain many of the positive aspects of authority relationships. IMUF systems are hybrid organizations where a firm achieves both global geographical scope and significant forward vertical integration using an organizational form that derives both the benefits of markets (through sale of inputs and services by franchisors to franchisees) and hierarchies (through the authority that the franchisor wields over franchisees). As discussed throughout this paper, IMUF seems to be better able to attain these advantages compared to ISUF due to its better ability to address problems inherent in agency relationships.