Foreign Market Servicing Strategies of UK Franchisors: An Emperical Enquiry from a Transactions Cost Perspective(1).
* This paper investigates the firm's choice between franchising-out directly and appointing an intermediary to develop host markets. This decision is modelled using a transaction costs framework.
* The costs associated with monitoring, searching for and servicing franchise buyers are incorporated, as are the costs of protecting intellectual property. The model is tested using data on the scale and scope of UK outward franchising activity.
* Direct franchising is used by larger franchisors to establish smaller franchise systems while more experienced franchisors with larger domestic operations employ intermediaries.
One of the most enduring trends to have emerged in international business over the past two decades has been the growth in the internationalisation of services. This is evident in the range of services now sold across borders, in the broad spatial distribution of this activity and in the volume of service-based transactions taking place. In conjunction, the rate at which firms have adopted one particular business vehicle -- the international franchise -- to effect the downstream delivery of services into foreign markets has also accelerated dramatically over this period (Hoffman/Preble 1993, Falbe/Dandridge 1992, Welch 1992). Today, many franchising firms -- notably those from the US -- own brand names and marks recognisable well beyond the confines of their domestic market. Nevertheless, this `new-form' of market entry remains a relatively neglected topic among researchers of international business (Kedia et al. 1994).(2) There has been research into the motives for franchising internationally and on operational concerns (for example, the degree of product/service standardisation undertaken, marketing issues addressed, and the legal and practical impediments to international franchising), and other work has compared the characteristics of domestic versus internationally-oriented franchising firms (see Julian/Castrogiovanni 1995, Kedia et al. 1994, Falbe/ Dandridge 1992, Huszagh et al. 1992, McCosker 1992, Preble 1992, Eroglu 1991, Hoffman/Preble 1991, Aydin/Kacker 1990, Walker/Cross 1989, Welch 1989, 1992, 1993, Hackett 1976). However, much of this research adopts a descriptive approach. Little effort has been made to craft a rigorous analysis of cross-border franchising as a method of servicing foreign markets. This paper is a first attempt at formally modelling the cross-border franchising process. In drawing upon the results of a survey of UK franchising firms, the paper also provides a first insight into the internationalisation of Europe's foremost franchising nation. As such, the paper represents a useful counter-point to the US and Australian focus of much of the existing research on the topic.
As accounts by legal experts reveal, firms that have developed and own a business format franchise use a variety of organisational forms to transfer this format to foreign markets (Mendelsohn 1992, Konigsburg 1991, Abell 1990). In addition to selling franchises directly to host country individuals and businesses, franchising firms also confer on an independent intermediary the contractual responsibility to develop the local market by selling the franchise on or by operating their own outlets. Less commonly, owners of a franchise also take equity positions in foreign operations, in the form of company-owned outlets -- run by employee managers -- and as affiliated subsidiaries and joint ventures. These affiliated subsidiaries and joint ventures, as well as selling franchises to independent local entrepreneurs, can also be used to operate networks of internal, company-owned outlets in particular markets (Burton/Cross 1995). The generic description of international franchising therefore embraces a diverse sub-set of organisational forms, which much empirical research on the subject acknowledges only in part (see Preble 1992, Hoffman/Preble 1991, Hackett 1976). Contributions, most notably from the legal profession (Mendelsohn 1992, Konigsburg 1991, Abell 1990) do describe in very general terms the various merits and demerits of each. However, little work to date has been conducted to investigate quantitatively the circumstances under which a franchising entrant adopts a particular method; a significant omission in our theoretical appreciation of international franchising which this paper seeks to address.
To advance our understanding, we introduce a model to investigate the factors that determine the propensity for a firm to adopt a given strategy when it sells -- or franchises-out -- its business format to entities (companies and individuals) in non-domestic markets. Foreign entries involving equity participation in the host market are thus beyond the remit of this study. Central to the model presented is a transaction cost framework. In the past, transaction cost analysis has been used to examine the efficiencies of different modes of entry ranging from hierarchies to market alternatives (Contractor 1990, Hill et al. 1990, Hennart 1989, Anderson/Gatignon 1986, Buckley/Casson 1976). The current study is distinctive in that it uses transaction cost analysis to tease out market entry issues within a single broad category of entry mode. In particular, we posit that a divergent set of transaction costs exist between franchising directly into a host market and using an intermediary to develop that market. Our model incorporates variables that theoretical research suggests will impact disproportionately on these distinct costs (or on the firm's ability to bear them), so rendering one method uneconomic relative to the other for a given circumstance. These variables include firm-specific factors (size, experience, scope of international operations and the relative importance of international versus domestic business) and location-specific factors (market size and geographic and psychic distance from the home country). The model is tested using data collected in a survey of the scale, scope and character of international activity among UK franchisors.
The paper is organised as follows. First, we detail the various organisational forms of cross-border franchising available to a franchising firm. Second, we present a transaction cost framework for analysing the determinant factors that favour the selection of one of these forms. Third, we review the methodological approach to the survey upon which this study draws and highlight the main findings. Fourth, we put forward a model used to test the theory and describe the exogenous variables employed. We then discuss the empirical analysis, present the results and describe the implications of these findings. We conclude by recommending avenues for further work.
A Taxonomy of Cross-Border Franchising
Because the term franchising is used to describe a diversity of business arrangements, under which a broad range of products and services can now be sold, as a concept it defies definition. It is nevertheless usual to distinguish between first and second generation franchising. In first generation franchising, the franchise purchaser is granted access to a relatively limited set of rights and resources from the franchisor -- often merely the right to produce a particular product or provide a service under a particular trade name (see Felstead 1993, Sanghavi 1990). Instances are found in the forecourt distribution of automobiles and petroleum products and also in the bottling and marketing of soft drinks. In an international setting, first generation franchising can be regarded, to all intents, as a variant of the `classic' license entry mode (Burton/Cross 1997). On the other hand, second generation franchising, which, as a means of serving foreign markets, is growing at the faster rate (Hoffman/Preble 1993, Falbe/Dandridge 1992, Welch 1992) involves the purchase of a far more comprehensive `package' of rights and resources from the franchisor. This `business format' usually contains most if not all of the elements needed by the buyer to replicate and operate the business successfully for the term of the contract. Depending upon the nature of the business format and contractual obligations, buyers may receive access to the franchisor's intellectual property (encapsulated in, for instance, trademarks, trade names and copyright), business know-how, managerial assistance and support (in, for example, finance, lease negotiation, site selection and outlet design). Buyers of the franchise may also receive from the franchisor training and instruction on the operation of the business and, often, be granted some degree of geographic exclusivity. They may also benefit from centrally co-ordinated data processing, advertising and purchasing by the franchisor. In return for the format, the franchisor can receive an initial up-front fee and/or continuing fees (calculated, commonly, as a percentage of sales or turnover). As the business format is transferred in a highly regulated and controlled manner (to protect the goodwill contained therein) and given the high degree of additional support received by recipients, business format franchising typically involves a much closer, if not intimate, relationship between contracted parties than does first generation franchising. This study investigates factors impacting upon the effective management of this relationship (although much of our analysis will have relevance for arrangements regarded as first generation franchising).
Depending on the nature of the business being franchised, the downstream facilities developed by the franchisor take the form of shops, restaurants, offices, hotels, service outlets and so on. Here, we collectively refer to these host country facilities as units. It has long been recognised -- especially in the domestic setting -- that, in addition to franchising-out businesses to independent operators, franchising companies also operate a significant share of company owned units run by employee-managers. The decision whether or not to internalise the downstream activities of franchising firms has been well examined in the domestic context (see, Norton 1988, Brickley/Dark 1987, among others), and, more recently, has begun to receive attention in the international context (Fladmoe-Lindquist/ Jacque 1995). However, it is important to appreciate that, as already inferred, franchising firms use various organisational forms to establish foreign networks of units (Burton/Cross 1997, Mendelsohn 1992, Konigsburg 1991, Abell 1990), an observation that prior empirical investigation of this entry mode often disregards.
We now look at the various organisational forms by which business formats are transferred to host markets. The simplest of these is direct international franchising (see Figure 1).
[Figure 1 ILLUSTRATION OMITTED]
In direct franchising, the franchising entrant sells its franchise on an individual basis to buyers -- or franchisees -- in the host country. These franchisees are serviced directly from the franchisor's home country. (Dotted arrows in Figure 1 represent the `arm's length' transactions involved). A presupposition is apparent in some contributions to the literature that an international franchising entry must, primafacie, occur in this manner (see, for example, Moyne 1997, Fladmoe-Lindquist/Jacque 1995, Julian/Castrogiovanni 1995). However, it is important to recognise that it is common for franchisors to also transact with some form of intermediary, who purchases from the franchising entrant the right to develop their own network of units in the host market (Konigsburg 1991). These intermediaries (represented by an elongated rectangle in Figure 1) can be local or, but more rarely, third country individuals and companies. There are two main sub-variants to this method of cross-border franchising. The first is master international franchising. Here, the intermediary -- the sub-franchisor -- is permitted by the franchisor to sub-franchise and sell the format on to independent sub-franchisees, in addition to owning and operating its own units. In this instance, the sub-franchisor effectively adopts the role of franchisor in the host market. In the second sub-variant, namely area development franchising, the intermediary -- the area developer -- is permitted only to establish and operate its own units (run by employee managers of the area developer). This franchise agreement does not permit the area developer to sell the franchise on to independent sub-franchisees.
`Arm's length' transactions between the franchisor and independent franchisees, sub-franchisors and area developers give rise to the characterisation of cross border franchising commonly found in the international business literature: that it is a low cost, low risk entry strategy that allows the smaller internationalising firm to develop new and distant markets more quickly and on a larger scale than might otherwise be possible (Preble 1992, Young et al. 1989, Brooke 1986, but see Burton/Cross 1997). Nevertheless, the following question arises. Given that both direct franchising and franchising with an intermediary are both market-based, `arm's length' transactions, and, consequently, would seem to yield similar strategic benefits to the international firm, why do the various strategy subsets of cross-border franchising prevail? Rephrased; why, and under what circumstances, do franchising firms prefer direct international franchising to franchising with an intermediary, and vice versa? Therein lies the central question of the current investigation.
The literature of managerial economics asserts that franchising is a hybrid organisational form, located somewhere between the extremes of vertical integration on the one hand and completely independent operations on the other (Carney/ Gedajlovic 1991, Thompson 1988). Although the parties to the agreement are independent (Rubin 1978, Caves/Murphy 1976), transactions which take place between them can be likened to those of intra-firm vertical integration (Norton 1988), because features of the contract encourage the franchise buyer, through their self-enforced effort, to act in the interests of the franchisor. For the franchisor, this agent-principal relation, and in particular the nature of the contract between the parties, will impact appreciably on the success or failure of foreign market entry using a particular organisational form. In the following section we develop a transaction cost framework to infer the relative costs that the franchising entrant :incurs when transacting with external local entities. Whilst the associated agency costs are likely to be transaction-specific and dependent upon the precise nature of the relationship between contracted parties, these costs are implicit to the following discussion.
The argument is made (for example, by Guisinger/Brewer 1998) that transaction cost economics is a `mature' analytic tool in international business; even a "most travelled boulevard" (Caves 1998, p. 6). Transaction cost economics is a deterministic approach in which efficiency and cost are central to the resource allocation decision. Consequently, in addition to lacking a temporal or experiential dimension, in transaction cost economics factors such as strategic issues (Hill et al. 1990), the core competencies of the firm (especially management capabilities) and the behaviourist view of the international firm and their role in entry mode choice are generally subsumed in an analysis of economic efficiency and its achievement (Bjorkman/Forsgrens 1997). Given these limitations, alternative approaches to investigating the choice of organisational form of franchising entries may prove of value. One possibility is the networks approach (for example, McGaughey/Welch/Welch 1997), which recognises that the experience and personal characteristics of owners and key personnel in small firms can have significant impact on decision-taking and performance in small firms.(3) Nevertheless, despite its critics, transaction cost analysis does yield valuable insight into the 'firm versus market' problem and the cost of `arm's length' spot transactions relative to internalization within business organizations. As such, it provides a useful framework by which to investigate the relative costs of selling franchises directly to host markets or designating an intermediary to accomplish this, and hence is the approach adopted in this study.(4)
Transaction Cost Factors
Five interdependent sets of transaction costs associated with franchising-out into host markets can be envisaged (Welch 1993, Williamson 1975).
(1) Monitoring costs: Franchising firms invest considerable resources to develop, through centralised advertising, marketing and other means, a trademark and brand that is readily recognisable both in the home and, in the context of this paper, in international markets. These marks signal to consumers a certain quality of product or service, irrespective of location. They therefore represent an important if not vital asset for the franchisor, and the franchisor has every incentive to protect the goodwill of the franchise format and its intellectual property. A number of mechanisms are available to the franchisor to ensure the compliance of contracted parties and so militate against dilution of this goodwill. These mechanisms can be grouped into hostage effects and remuneration effects. Buyers of the franchise are held hostage to the agreement (i) by unilateral termination clauses in the contract (Rubin 1978); (ii) by the `up-front' fee they pay, (iii) by format-specific assets which they acquire (Williamson 1983), and (iv) by their dependency on the franchisor for, inter alia, training, technical improvements, sub-supplies and, if permitted in law, tied purchases. Remuneration effects arise because franchise buyers are compensated through the appropriation of a share of the profits they generate (Matthewson/Winter 1985). Despite these hostage and remuneration effects, independent buyers of the franchise nevertheless have the incentive to behave opportunistically and `free ride' (Anderson/Gatignon 1986). For example, declaring incorrect sales and submitting inappropriate fees may increase retained surpluses. More significantly for the franchisor, independent buyers can also economise on overheads to reduce costs and increase residual claims (Fladmoe-Linquist/Jacque 1995). Should a local franchisee or, perhaps of greater concern, an intermediary attempt to achieve this by providing a sub-standard product or service, the reputation of the franchise may diminish. Such an outcome may have important ramifications for the franchisor, in terms of future revenue streams, the rate of sale of franchises and the value of the initial fee commanded, not only in a particular territory where opportunism has occurred, but possibly further afield as well (Norton 1988). Franchising firms therefore incur ex-post monitoring costs as they constrain opportunism and defend the goodwill of the format by policing and supervising independent operators (either `in-house' or by using third parties) through, for example, quality and other control mechanisms, inspection visits, the auditing of procedures and accounts, and so on.
(2) Search costs arise as the franchisor identifies, solicits and evaluates potential franchise buyers in the target market, and as contracts are negotiated and agreed with acceptable candidates. While unsolicited foreign enquiries (Welch 1989) and franchise opportunity brokers may help to lower the ex ante cost of finding suitable foreign parties to operate the franchise, search costs will nevertheless represent a significant sunk cost of entry for many franchisors. On the other hand, the commitment of greater resources during the initial stages of the business relation should improve the quality of the independent operator selected. This, in turn, may lead to lower servicing costs (see below), a lower likelihood that the chosen independent operator will under-perform, and, it follows, a lower risk that the goodwill of the franchise will be jeopardised.
(3) Servicing costs are borne as the franchising firm transfers its technology and expertise to independent buyers of the franchise. The franchisor has to codify the franchise format appropriately (in the form of manuals, written procedures and instructions) so that the recipient can assimilate and replicate the business. The format may also require modification to suit local market conditions. These represent one-off fixed costs for each foreign market entered (an element of which may be apportioned across subsequent entries into psychically similar host markets). The franchisor may also be contractually obliged to provide managerial assistance, technical support, initial training and continuing instruction, both onsite (in the host country) and in the classroom (in the home country) to each buyer of the franchise. Together, these constitute a variable cost.
(4) Property rights' protection costs: It is common for franchise agreements to include clauses forbidding contracted parties from operating a similar business in a given territory and/or time period once the agreement terminates. However, the threat remains that, once fully conversant with the format, these contracted parties (or indeed others) may establish operations in direct competition with the franchisor, before or after the contract officially ends. This threat probably increases with increasing standardisation of the franchise format (Preble 1992, Falbe/Dandridge 1992) and probably decreases, with contracted parties at least, as their dependency on the franchise agreement increases. Nevertheless, in each market entered, franchisors can incur a one-off sunk cost as they take steps to legally safeguard the intellectual property embodied within their franchise format (by registering this with the appropriate authorities), as well as incurring continuing costs to ensure that property rights are not violated ex-post by any local enterprise.
(5) Intermediary-related costs: As we show in the following section, in certain circumstances transacting with an intermediary lessens many of the monitoring, searching and servicing costs associated with franchising directly into host markets, which helps to explain the popularity of this mode of franchising. However, this organisational form introduces a number of additional costs not experienced by the franchisor when franchising directly. These include the resource costs and agency costs associated with maintaining an additional organisational tier. Franchising with an intermediary is very much a `hands off' strategy. Control of the foreign sub-system is relinquished to the local intermediary to a much greater degree than in direct franchising (Konigsburg 1991). There are agency costs associated with managing opposing objectives and regaining control if the local intermediary under-performs or fails altogether. Potential areas of disagreement between franchisor and intermediary commonly involve, for example, turnover targets, corporate image, the rate and manner by which the sub-system should be developed, and issues relating to fee structures, profit redistribution and so on (Mendelsohn 1992, Falbe/Dandridge 1992). Next, if the arrangement breaks down and/or the contract is terminated, the franchisor may incur the management and financial cost of imposing a new organisational structure in the host market. For example, the franchisor may need to reassure and assert control over local unit operators who had previously dealt solely with the intermediary and with whom the franchisor has had little or no prior contact. Indeed, these operators may prove reluctant to deal directly with the foreign franchisor or a new local intermediary (Abell 1990). In some instances, the franchisor may need to purchase the businesses of external operatives, perhaps at inflated prices if the contract penalises buy-back (Konigsburg 1991). Finally, in addition to these agency costs, the introduction of an additional organisational tier can also impact negatively on the compensation received by the franchisor for its franchise format. This is because, in direct international franchising, the franchisor retains all initial and continuing fees submitted by franchisees. An intermediary, on the other hand, appropriates a substantial share of the income generated by units under its control (Konigsburg 1991). For all these reasons, using an intermediary to develop a foreign market introduces an additional set of costs that are slight or even absent when franchising directly into a market.
This review of transaction costs provides a starting-point for analysing the franchising entrant's choice of appropriate organisational form for market entry. In particular, following Williamson (1975), the franchisor will select the organisational form that most economises on transaction costs. As we have shown, a nexus of transaction costs exists in cross-border franchising, the configuration of which is determined by whether or not an intermediary is present (see Figure 1). Transaction costs [TC.sub.DIR] are borne by the franchisor when it sells its franchise directly to an independent foreign franchisee. The total cost of establishing a network of host country units by direct means is therefore [Sigma][TC.sub.DIR]. Meanwhile, transaction costs [TC.sub.INT] are borne by the franchisor when it transacts with an unrelated foreign intermediary (a master sub-franchisor or an area developer). However, this intermediary, in turn, incurs transaction costs [Sigma][TC.sub.SF] as it services its own local network of independent sub-franchisees and units. These costs [Sigma][TC.sub.SF] are incorporated into our model under the assumption that they are passed on to the franchising entrant as a lower return on entry.(5) Transaction cost analysis predicts that under certain circumstances, but especially in instances of appreciable geographic distance and psychic distance (arising from differing language, culture, political and legal systems) between home and host countries, the intermediary's costs of developing a local network of outlets, [Sigma][TC.sub.SF] will be significantly less than those experienced by the franchisor if it had developed the operation directly, [Sigma][TC.sub.DIR]. In such instances, the franchisor's transaction costs [TC.sub.INT] of servicing the market via an intermediary plus the intermediary's transaction costs of servicing its own network of units, [Sigma][TC.sub.SF], may prove lower than if the franchisor transacted with those host market units directly, [Sigma][TC.sub.DIR]. Therefore, franchising with an intermediary is preferred when:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where n is the number of units in the host country operating the franchise format. Conversely, direct franchising will the observed entry mode when transaction costs [TC.sub.INT] + [Sigma][TC.sub.SF] exceed [Sigma][TC.sub.DIR].
Implications for Choice of Organisational Form of Franchising
Using this transactions costs framework, we develop in this section eleven testable hypotheses that predict the impact of exogenous variables on the relative transaction costs, [TC.sub.DIR], [TC.sub.INT] and [TC.sub.SF] of franchising into a foreign market either directly or via an intermediary. A summary of the hypothesised effect of an increasing value for the exogenous variables on the choice of organisation form is provided in Table 1, where the effect is positive (+), negative (-) or indeterminate ([circle]).
Table 1. Hypothesised Influence of Exogenous Factors on Organisational Form
Preferred Organisational Form of Cross Border Franchising as Exogenous Exogenous Variable Variable Increases Direct Intermediary size of franchisor (in terms of + - managerial resources) franchising experience of the franchisor + - (both of domestic and international franchising) size of the host market franchise system - + (in terms of number of units present) total number of foreign market entries - + made by franchisor cultural distance between home and host - + country geographic distance between home and - + host country size of the franchisor's domestic + - franchise system (number of units present) proportion of the franchisor's total + - annual sales generated by its franchising activities proportion of the franchisor's total - + franchising-related revenue earned by its domestic operations affluence of the host market + - firms from walk-in retail or motorist [circle] [circle] services industries
As already indicated, no empirical work exists on the factors influencing the franchisor's choice of organisational form in the international context. However, `anecdotal' accounts by lawyers familiar with formulating cross-border franchising agreements provide some evidence. In particular, they observe the tendency for direct franchising to occur between firms from countries that are geographically close and psychically proximate and also in instances when relatively small host-country franchise systems are developed (Mendelsohn 1992, Konigsburg 1991, Abell 1990).
It is well known that both geographic distance and psychic distance can influence the selection of foreign market servicing strategy (Hill et al. 1990, Contractor 1990, Anderson/Gatignon 1986, Hofstede 1983). With particular reference to franchising, we may envisage geographic proximity to attenuate the communication and transportation costs (particularly in respect to the international travel of home-country managers) of searching for, contracting with, monitoring and servicing local buyers of the franchise (Konigsburg 1991). Similarly, transaction costs may abate when markets are entered that resemble culturally (in terms of language, tastes, laws and commercial practice) the franchisor's home market (Kogut/Singh 1988). In such a circumstance, servicing costs may diminish as the franchise package will probably require little, if any, modification to be viable locally (Sanghavi 1990). Similarly, cultural proximity is likely to promote a convergence of organisational culture between franchisor and buyers of the franchise. Such convergence should facilitate the quick and efficient assimilation of transferred technologies and operational requirements on the part of the recipient, together with the adoption of any necessary control and monitoring mechanisms (Konigsburg 1991). Cultural proximity may also foster the close working relationship between contracted parties often cited as fundamental to the success of franchising (Falbe/Dandridge 1992, Hoffman/Preble 1991). When host and home markets are geographically and culturally proximate, therefore, the franchisor's cost of transacting with franchisees, [Sigma][TC.sub.DIR], may arguably be of a similar order of magnitude to that incurred by an intermediary as it transacts with units and sub-franchisees, [Sigma][TC.sub.SF]. In such circumstances, however, the introduction of an additional set of transaction costs, [TC.sub.INT] (and especially intermediary-related costs) may render uneconomic franchising with an intermediary. Transaction cost analysis thus predicts that direct international franchising is preferred when entering markets that are geographically and culturally close to the home country, a trend which has been reported, for example, between firms from the US and Canada, the UK and Eire, and Australia and New Zealand (Mendelsohn 1992, Abell 1990).
As spatial and psychic distance increase, however, so too will transaction costs [TC.sub.DIR] and [TC.sub.INT]. Search costs may increase because franchisors need to expend greater resources to identify and contract with acceptable candidates for franchisees and intermediaries. In addition, franchise contracts will almost certainly need to be redrawn to make them enforceable in markets with differing legislative regimes to those previously encountered. Servicing costs may increase if elements of the franchise package need tailoring to accommodate local market conditions (Eroglu 1992, Preble 1992, Hackett 1976). Finally, monitoring costs are likely to increase if differences in business ethics and practices between the franchising entrant and locally contracted parties become more pronounced, rendering it less easy (or more costly) to ensure the satisfactory performance of the latter. For the franchisor, these increasing costs are likely to be most burdensome when franchising directly, particularly when developing larger host sub-systems, because this method requires multiple contracts to be negotiated and policed on a one-to-one basis with individual foreign franchisees. Franchising with an intermediary, on the other hand, involves the ratification and policing of just a single contract. As this intermediary effectively adopts the role of franchisor in the host market, it can use its own local knowledge and expertise to adapt and modify the transferred business package to suit local conditions. This role may lower the cultural and legal barriers to entry (Falbe/Dandridge 1992). In addition, the intermediary's own monitoring and servicing costs abate as it is spatially and culturally close to its local operations. In such circumstances, [TC.sub.INT] + [Sigma][TC.sub.SF] may prove to be lower than [Sigma][TC.sub.DIR], and for a firm to use direct franchising to establish numerous host country units in geographically and psychically distant markets may become uneconomic (Sanghavi 1990). As Abell (1990, p. 21) states, direct franchising "is not ... usually suited to distant markets with a different language and business culture". In such circumstances, franchising with an intermediary may prove superior.
Features of the franchisor's international franchise system may also influence the choice of organisational form. Firstly, as already indicated, transaction costs [TC.sub.DIR] and [TC.sub.SF], particularly monitoring and servicing costs, should increase in proportion to the number of host country units established. Transaction costs [[Sigma]TC.sub.DIR] therefore effectively constrain the number of units in the foreign operation that can be adequately serviced directly by the franchisor. Consequently, direct franchising should tend to be observed when firms develop smaller nondomestic systems (Konigsburg 1991). On the other hand, the proximity of an intermediary to its host market operation should moderate transaction costs [Sigma][TC.sub.SF] relative to [Sigma][TC.sub.DIR] and -- providing this intermediary has the requisite resources to hand -- should favour the establishment of relatively larger host country franchise systems. Secondly, significant diseconomies of scale, particularly in the co-ordination and management of increasingly disparate groups of franchisees, accrue to the franchisor as the number of host countries it has entered rises. This, we suggest, might act disproportionately upon transaction cost [Sigma][TC.sub.DIR], relative to the cost [TC.sub.INT], + [Sigma][TC.sub.SF], of appointing intermediaries to effect market development. Conversely, it may be efficacious to employ intermediaries across several different host markets as this enables the franchisor to `benchmark' relative performances. We surmise, therefore, that a correlation will exist between a preference for direct franchising and fewer and less extensive operations in host markets, and vice versa for entries using intermediaries.
International business theory suggests that a number of firm-specific factors influence the firm's choice of foreign market servicing strategy (Contractor 1990, Anderson/Gatignon 1986). However, as no empirical work exists on the relationship between characteristics of the franchising firm and its predilection for a particular method of cross-border franchising, we now take a first look at this relationship.
Market entry theory asserts that `arm's length' transactions are preferred by smaller and/or inexperienced firms (Contractor 1990, Young et al. 1989) that lack the requisite managerial or financial capacity to internalise foreign operations. As we have seen, both franchising directly into a host market and using an intermediary are `arm's length' arrangements, insofar as they involve transactions with non-affiliated concerns. Nevertheless, the size and experience of the firm may still impact upon the choice of franchising form. For example, larger franchising firms should have the requisite managerial and/or financial resources to bear the transaction costs [TC.sub.DIR] of direct franchising, and so avoid intermediary-related costs, compared to their smaller counterparts. The latter, because of their inability to finance the indivisible costs of direct franchising, may prefer to appoint an intermediary to effect entry. Experienced firms should also be better skilled than novice firms at identifying suitable franchise candidates; and be better able to efficiently contract, codify and transfer know-how and procedures and monitor the performance of independent parties to constrain agency costs (Julian/Castrogiovanni 1995, Aydin/Kacker 1990, Davidson 1980). We incorporate the experiential dimension into our model in terms of the length of time that the firm has franchised its business, in both the home and foreign markets, and in terms of the size of its domestic franchise network (with a larger domestic operation indicating, we suggest, a greater appreciation of the franchising process overall). We surmise that experienced firms will prefer to apply their knowledge and skills and franchise directly into host markets rather than appoint an intermediary, with its associated resource and agency costs.
We now consider three exogenous variables that relate to revenue streams associated with the organisational form of cross-border franchising. For some firms, franchising-related activities represent only a small element of total business; for others it is their sole source of income. While the former may find it cost effective to develop markets via an intermediary, the latter may seek to militate against the concomitant potential threat of control loss and instead enter a market by direct means. For similar reasons, firms more heavily reliant on foreign franchise-related revenue compared with revenue generated by their domestic franchise operations may prove reluctant to adopt a `hands off' strategy and prefer instead the greater control afforded by franchising directly. This may also prove to be the case when firms enter relatively affluent markets, since direct franchising permits the franchisor to appropriate a greater percentage of operating surpluses. In other words, the more important international business is -- as a share of total business -- and the more affluent the host market is perceived to be, the greater the opportunity cost of allowing an independent intermediary to develop it. Lastly, we incorporate into the model a variable to capture any industry effects on the choice of franchising entry mode that might otherwise not be modelled by the other variables employed.
This section presents the research methods for the collection of the data used to test the model.
Population and Sample
No official cross-sectional data is collected on the scale, scope and characteristics of the non-domestic operations of franchisors originating from and headquartered in the UK. We therefore test the model using data obtained in a postal survey of outward UK franchising activity (Burton/Cross 1995).
Three trade directories, the UK Franchise Directory (1994), the World-wide Franchise Directory (1991) and the Franchise World Directory 1992 (1993) and other franchising-related press were used to identify the population of UK-based international franchisors, comprised of those firms advertising that they had franchised internationally or that were soliciting foreign enquiries (and so may already have been internationally active at the time of the survey). It is common practice for franchising firms to publicise the extent to which the format they are selling has been internationalized -- effectively enhancing its attractiveness to potential buyers -- when in fact these firms have merely purchased the right to develop a particular host market and have had no involvement in the internationalization of the format elsewhere. Care was therefore taken to ensure that the population included only those UK firms actually responsible for establishing the foreign operations they purported to possess.
The population derived in this way consisted of 240 UK-based franchising firms that could be active or potentially active in non-domestic markets. It is likely, however, that this figure over-estimates considerably the true population of internationally-active UK-based franchisors. In particular, a study by the National Westminster Bank and the British Franchise Association (BFA) (1994) carried out just 18 months prior to the survey reported here stated that, of the 396 business format franchise systems active in the UK in 1993, some 22 per cent, that is 87 franchise formats, were also in operation in non-UK markets. Of these 87 franchises, an unreported yet potentially significant proportion will have been developed by non-UK franchisors. Indeed, the survey reported here identifies eleven such firms. It is therefore likely that the 240 firms sampled exaggerates the true number of UK franchisors possessing non-domestic operations, with many firms possibly soliciting foreign enquiries with the primary objective of improving the perception of their format amongst potential domestic buyers of the franchise.
The survey instrument, a self-administered questionnaire, was mailed to the identified population of firms together with a personalised covering letter to their owners, managing directors or international franchise managers, as appropriate.(6) A brief glossary of terms and pre-paid, addressed envelope were also included. Since franchising firms today provide a diverse range of products and services, and because a varied and indeed confusing taxonomy of terms exist, considerable effort was made to make the questionnaire as unambiguous and as universally applicable as possible. Academics familiar with questionnaire design tested drafts prior to administration. A summary of responses is presented in Table 2.
Table 2. Summary of Responses Response to Mailing Number Total sent 240 Returned undelivered (address gone away) 20 Delivered 220 Returned 74 Removed from population 49 (firm yet to internationalize) (26) (firm not responsible for developing non-UK (11) franchise operation) (firm no longer trading/franchising) (12) Total identified population 171 Unusable replies and refusals to participate 10 Number of usable replies 15 Response rate 8.7%
As this Table shows, if the 69 firms known to be inactive internationally (addressee gone way, firm yet to internationalise, firm not responsible for developing non-UK franchise operation, firm no longer trading or franchising) are removed from the original population, the fifteen usable replies represents a response rate of just under 9 per cent of our sample. However, it is likely, given the comments above, that these are more representative of the true population of internationally active UK franchisors than the response rate at first suggests. In particular, taking the National Westminster Bank/BFA survey estimate of 87 internationally active franchises in the UK as a more accurate assessment of the true population, we can infer that over 17 per cent of these firms responded to our survey, lending support to the generality of our findings.
The respondent firms were involved in a wide range of business activities: property and estate-related services, walk-in retail, motorist services (three firms each); distribution services, business services (two firms each); cleaning services, health and beauty (one firm each). Between them, these franchisors had made 93 foreign market entries into 46 countries. Seven foreign market entries were made using local affiliates (joint ventures). Since we focus here on non-affiliate franchising, these entries were removed from the data set. A further five were removed because of incomplete data. Removing these entries from the data set deleted all the entries made by one firm. The data set, therefore, consists of 81 foreign market entries (the unit of analysis), 466 units and 14 franchising entrants. Table 3 summarises the distribution of the foreign market entries made by these firms.
Table 3. International Operations of UK Franchisors in Data Set Total Number of Number of Units Operating Host Market Respondents with Respondents' Entered by Operations in Format in Respondent Host Market Host Market European Union Austria 1 1 Belgium 3 10 Denmark 2 8 Eire 5 46 Finland 1 1 France 5 18 Greece 2 7 Holland 5 30 Luxembourg 2 1 Portugal 3 20 Spain 4 36 Sweden 2 2 Total EU 35 180 Rest of Europe Croatia 1 1 Cyprus 2 8 Hungary 2 23 Malta 1 7 Norway 3 16 Switzerland 2 2 Total R.o.E. 12 57 North America Canada 1 1 USA 1 10 Total N. America 2 11 Africa Ghana 1 2 Kenya 1 1 Libya 1 2 Mauritius 1 1 South Africa 1 85 Total Africa 5 91 Middle East Bahrain 1 1 Jordan 1 1 Kuwait 1 7 Oman 1 1 Saudi Arabia 4 17 Turkey 3 18 UAE 2 21 Total ME 13 66 Far East Australia 4 33 Hong Kong 3 8 Indonesia 1 1 Japan 1 7 Malaysia 2 7 New Zealand 1 1 Singapore 2 4 Total FE 14 61 Grand Total 81 466 All Markets
Table 4 shows how often the 14 respondent firms used each organisational form of international franchising to develop these non-domestic markets. These firms did not confine themselves to exclusively employing a particular organisational form.
Table 4. Organisational Form of Cross-Border Franchising Employed by UK Franchisors in Data Set
Total Number Frequency of Non-domestic Organisational Form Used Units Established Direct franchising 22 73 Franchising with intermediary 59 393 (Master) (34) (282) (Area developer) (25) (111) Total 81 464
Estimating a model by Ordinary Least Squares where the dependent variable is dichotomous or limited induces bias and heteroscedasticity. The solution to this is to employ a symmetric distribution (a standard normal or logistic) and a maximum likelihood estimator, which results in either a probit or a logit model. In practice there is little difference between the probit or logit models and the most convenient is usually employed (Judge et al. 1988). For this study, a multinomial probit model was estimated (using the variables as described below). Direct international franchising was given the coding DIF, while master international franchising and area development franchising were each given a separate coding. This allowed for a choice of three organisational forms by the franchisor. A binomial probit model was then estimated combining master and area development franchising under the same coding, MADF. There were no significant differences between the coefficients obtained using each model, which is not surprising given the similarity in theory of master and area development franchising. The transaction costs [TC.sub.INT] that arise when an intermediary is present are of a similar order of magnitude irrespective of whether a master sub-franchisor or an area-developer is employed. The preferred estimator is therefore a binomial probit.
The dependent variable METH in the empirical model is dichotomous, taking the value 0 if the market is served by DIF and 1 if served by MADF. Only the actual organisational form used in a particular entry is observed from the data set. This is assumed to reflect partially the franchisor's strength of preference for MADF as opposed to DIF. The rate of return to these organisational forms (and hence the firm's preference) is modelled as dependent upon the exogenous variables given in Table 5:
METH = [[Beta].sub.0] + [[Beta].sub.1] SIZF + [[Beta].sub.2] YRCO + [[Beta].sub.3] NRCT + [[Beta].sub.4] NRUN + [[Beta].sub.5] NUMU + [[Beta].sub.6] CULT + [[Beta].sub.7] DIST + [[Beta].sub.8] UKUN + [[Beta].sub.9] PSFR + [[Beta].sub.10] PFOV + [[Beta].sub.11] GDP + [[Beta].sub.12] WINT + [[Mu].sub.i]
The reported coefficients for these explanatory variables best estimate, overall, the probability that the respondent firm would use MADF as opposed to DIF to develop a particular market.
Table 5. Exogenous Variables Employed Notes and Source (from the survey instrument unless Variable Definition otherwise stated) SIZF Size of franchisor Using a 5-point scale (number of employees in from < 10 (1), between the UK headquarters 11-30, 31-60, 61-100 or used as proxy) 100+ (5) UK employees. YRCO Year host market first See text entered (a proxy for international franchising experience) NRCT Number of years between See text year of first UK franchise and entry into first foreign market (a proxy for domestic franchising experience) NRUN Number of units in host country NUMU Scale of international Total number of countries operations entered. CULT Cultural distance between A dichotomous variable home and host country where host country is Anglo (CULT=0) or non-Anglo (CULT=1) (after Ronen and Shenkar 1985). Anglo countries are the US, Canada, Australia, New Zealand, Ireland and South Africa. DIST Geographic distance Statute air miles between between home and host London, Heathrow and country capital city of host country as reported in Whitacker's Almanac (1996, p. 117). UKUN Size of the domestic Current number of units operations operating franchisor's format in UK. PSFR % of total annual revenue Using 5-point scale from generated by the firm's 0-19% (1), 20-39%, 40-59%, franchising business 60-79%, 80-100% (5). PFOV % of global annual As for PSFR. franchising-related revenue generated by UK franchise operation GDP Host market size Gross domestic product (affluence) (GDP) per capita for 1992, in US$ from United Nations (1995). WINT Walk-in retail or motorist A dichotomous variable services taking the value 1 if the firm is from the walk-in retail or motorist services sectors and 0 if not (see text).
These exogenous variables were obtained from the survey instrument. This requested respondents to provide general information about their company, and then more detailed information about each of the foreign markets into which they had sold their franchise. Respondents were asked to name the country entered; indicate the year of the first entry; the franchise method used to effect entry (direct, master, or area development franchising); and the number of units established by either the franchisor or the intermediary. From this information, two measures of firm experience were calculated. The year of market entry provided YRCO, a higher value for which indicates a more recent foreign market entry and therefore less experience of operating in that market. We then calculated NRCT, which is the number of years between the firm's first entry into the foreign market concerned and the year of its first UK franchise. The larger this number, the greater the franchisor's overall experience of operating a franchise prior to a foreign entry. Finally, as it was necessary to capture the effect of industry type which would not otherwise be observable in the other exogenous variables a number of dummy variables were introduced for each industry type represented by the respondents. Of these dummy variables only `walk-in retailing' and `motorist services' had any measurable effect. There was no difference in the effects of these variables in estimation and so they were combined. The final specification of the model therefore included the dichotomous variable WINT to describe whether or not the franchise of the respondent firms was in `walk-in retailing' and `motorist services' business.
The estimated results, using Limdep (version 6.0) are reported in Table 6.
Table 6. Summary of Results Variable Coefficient t-ratio Prob |t| > x Constant -29.82518 -1.028 0.30418 SIZF -7.591281 -2.459 0.01393(**) YRCO 0.2694773 0.842 0.40001 NRCT 0.5629821 1.753 0.07962(*) NRUN 0.5899420 2.351 0.01874(**) NUMU -1.506319 -2.344 0.01905(**) CULT 0.2249635 1.107 0.26848 DIST 0.2311607E-03 1.259 0.20818 UKUN 0.8560355E-01 2.559 0.01050(**) PSFR -0.4004493 -0.398 0.69088 PFOV 7.000541 2.526 0.01155(**) GDP 0.1088126E-03 1.619 0.10542(#) WINT 10.95086 1.991 0.04645(**)
(#) Significant at p < 0.15
(*) Significant at p < 0.10
(**) Significant at p < 0.05
Overall, the fit of the model is very good. A [chi square] test of the joint significance of the variables produced a value of 71 against a critical value of 27 at a 99% level of significance. Given a dichotomous dependent variable, goodness of fit is slightly more complex to derive than under Ordinary Least Squares. The matrix of actual and predicted outcomes (see Table 7 below) shows that the model correctly predicts the observed outcome in over 90% of cases. There seems little skew in the predictive errors. This supports our claim of an appropriate specification.
Table 7. Matrix of Actual versus Predicted Outcomes Predicted Actual 0 1 Total 0 20 2 22 1 3 56 59 Total 23 58 81
The results of probit analysis do not give marginal effects. These effects are conditional upon the level of all exogenous variables. We therefore adopted the standard approach of deriving these marginal effects at the mean of the exogenous variables. Not one of the coefficients switched sign. A correlation matrix was produced, which suggested that bivariate collinearity is not a problem in the model. For example, SIZF and UKUN, two variables most likely to demonstrate collinearity, have a low correlation of 0.46. Much of the variation in the dependent variable is between rather than within firms. It is accepted practice to exclude group effects in panel estimates where between variation is of interest, and this paper uses an analogous justification. The firm's preference for a particular organisational form would be driven by those variables included in the model. The inclusion of dummy variables for each firm does not provide additional explanatory power, while multicollinearity confuses the results. As discussed above, dummies were included to examine whether or not different industries reveal a tendency towards a particular entry mode of franchising. The high degree of explanatory power for the preferred model confirms that any omitted variable bias must be small.
The t-test confirms that the coefficients on the variables SIZF, NRUN, NUMU, UKUN, PFOV and WINT are significantly different from zero at a 95 per cent significance level. The coefficients for the variables SIZF, NRUN and PFOV have the a priori sign. These results indicate that the probability that direct international franchising is the observed entry method increases (and decreases for franchising with an intermediary) as the size of the franchisor increases (in terms of the number of employees in its headquarters) and as the size of the host country operations decreases (in terms of numbers of units established). With respect to PFOV, the probability that direct franchising is the observed entry method increases as the contribution made by foreign (domestic) business to the total franchise revenue increases (decreases). Naturally, for each result the converse holds for franchising with an intermediary (see Figure 2).
[Figure 2 ILLUSTRATION OMITTED]
Two of the variables found to be significant at a 95 per cent level, NUMU and UKUN, had signs counter to that predicted by theory. These findings indicates that, contrary to expectation, the probability that direct international franchising is preferred over using an intermediary increases as the number of foreign markets entered by a franchisor increases and as the size of the franchisor's domestic operations decreases (in terms of the numbers of units present). Also contrary to expectation is the finding for NRCT, which has a positive coefficient and was significant at a 90% level. This finding indicates that firms are more likely to franchise into foreign markets using an intermediary the greater the period of time between establishing its initial home franchise network and its first venture into a non-domestic market (or, in other words, the greater the `hand's on' experience gained of franchising in the domestic market). Again, for each result, the converse holds for franchising with an intermediary.
Theory was unable to ascertain a priori the direction of effect for the variable WINT on the decision between franchising directly or via an intermediary. The positive sign on the coefficient indicates that firms from the `walk-in retail' and `motorist services' sectors are more likely to internationalise using an intermediary than by franchising directly into foreign markets. Firms in other sectors show no preference for either option, confirming the strength of the results for the other variables.
The remaining coefficients are insignificant and have no statistical effect on the choice of organisational form. These are firm experience, measured in terms of the length of time the franchisor has serviced a particular foreign market, YRCO; the proportion of total annual revenue generated by the firm's franchising activities, PSFR; the psychic distance between the UK and host markets, CULT (despite employing a dichotomous variable to capture the greatest variance); the geographical distance of host markets from the UK, DIST; and market affluence in terms of GDP per capita, GDP (although a t-test almost rejected the null for this variable at a 90 per cent significance level).
Discussion and Conclusions
Once the manager of a franchising firm has chosen to sell this format into a new foreign market, a decision must be taken between selling the franchise directly to franchisees or delegating the development of the market to sub-franchisors and area developers. Central to this paper is the notion that this decision is predicated upon the desire of the firm to protect the intellectual property and goodwill associated with its format, while simultaneously economising on the costs associated with searching for, monitoring and servicing the activities of independent purchasers of the franchise. We introduce a transaction cost framework to provide a rigorous and novel theoretical approach to analysing the decision-making process that attempts to reconcile these often mutually exclusive objectives.
The UK firms studied reveal a tendency to adopt direct franchising when theory predicts that the transactions cost associated with searching for, policing and monitoring independent buyers of the franchise are minimised. In particular, the costs of negotiating and servicing numerous local franchisee businesses directly would seem to inhibit the development of larger scale host country networks by direct means. Moreover, firms with greater managerial resources with which to service host country units themselves and those that earn a greater proportion of their franchising-related revenue from their international operations (and who might therefore prove unwilling see control over these operations diluted by an intermediary) do exhibit a preference for franchising directly. On the other hand, smaller franchisors and those that generate a greater share of their franchising-related revenue from their domestic activities demonstrate a propensity to employ intermediaries. For these firms, contracting with local intermediaries would seem to detract less upon their resources and marginal international activities than does direct franchising. Indeed, using intermediaries may even allow firms to streamline the size of their domestic workforce. Intermediaries, because of their spatial and cultural proximity to their local market, are also more likely to be used to oversee larger-scale operations in host countries. These results are the first empirical confirmation of `anecdotal' evidence presented by lawyers regarding the circumstances in which direct, master and area development franchising are observed.
Two of the results obtained run counter to expectation. One might have expected firms that possess a large network of units at home to show a preference for extending into the international arena directly their knowledge of franchising-out that they have gained domestically, so avoiding the resource costs and agency costs associated with delegating local market development to an intermediary. The same might be expected for firms with greater experience of franchising at home. Instead, this study reveals a tendency for such firms to transact with local intermediaries, for which several explanations can be envisaged. It is possible that the expertise and knowledge gained in franchising out to a larger number of domestic franchisees, over a longer period of time, provides the franchisor with greater confidence and ability to select and police competent intermediaries, despite the agency costs introduced by this party. A successful home operation may also attract pro-active host country entrepreneurs with the requisite resources and ambition to take the franchise to the foreign market as a successful sub-franchisor or area developer (Moyne 1997). That direct franchising is preferred by respondent firms that have internationalised relatively soon after first franchising at home may reflect the tendency for this method to be used as a `toe in the water' internationalization strategy by relatively inexperienced firms unable as yet to manage the complexities of the franchisor-intermediary relationship. This may partly explain why franchising with intermediaries is found to occur in conjunction with the development of fewer international markets.
Transaction cost theory predicts that the diseconomies of scale associated in co-ordinating and managing networks of direct franchisees across numerous international markets favours instead the use of intermediaries. However, the management of the franchisor-intermediary relationship and its associated agency costs may require a degree of managerial commitment and finesse on the part of the franchisor that prevents the mode from being employed too frequently across international markets. Industry effects may also be at work, since, in our sample, firms involved in the walk-in retail and motorist services sectors showed a clear preference to enter markets using an intermediary. It is possible that, in these sectors, greater need exists for the format to be tailored to suit the requirements of individual customers while maintaining uniformity in service delivery across various outlets in the host market. Employing a single intermediary, with its particular local knowledge and resources may achieve this more readily than would using a number of individual franchisees.
A number of variables are found to have no demonstrable effect on the choice of organisational form by firms in our sample. No particular preference is shown by experienced or novice franchisors (in terms of the time they have been active internationally) for franchising directly or using an intermediary. The same is true for the proportion of total earnings that the sampled firms generate from their franchising business. These results may be explained in part because, since both are `arm's length' transactions, franchising directly and franchising via an intermediary may yield equivalent costs (and benefits) to the internationalising firm with respect to these firm-specific variables.
Surprisingly, we observe psychic and geographic distance between home and host market and the market potential of the host territory to have no influence on the choice of franchising form among our sampled firms. One would expect cultural distance in particular to impact upon the franchisor's search, servicing and monitoring costs of market servicing and therefore have a determinant role in entry mode selection. However, a process of `self-selection' may be in operation. For example, a local enterprise may only purchase a franchise which it is confident will be viable and profitable in the host market without significant modification and fine-tuning, irrespective of whether that enterprise intends to be a franchisee or intermediary. The franchisor may also only franchise into markets where it knows a demand exists for the products and services provided by the format, and where the transfer to local enterprises can be effected with a minimum of cost. In the case of franchising-out, psychic and geographic distance and market affluence may perhaps have greater influence on whether market entry takes place in the first instance, rather than on the manner by which the entry occurs.
As we have previously indicated, the model we present is exploratory. An interesting addition would be an examination of the dynamic aspect of this organisational choice and the motives for switching between modes subsequent to market entry. Issues regarding the local environment for franchising-out that might influence the selection of appropriate entry strategy could also be incorporated. These might include issues relating to local franchising-related legislation, fiscal regime, financial environment (for example, banks may prove unwilling to finance local franchisees when the franchisor is remote and lacks local representation) and additional agent-related issues (for example, potential franchisees may be reluctant to purchase franchises from non-resident franchisors) (Konigsburg 1991). The model would also benefit from incorporating a measure of absolute sales volume and/or profitability in a given market. This could prove to be a significant variable, as the set-up cost of establishing an intermediary could be spread over a greater volume of sales, rendering that mode more efficient. Access to more detailed data, and where necessary commercially sensitive information, would facilitate this. Finally, the model could also be usefully extended to other home countries, in particular the US, where the majority of international franchising activity has originated and where a larger sample size might be expected. This study represents the first formal and rigorous analysis of the organisational form for franchising in an international context and provides a useful foundation upon which our understanding of the franchising entry mode can be further advanced.
(1) The authors gratefully acknowledge the financial support of ESRC grant R-000-22-1121, Colin McCosker for making available to us a survey instrument and Peter Buckley, Jeremy Clegg and two anonymous referees for their comments on earlier drafts.
(2) This neglect is explained in part by the paucity of official data on the scale and scope of cross-border franchising and the organisational forms used to effect it, and in part by the misapprehension that international franchising is merely a sub-variant of non-affiliate licensing (for example, Boddewyn et al. 1986, but see Burton/Cross 1997).
(3) The authors are grateful for the comments of an anonymous referee on this point.
(4) By way of support, transaction cost analysis also continues to be applied to corporate governance problems in the context of purely domestic franchising (for example, Brown 1998).
(5) The intermediary's own transaction cost exist in the system and need to be endogenous to our model. However, we recognise that the degree to which these costs are passed on to the franchisor is a subject for further investigation. Note, in area development franchising the area developer acquires the franchise in the knowledge that it cannot sell this on. Here, transaction costs [TC.sub.SF] between the intermediary and its units thus become internalised.
(6) The survey instruments and detailed summary of responses are available from the authors on request.
Abell, M., The International Franchise Option, London: Waterlow 1990.
Anderson, E./Gatignon, H., Modes of Foreign Entry: A Transaction Cost Analysis and Propositions, Journal of International Business Studies, 17, 3, 1986, pp. 1-26.
Aydin, N./Kacker, M., International Outlook of US-based Franchisors, International Marketing Review, 7, 2, 1990, pp. 43-53.
Boddewyn, J. J./Halbrich, M. B./Perry, A. C., Service Multinationals: Conceptualization, Measurement and Theory, Journal of International Business Studies, 17, 3, 1986, pp. 41-57.
Bjorkman, I./Forsgren, M., Nordic Contributions to International Business Research, in Bjorkman, I./Forsgren, M., (eds) The Nature of the International Firm, Copenhagen: Copenhagen Business School Press 1997.
Brickley, J. A./Dark, F. H., The Choice of Organisational Form: The Case of Franchising, Journal of Financial Economics, 18, 1987, pp. 401-420.
Brooke, M. Z., International Management: A Review of Strategies and Operations, London: Hutchinson 1986.
Brown, W. O. Jr., Transaction costs, corporate hierarchies, and the theory of franchising, Journal of Economic Behavior and Organization, 36, 1998, pp. 319-329.
Buckley, P. J./Casson, M. C., The Future of the Multinational Enterprise, London: Macmillan 1976.
Burton, F. N./Cross, A. R., International Franchising: Market versus Hierarchy, Chapter 7 in Chryssochoidis, G./Millar, C./Clegg, L. J. (eds), Internationalisation Strategies, Basingstoke: Macmillan 1997.
Burton, F. N./Cross, A. R., Foreign Market entry and Development Strategies of Franchising Firms: Some Evidence from the UK, in Schiattarella, R. (ed.), New Challenges for European and International Business, Proceedings of the European International Business Academy, Rome Conference 1995, pp. 89-114.
Carney, M./Gedajlovic, E., Vertical Integration and Franchise Systems: Agency Theory and Resource Explanations, Strategic Management Journal, 12, 1991, pp. 607-629.
Caves, R. E. Research on International Business: Problems and Prospects, Journal of International Business Studies, 29, 1, 1998, pp. 5-19.
Caves, R. E/Murphy, W. F., Franchising: Firms, Markets and Intangible Assets, Southern Economic Journal, 42, 1976, pp. 572-586.
Contractor, F. J., Contractual and Cooperative Forms of International Business: Towards a Unified Theory of Modal Choice, Management International Review, 30, 1, 1990, pp. 31-54.
Davidson, W., The Location of Foreign Direct Investment Activity: Country Characteristics and Experience Effects, Journal of International Business Studies, 11, 2, 1980, pp. 9-22.
Eroglu, S., The Internationalization Process of Franchise Systems: A Conceptual Model, International Marketing Review, 9, 5, 1992, pp. 19-30.
Falbe, C. M./Dandridge, T. C., Franchising as Strategic Partnership: Issues of Co-operation and Conflict in a Global Market, International Small Business Journal, 10, 3, 1992, pp. 40-52.
Felstead, A., The Corporate Paradox: Power and Control in the Business Franchise, Routledge: London 1993.
Fladmoe-Lindquist, K./Jacque, L. L., Control Modes in International Service Operations: The Propensity to Franchise, Management Science, 41, 7, 1995, pp. 1238-1249.
Franchise World Directory 1992, London: Franchise World Magazine 1993.
Guisinger, S./Brewer, T. L., Multinational Enterprise and Economic Analysis: Introduction to the Symposium, Journal of International Business Studies, 29, 1, 1998, pp. 1-3.
Hackett, D. W., The International Expansion of US Franchise Systems: Status and Strategies, Journal of International Business Studies, 7, 1976, pp. 65-75.
Hennart, J-F., Can the `New Forms of Investment' Substitute for the `Old Forms': A Transactions Costs Perspective, Journal Of International Business Studies, 20, 1989, pp. 211-234.
Hill, C. W. L./Hwang, P./Kim, W. C., An Eclectic Theory of the Choice of International Entry Mode, Strategic Management Journal, 11, 1990, pp. 117-128.
Hoffman, R. C./Preble, J. F., Franchising: Selecting a Strategy for Rapid Growth, Long Range Planning, 24, 4, 1991, pp. 74-85.
Hoffman, R. C./Preble, J. F., Franchising into the Twenty-First Century, Business Horizons, 36, 6, 1993, pp. 35-42.
Hofstede, G., The Cultural Relativity of Organizational Practices and Theories, Journal of International Business Studies, 14, 2, 1983, pp. 75-90.
Huszagh, S. M./Huszagh, F. W./McIntyre, F. S., International Franchising in the Context of Competitive Strategy and the Theory of the Firm, International Marketing Review, 9, 5, 1992, pp. 5-18.
Judge, G. G./Hill, R. C./Griffiths, W. E./Luthepohl, H./Lee, T. C., Introduction to the Theory and Practice of Econometrics, 2nd ed., Chichester: John Wiley and Sons 1988.
Julian, S. D./Castrogiovanni, G. J., Franchisor Geographic Expansion, Journal of Small Business Management, 33, 2, 1995, pp 1-11.
Kedia, B. L./Ackerman, D. J./Bush, D. E./Justis, R. T., Determinants of Internationalisation of Franchise Operations by US Franchisors, International Marketing Review, 11, 4, 1994, pp. 56-68.
Kogut, B./Singh, H., The Effect of National Culture on the Choice of Entry Mode, Journal of International Business Studies, 19, 3, 1988, pp. 411-432.
Konigsburg, A. S., Analyzing the International Franchise Opportunity, in Gramatidis, Y./Campbell, D. (eds), International Franchising: An In-Depth Treatment of Business and Legal Techniques, Deventer: Kluwer 1991.
Mathewson, G. F./Winter, R. A., The Economics of Franchise Contracts, Journal of Law and Economics, 28, 3, 1985, pp. 503-526.
McCosker, C. F., Motivations and Methods of Australian Firms Franchising Internationally, University of Southern Queensland 1992.
McGaughey, S./Welch, D./Welch, L., Managerial Influences and SME Internationalisation, in Bjorkman, I./Forsgren, M. (eds) The Nature of the International Firm, Copenhagen: Copenhagen Business School Press 1997.
Mendelsohn, M., The Guide to Franchising, 5th ed., London: Cassell 1992.
Moyne, S. O., Transactions Costs and the Internationalisation of Business Franchising, Centre for International Business Studies, Research Paper 1-97, South Bank University 1997.
National Westminster/British Franchise Association, The 1993 NatWest/BFA Franchise Survey, Henley-on-Thames: British Franchise Association, March 1994.
Norton, S. W., An Empirical Look at Franchising as an Organizational Form, Journal of Business, 61, 2, 1988, pp. 197-218.
Preble, J. F., Global Expansion: The Case of U.S. Fast-Food Franchisors, Journal of Global Marketing, 6, 1/2, 1992, pp. 185-205.
Ronen, S./Shenkar, O., Clustering Countries on Attitudinal Dimensions: A Review and Synthesis, Academy of Management Review, 10, 3, 1985, pp. 435-454.
Rubin, P. H., The Theory of the Firm and the Structure of the Franchise Contract, Journal of Law and Economics, 21, 1978, pp. 223-233.
Sanghavi, N., Retail Franchising in the 1990s, London: Longman 1990.
Thompson, R. S., Agency Costs of Internal Organisation, Chapter 4 in Thompson, R. S./Wright, M. (eds), Internal Organization, Efficiency and Profit, Oxford: Philip Allen 1988.
UK Franchise Directory, 9th ed., Norwich: Franchise Development Services 1994.
United Nations, Statistical Yearbook, United Nations Department for Economic and Social Information and Policy Analysis, New York: United Nations 1995.
Walker, B. J./Cross, J., "A Progress Report on the Scope of International Expansion by U.S. Franchise Systems", in Brown, J. R. (ed.), Franchising Challenges and Opportunities in the 1990s and Beyond, Proceedings of the Third Conference of the Society of Franchising 1989.
Welch, L. S., Diffusion of Franchise System Use in International Operations, International Marketing Review, 6, 5, 1989, pp. 7-19.
Welch, L. S., Internationalisation by Australian Franchisors, Asia Pacific Journal of Management, 7, 2, 1990, pp. 101-121.
Welch, L. S., Developments in International Franchising, Journal of Global Marketing, 6, 1/2, 1992, pp. 81-96.
Welch, L. S., Outward Foreign Licensing by Australian Companies, Chapter 6 in Buckley, P. J./ Ghauri, P. (eds), The Internationalization of the Firm, London: Academic Press 1993.
Whitaker's Almanack, 128th annual ed., London: Whitaker 1996.
Williamson, O. E., Markets and Hierarchies: Analysis and Antitrust Implications -- A Study in the Economics of Internal Organizations, New York: Free Press 1975.
Williamson, O. E., Credible Commitments: Using Hostages to Support Exchange, American Economic Review, 83, 4, 1983, pp. 519-540.
Worldwide Franchise Director, edited by S. Boyles-Martin, Detroit: Gale Research Inc. 1991.
Young, S./Hamill, J./Wheeler, C./Davies, J. R., International Market Entry and Development: Strategies and Management, Hemel Hempstead: Harvester Wheatsheaf 1989.
Manuscript received October 1997, revised June 1999
Fred Burton, Senior Lecturer, Manchester School of Management, UMIST, Manchester, UK.
Adam R. Cross, Lecturer in International Business, Centre for International Business, (CIBUL), Leeds University Business School, University of Leeds, Leeds, UK.
Dr. Mark Rhodes, Department of Economics, University of Warwick, Coventry, UK.
|Printer friendly Cite/link Email Feedback|
|Author:||Burton, Fred; Cross, Adam R.; Rhodes, Mark|
|Publication:||Management International Review|
|Date:||Oct 1, 2000|
|Previous Article:||The Downward Spiral and the US Model Business Principles -- Why MNEs Should Take Responsibility for the Improvement of World-Wide Social and...|
|Next Article:||In memoriam Hans Gunther Meissner.|