Abstract
Royalty rates are an essential contractual provision to reduce the risk of opportunism in franchising partnerships, many of which are international. However, extant research provides limited insights into the factors that determine the level of royalty rates in international franchise agreements. To address this gap, the authors conceptualize and empirically test a model that treats country characteristics (economic potential, legal rights protection, and cultural distance) and contract characteristics (territorial exclusivity and contract duration) as drivers of royalty rates, accounting for product-market profile and service type as potential contextual factors. Using a unique data set comprising 125 international franchising contracts between franchisors and franchisees from 19 countries, the authors find that economic potential (but not territorial exclusivity) is associated with higher royalty rates, whereas legal rights protection, cultural distance, and contract duration are associated with lower royalty rates. Although these relationships are robust across business-to-consumer and business-to-business markets, the impact on royalty rates of economic potential is more pronounced, and that of legal rights protection is less pronounced, for services targeting people (e.g., hospitality) than for services targeting their possessions (e.g., financial services). This work extends the literature exploring the contracting stage of international franchising and provides insights that inform franchisors’ and franchisees’ decisions related to the design of such contracts.
Franchising is a business arrangement in which a firm—the franchisor—permits another firm—the franchisee—to use its brand name and business format to market goods and services. In return, the franchisee pays the franchisor a one-time fee and ongoing royalties (Combs et al. 2011). Originally conceived in the United States, the concept of franchising has gained increased popularity around the world (Dant, Grünhagen, and Windsperger 2011). Today, the franchising industry contributes more than $2 trillion per year to the global economy and provides work to more than 19 million people worldwide (Alon, Apriliyanti, and Henríquez Parodi 2021).
Owing to its substantial economic importance and innate intricacies as an interorganizational business structure, franchising has received much research attention across various disciplines, including management (e.g., Fladmoe-Lindquist and Jacque 1995), international business (e.g., Contractor and Kundu 1998a), entrepreneurship (e.g., Combs and Ketchen 2003), and marketing (e.g., Walker and Etzel 1973). Although researchers adopt different perspectives to investigate franchising practices, most studies are set in a domestic context. However, in practice, franchising across borders is the rule, not the exception. According to the International Franchise Association (2016), four of five franchise companies operate internationally and generate a substantial share of their revenues abroad.
Nevertheless, international franchising is still considered a “Cinderella topic” in the marketing literature (Robson et al. 2018, p. 428); it has received considerably less attention than other foreign market entry modes, such as wholly owned subsidiaries (e.g., Birkinshaw and Morrison 1995; Homburg and Prigge 2014; Leung, Tse, and Yim 2019) and joint ventures (e.g., Calantone and Zhao 2001; Shu, Jin, and Zhou 2017; Tower, Hewett, and Fenik 2019). Previous studies mostly focus on (1) why firms choose to enter foreign markets through franchising (e.g., Erramilli and Rao 1993; Fladmoe-Lindquist and Jacque 1995), (2) what criteria franchisors apply when selecting foreign partners and managing the relationships with them (e.g., Altinay 2006; Brookes and Altinay 2011), and (3) how franchises operate in foreign markets (e.g., Alon 2006; Garg and Rasheed 2003).
Although these contributions shed light on various aspects in the precontracting (i.e., decision to franchise and selection of partners) and postcontracting (i.e., process and conflict management) stages, little is known about the contracting stage of international franchising. This is surprising given that formal contracts are considered a key instrument to govern interfirm relationships, such as franchising partnerships, in an effective and cost-efficient way (Griffith and Zhao 2015). In general, franchisors and franchisees face the risk of opportunistic behavior, such that “one party acts in its own self-interest at the expense of the other party” (Jayachandran et al. 2013, p. 110), a problem referred to as a “double-sided moral hazard” (Lafontaine 1992, p. 279; Milgrom and Roberts 1992). For example, the franchisor may shirk on investments in the franchise brand, or the franchisee may harm the franchisor's brand by lowering its service level to cut costs (Brickley and Dark 1987; Fladmoe-Lindquist and Jacque 1995).
The problem of moral hazard is particularly pronounced in cross-border business relationships (Barnes et al. 2010; Katsikeas, Skarmeas, and Bello 2009; Leonidou et al. 2017). In a domestic setting, franchisors address this problem using monitoring procedures and incentives to encourage compliance (Antia, Mani, and Wathne 2017). However, in an international context, monitoring franchisees is difficult and expensive, which highlights the role of incentives as a governing mechanism. Contracts that are designed to provide each party with just enough incentive to comply can align both parties’ interests and actions and mitigate potential conflicts that could adversely affect business performance (Kashyap, Antia, and Frazier 2012; Zheng et al. 2020).
A key provision to reduce the risk of opportunism in a franchising relationship is the royalty rate (Agrawal and Lal 1995; Kaufmann and Dant 2001). Also referred to as the “sharing parameter” (Brickley 2002, p. 511), the royalty rate represents the percentage of gross sales that a franchisor receives from the franchisee on an ongoing basis as payment for the granted rights (Lal 1990). To effectively reduce behavioral uncertainty, royalty rates need to provide both partners with a sufficiently large incentive not to take advantage of each other. Therefore, franchisors and franchisees are naturally inclined to accept only royalty rates they both perceive as “fair,” keeping in mind their individual goals and risk perceptions with respect to moral hazard. But what factors influence the perceived risk of opportunism in international franchising partnerships and thus determine the accepted level of royalty rates?
Although extant literature offers some insights into the determinants of royalty rates in a national context (e.g., Brickley 2002; Lafontaine 1992; Lafontaine and Shaw 1999; Lanchimba, Windsperger, and Fadairo 2018), empirical investigations of factors that drive royalty rates in an international context are scant. The few empirical studies on international franchising (see Table 1) mostly focus on the role of certain country characteristics for companies’ choice of international franchising as an entry mode (e.g., Burton, Cross, and Rhodes 2000; Contractor and Kundu 1998b). Only Perrigot, López-Fernández, and Eroglu (2013) consider the role of a contract characteristic and show that the duration of a contract (among other factors) matters to the internationalization of franchise networks.
Related Literature.
Most studies also include variables drawn from secondary data sources (e.g., public data related to country characteristics).
As another notable exception, Lafontaine and Oxley (2004) investigate whether and to what extent U.S. franchisors modify their royalty rates when entering the Mexican market and find little evidence of meaningful variations in royalty rates. Although that study represents an important foray into the matter of royalty rates in international franchising, it has two noteworthy limitations. First, the sample is limited to 180 U.S. franchisors, most of which operated few units in Mexico. This raises the possibility that the observed practices are idiosyncratic to U.S. franchisors or franchisors that presumably “test the waters” by authorizing only a few units abroad. Second, the investigation is limited to a single dyad of bordering countries. Thus, the extent to which the findings can be generalized to other country dyads with different characteristics (e.g., geographically and culturally distant countries) is unclear. Overall, the role of different country and contract characteristics as determinants of royalty rates in international franchising agreements remains poorly understood.
To address this gap, we adopt a moral hazard perspective and draw on incentive theory to conceptualize a model of royalty rate determinants. In line with related work, we propose that several country characteristics may determine the level of royalty rates, including the host market's economic potential, the level of legal rights protection, and the cultural distance from the franchise's country of origin. We complement these macro-level determinants with two key contract characteristics that are likely to influence the level of royalty rates: territorial exclusivity and contract duration. We empirically test the proposed model using a unique data set that combines information on 125 international franchising contracts with trademark- and country-related data from several secondary data sources. Our findings confirm all but one of our propositions. Although the economic potential of the target market increases the payable royalties, the strength of legal rights protection in that market and the cultural distance from the franchise's country of origin both lead to lower royalty rates. Furthermore, we find that contract duration is associated with lower royalty rates, while territorial exclusivity appears to be of little relevance. Most relationships are robust across contexts.
This study contributes to the literature in several ways. In general, our findings extend knowledge on international franchising by shifting the spotlight to the contracting stage and offering novel insights into the determinants of royalty rates, an important but understudied aspect of international franchising. Our findings complement those of studies focusing on precontractual aspects, such as partner selection (e.g., Brookes and Altinay 2011), and postcontractual aspects, such as performance (e.g., Brouthers 2002). By focusing on royalty rates—as a specific formal contractual provision that serves as an incentivization mechanism—and its determinants, this study adds to extant research on governance mechanisms in international interfirm relationships (e.g., Chu, Lai, and Wang 2020; Griffith and Zhao 2015; Zheng et al. 2020). It offers rare quantitative empirical evidence in support of the relevance of a host market's economic potential, level of legal rights protection, and cultural distance for international franchising phenomena (Rosado-Serrano, Paul, and Dikova 2018). Furthermore, our findings regarding the role of specific contractual terms, such as the duration of a contract, complement recent findings on global contract features, such as their overall level of completeness or specificity (Griffith and Zhao 2015; Kashyap and Murtha 2017; Mooi and Ghosh 2010).
We organize the article as follows: We first elaborate on the problem of moral hazard, which is grounded in agency theory, and draw on incentive theory to explain the role of royalty rates in reducing behavioral uncertainty within international franchising partnerships. We then theorize how country characteristics and contract characteristics are related to royalty rates and how contextual factors (product-market profile and service type) may moderate some of these relationships. Next, we describe the data and methods we employ to empirically test the proposed model, after which we present the results, including several robustness checks. We discuss how our findings extend international marketing theory and provide recommendations for managers concerned with international franchise contracts. Finally, we conclude by addressing noteworthy limitations of our work and outlining avenues for future research.
Theoretical Background and Hypothesis Development
Moral Hazard in International Franchising Partnerships
In general, franchising involves two legally independent firms that start a business relationship in which the franchisor is contractually authorized to enforce certain actions and the cooperating franchisee is contractually bound to perform and comply with said actions (Alon, Ni, and Wang 2012). Thus, franchising research has a long tradition of employing agency theory to explain the partnering firms’ behavior (Agrawal and Lal 1995; Kashyap, Antia, and Frazier 2012; Ni and Alon 2010). Agency theory explains how business principals (here, franchisors) control agents’ (here, franchisees) behaviors efficiently (Bergen, Dutta, and Walker 1992). Principal–agent relationships, such as franchisor–franchisee relationships, are typically characterized by divergent goals, risk preferences, and levels of information (Kashyap, Antia, and Frazier 2012). Whereas franchisors aim to expand into new markets and develop additional revenue streams without harming or jeopardizing their brand, franchisees strive for profit maximization and usually know the local market better than the franchisor (Kidwell, Nygaard, and Silkoset 2007; Wathne and Heide 2000; Zheng et al. 2020). The franchisors’ goal of maximizing systemwide sales and the franchisees’ goal of maximizing their own net profits are not perfectly aligned (Frazer and Grace 2017; Nair, Tikoo, and Liu 2009).
This goal misalignment leads to behavioral uncertainty, as both parties face the risk of opportunistic behavior from the other firm. One party may not act in good faith, shirk on its performance obligations, or leverage possible information advantages to achieve its goal while sacrificing or inhibiting the other party's goals, a situation referred to as “moral hazard” (Lafontaine 1992; Milgrom and Roberts 1992). Importantly, moral hazard can occur on both the franchisor's and the franchisee's sides, which implies a “double-sided moral hazard” (Lafontaine 1992, p. 279). For example, a franchisor may shirk on investing in the franchise brand or other expected efforts, such as support and training (Agrawal and Lal 1995; Brickley 2002; Shane 1998). Conversely, a franchisee may provide a suboptimal service level to cut costs and increase profitability, which could hurt the franchisor's brand (Brickley 2002; Brickley and Dark 1987; Hoffman, Munemo, and Watson 2016).
The issue of moral hazard is particularly pronounced in international partnerships (Barnes et al. 2010; Katsikeas, Skarmeas, and Bello 2009; Leonidou et al. 2017). Compared with domestic partnerships, information asymmetries with respect to the conditions in the target market, such as economic, legal, and cultural aspects, tend to be greater in international partnerships (Doherty and Quinn 1999; Kidwell, Nygaard, and Silkoset 2007). Furthermore, predicting a foreign partner's behavior can be more difficult due to cultural differences (Durand, Turkina, and Robson 2016; Elango 2007; Erramilli and Rao 1993). Similarly, the propensity to trust others varies across countries, increasing the likelihood of different levels of trust between international business partners (Wang, Zhang, and Zhou 2020; Westjohn et al. 2021). As a result, franchisors and franchisees perceive a significant risk of opportunistic behavior that can pose considerable challenges in establishing a strong and mutually beneficial business relationship.
Royalty Rates as an Incentivization Mechanism
In general, the problem of moral hazard can be addressed using motivation-influencing governance mechanisms (Antia, Mani, and Wathne 2017), namely monitoring and incentivization. “Monitoring” describes an auditing process of the partner's performance, based on some ex ante agreed-on criteria and standards (Elango 2007; Heide and Wathne 2006). Though effective, monitoring is more challenging and expensive in international franchising than in national franchising in at least two ways. First, as geographic distance between the franchisor's headquarters and the host market increases, monitoring costs tend to increase as well (Brickley and Dark 1987; Rubin 1978). Second, aside from more costly and time-consuming inspection visits (Agrawal and Lal 1995), international differences in terms of language or accounting systems may further add to the complexity of the monitoring process (Kretinin, Anokhin, and Wincent 2020). Accordingly, empirical research frequently employs measures of (geographic) distance as proxies for monitoring costs (Brickley and Dark 1987; Burton, Cross, and Rhodes 2000; Combs and Ketchen 2003).
The difficulty of monitoring foreign franchisees underscores the importance of incentives as a mechanism to address behavioral uncertainty in international franchising arrangements. According to incentive theory, incentives are “monetary [and nonmonetary] inducements for an independent reselling agent to resell a principal's product” (Gilliland and Kim 2014, p. 362). As such, incentives are viewed as a mechanism to modify the behavior of channel members. By structuring interorganizational relationships “in such a way that particular actions are explicitly rewarded or punished” (Heide and Wathne 2006, p. 95), incentives induce compliance with contractual demands and motivate aligned behaviors (Gilliland and Kim 2014).
The literature typically highlights the role of two monetary components of franchise contracts, which may be used as incentive instruments: the (initial) franchise fee, a single, lump-sum payment that the franchisee pays to the franchisor at the beginning of the contract, and the (ongoing) royalty rate, which specifies the amount of royalties the franchisee needs to pay to the franchisor on an ongoing basis, typically calculated as a percentage of gross sales (Brickley 2002; Lafontaine 1992; Lal 1990).
Theoretically, these two monetary provisions are likely to be inversely related, such that higher franchise fees imply lower royalty rates, and vice versa (Blair and Kaserman 1982; Rubin 1978; Shane 1998). The underlying notion is that franchise fees act as a rent extraction mechanism. The more additional profits can be extracted up front, the less surplus rent needs to be extracted through ongoing royalties (assuming that future revenues are predictable). However, extant empirical evidence of this inverse relationship is mixed. Several studies show no significant relationship between royalty rates and franchise fees (e.g., Dnes 1992; Lafontaine 1992; Lafontaine and Shaw 1999; Maruyama and Yamashita 2012). 1 Thus, scholars argue that initial fees are set not to systematically recover future rents but “mostly to cover the costs they incur in setting up a new franchisee” (Lafontaine and Shaw 1999, p. 1064). 2
This finding has two important implications. First, in the long run, franchise fees account for a relatively small portion of the revenues that franchisors extract from franchisees. Lafontaine and Shaw (1999) estimate that, on average, franchisee fees account for only approximately 8% of the total revenues paid over the duration of a contract. Second, franchise fees are primarily a “break-even proposition as far as the franchisor is concerned” (Bond 1998, p. 21). Thus, they are not related to incentive issues (Lanchimba, Windsperger, and Fadairo 2018), which renders the royalty rate the main incentive instrument available to reduce behavioral uncertainty in a franchising relationship (Agrawal and Lal 1995; Kaufmann and Dant 2001).
From the franchisor's perspective, a high royalty rate is desirable because it implies larger revenues. A high royalty rate thus leads to larger returns on prior investments and facilitates new investments (Agrawal and Lal 1995). At the same time, a high royalty rate reinforces franchisees’ expectations of continuous investments in the joint business (e.g., advertising expenditures aimed to maintain the brand name value), providing the franchisor an incentive to commit to such investments (Rubin 1978; Vásquez 2005). From the franchisee's perspective, a low royalty rate is desirable because lower payments to the franchisor leave the franchisee a larger fraction of residual income (Jell-Ojobor and Windsperger 2017). Conversely, a low royalty rate also implies less available resources for the franchisor to support the franchisee directly (through the provision of training and support) and indirectly (through brand name investments). Thus, agreeing on a low royalty rate provides an incentive to the franchisee to maintain and grow demand through service improvement from its own investments (Shane and Foo 1999).
To effectively reduce behavioral uncertainty, royalty rates need to provide both the franchisor and the franchisee with just enough incentive to comply and refrain from acting opportunistically. Therefore, both parties will accept only royalty rates they perceive as “fair,” taking into account their individual goals and perceived risk of moral hazard. On the one hand, franchisors and franchisees pursue divergent goals: whereas franchisors are primarily interested in growing the franchise to maximize sales (i.e., more revenue through royalties), franchisees are primarily interested in maximizing the profitability of their franchise units (i.e., more residual income after paying the fees to the franchisor) (Frazer and Grace 2017; Nair, Tikoo, and Liu 2009). 3 On the other hand, franchisors’ and franchisees’ perceptions of risk of opportunistic behavior likely depend on characteristics of the host market and the contract itself. For example, information asymmetry related to the local conditions in the host market exacerbates the behavioral uncertainty between the partners (Wathne and Heide 2000; Welsh, Alon, and Falbe 2006).
As a consequence, we expect royalty rates to vary as a function of certain country and contract characteristics. Figure 1 depicts our conceptual model, which contains the proposed determinants of royalty rates alongside the product-market profile and service type as contextual factors. Next, we elaborate on how each driver relates to the issue of double-sided moral hazard and the associated implications for the royalty rates both partners are willing to accept.
Conceptual Model.
Country Characteristics as Drivers of Royalty Rates
To conceptualize the determinants of royalty rates at the country level, we follow previous research in international marketing (Burgess and Steenkamp 2006; Eisend, Evanschitzky, and Calantone 2016; Peng, Wang, and Jiang 2008) that highlights the importance of three distinct but interrelated country characteristics: a country's economic, regulatory, and cultural characteristics. Jointly, these factors “provide structure to society, albeit in different ways” (Burgess and Steenkamp 2006, p. 341) and “shape the way how companies operate and … perform” in that country (Eisend, Evanschitzky, and Calantone 2016, p. 42).
In general, a country's economic conditions determine its economic development, with direct implications for consumers’ purchasing power and consumption preferences (Whitley 1992; Wu 2013); its regulative (or legislative) framework affects how effectively it can establish and enforce legislations and regulations, which is crucial for reducing uncertainty in business transactions (Erramilli and Rao 1993; Hoffman, Munemo, and Watson 2016; Wu 2013); and its culture has a strong influence on the behavior of individuals, as both consumers and members of organizations such as firms (Hofstede 2001). Thus, a country's economic, legal, and cultural contexts can be sources of opportunities (e.g., favorable economic conditions) and risks (e.g., weak protection of legal rights), which may affect the royalty rate a franchisor and franchisee consider appropriate.
Economic potential
Economic potential refers to the host market's attractiveness in terms of prospective revenues. Because larger country markets may reap larger sales volumes, they are generally considered more attractive than smaller markets (Aliouche 2017; Alon and Banai 2000; Dunning and Lundan 2008). To measure the size of a country market, researchers commonly rely on a country's gross domestic product (GDP), or the total value of products and services produced in a year, or GDP per capita, to account for countries’ economic output generated by populations of different sizes (Jell-Ojobor and Windsperger 2017). Thus, GDP per capita reflects a citizen's average economic output and is indicative of the material wealth and living standard in a country (Magnusson, Westjohn, and Boggs 2009; United Nations Development Programme 2021). Higher levels of economic output go hand in hand with consumers’ increased purchase power and discretionary spending (Kravis, Heston, and Summers 1978; Summers and Heston 1991).
From the franchisor's perspective, the economic potential of a host market presents the prospect of generating large sales volumes and, thus, royalties. Therefore, the franchisor has a strong incentive to fulfill its obligations in terms of brand investments and business process support if the market is of significant economic importance (Melo, Borini, and Ogasavara 2019; Quelch 1985). These activities require resources that the franchisor may account for through higher royalty rates. At the same time, with increasing market size, the franchisor's opportunity costs of not entering the market directly also increase (Jayachandran et al. 2013). These opportunity costs of notentering an economically promising market directly (i.e., through own units) may further increase the royalty rate the franchisor deems appropriate.
From a profitability standpoint, high royalty rates are generally unattractive to a franchisee and provide it little incentive to invest in developing the business. A franchisee might even shirk on its efforts to maintain high service quality to increase short-term profitability. Yet, with increasing market size, the franchisee's dependence on the franchisor's efforts to maintain and enhance the reputation of the franchise also increases. Thus, in economically promising markets, we expect franchisees to be more accepting of higher royalty rates to prevent the franchisor from shirking the needed support (Jayachandran et al. 2013). This expectation is in line with previous findings in the contexts of technology licensing (Caves, Crookell, and Killing 1983) and brand licensing (Jayachandran et al. 2013) that suggest that licensees tend to pay more in larger/more lucrative markets.
Legal rights protection
Legal rules, as formalized in written laws (e.g., commercial, contractual, and competition law), structure business relationships and, ideally, support fair and predictable business activities in a specific market (Buchan 2014; Jell-Ojobor and Windsperger 2017). Laws inform market participants about “allowed” actions, enabling them to better anticipate the actions of other market participants and mitigate uncertainty. Furthermore, laws specify enforceable consequences if an actor crosses a legal limit (e.g., intellectual property rights violations) and thus protect legitimate interests (Cavusgil, Deligonul, and Zhang 2004).
Under the rule of law, firms typically have two seemingly opposing motivations. On the one hand, they favor unrestricted freedom in doing business and perceive market and business regulations as a burden. In the context of franchising, Melo, Borini, and Ogasavara (2019) show that business freedom is positively associated with a franchisor's commitment to internationalization. Burdensome and redundant regulations that restrict firms in closing contracts and conducting transactions as they deem fit are the most common barriers that prevent them from entering a new foreign market (Brouthers 2002; Gatignon and Anderson 1988). On the other hand, firms have a vital interest in reliable enforcement mechanisms (Jell-Ojobor and Windsperger 2017). Weak legal systems that are complex, nontransparent, and potentially corrupt pose a significant risk to the protection of the franchisor's intellectual property and can seriously harm the franchise (Jell-Ojobor and Alon 2017). Under a misalignment of goals, “conflicts on contractual matters …, which escalate to costly litigation” (Zheng et al. 2020, p. 150), are not uncommon between franchisors and franchisees. Such conflicts can severely harm the performance of the business relationship and potentially result in its termination (Griffith and Zhao 2015; Nair, Tikoo, and Liu 2009; Zhang, Griffith, and Tamer Cavusgil 2006). A weak legal system with inefficient legal enforcement mechanisms exacerbates these risks, as disputes in court take time to resolve and their outcome is less certain.
This uncertainty may be a burden to the franchisor, as a franchisee might have an information advantage regarding the idiosyncrasies of the host country's legal environment. Therefore, the franchisor may see a need to financially hedge against legal risks through increased royalty rates (Brouthers 2002; Melo, Borini, and Ogasavara 2019). Conversely, a strong legal system that effectively protects and enforces a franchisor's legal rights reduces the need for hedging. Thus, we project royalty rates to be lower (higher) in host markets that have strong (weak) protection of legal rights. This notion receives support from Jayachandran et al.’s (2013) finding that royalty rates for licensed brands tend to be lower in countries with strong intellectual property rights protection.
Cultural distance
Cultural distance, or “the degree to which culture in one country is different from another country's culture” (Griffith, Dean, and Hoppner 2021, p. 23), is a crucial factor in international business considerations, including international franchising (Contractor and Kundu 1998b; Elango 2007). In general, the term “culture” refers to the set of values that are shared by members of a social group, such as a nation, and that distinguish them from other groups (Schwartz 2014). These shared values reflect the group's “collective programming of the mind” (Hofstede 2001, p. 1) and determine social norms and expectations, ultimately shaping the behavior of individuals and organizations (Beugelsdijk, Kostova, and Roth 2017).
In short, the more (less) two countries differ in terms of “social norms, religions, languages and ethnicities” (Hewett and Krasnikov 2016, p. 62), the more (less) “distant” they are considered (Kogut and Singh 1988). The general rationale is that, with increasing cultural distance, behavioral patterns and normative expectations tend to diverge, which renders agreement and coordination with foreign business partners, as well as marketing to foreign consumers, more challenging (Durand, Turkina, and Robson 2016; Elango 2007; López-Duarte and Vidal-Suárez 2010). Previous research corroborates this notion by demonstrating that the cultural distance of a specific market (to a firm's home market) influences how a firm chooses to enter that market (e.g., Dow and Larimo 2009), consumers’ acceptance of the newly introduced product (e.g., Craig, Greene, and Douglas 2005), and the duration of cross-cultural business relationships (e.g., Tower, Hewett, and Fenik 2019). We argue that a host market's cultural distance from the franchisor's home market has important implications for both parties’ perceived risk of moral hazard and, in turn, for the royalty rate that provides them sufficient incentives to refrain from acting opportunistically.
From the franchisor's perspective, the host market's cultural distance increases the perceived threat of moral hazard (Contractor and Kundu 1998b; Elango 2007). The franchisee is a cultural insider that bridges the culture gap between local consumers and the franchisor (Erramilli and Rao 1993). The more pronounced cultural differences are, the more the franchisor relies on the franchisee's accurate “cultural translation” of the business concept (Kretinin, Anokhin, and Wincent 2020). The franchisee could take advantage of the information asymmetry with respect to local customs and business practices, for example, by misrepresenting or withholding relevant information. Such opportunistic behavior is more likely to occur when a franchisee's profitability is at risk. Thus, lower royalty rates give the franchisee an incentive to refrain from actions that hurt the franchisor.
Furthermore, lower royalty rates account for the increased uncertainty franchisees face regarding consumer acceptance of the foreign business concept (Wirtz and Lovelock 2016). The greater the cultural distance between the franchise's country of origin and a foreign host market, the more difficult it is to predict its success, given differences in consumers’ perceptions and expectations of services (Ozdemir and Hewett 2010). Examples such as McDonald's in Vietnam (CNBC 2018b), Starbucks in Australia (CNBC 2018c), and Dunkin’ Donuts in India (CNBC 2018a) show that international franchising attempts can fail despite a franchisor's high quality standards and commercial support. At the same time, a franchisee's initial investments are considerable (e.g., the up-front fee to gain access to the franchisor's brand name, service procedures, training, and uniforms; Brickley and Dark 1987; Shane 1998), which renders franchising in culturally distant countries risky for franchisees. The prospect of high returns (due to lower royalty rates) accounts for this risk and reduces incentives to increase profitability through deceitful conduct. Thus, we expect the following:
Contract Characteristics as Drivers of Royalty Rates
International franchising contracts are complex and nonstandardized legally binding business agreements (Melo, Borini, and Ogasavara 2019). Although the composition and naming of the specific contractual elements vary depending on the service industry (e.g., hotel, restaurant; Alon, Ni, and Wang 2012; Ni and Alon 2010), all contracts have a few key elements in common, such as information on the focal trademark, the territory granted to the franchisee, the exclusivity of the granted territorial rights, the duration of the agreement, and the fees payable to the franchisor. Additional terms provide further details on each party's obligations and rights regarding training, quality standards, and other operational aspects (International Franchise Association 2022). With the trademark and territory specified, we focus on the two remaining key contract elements that may affect both parties’ perceived risk of moral hazard: territorial exclusivity and contract duration (Lafontaine and Shaw 1999).
Territorial exclusivity
Territorial exclusivity rights entitle a single franchisee to operate all units of a franchise in a given geographical territory over a specified period of time. Such exclusive business arrangements can take either the direct form of “area development franchising,” where the franchisee maintains the right to open and operate multiple units on its own (Bodey, Weaven, and Grace 2013), or the indirect form of “master franchising,” where the franchisee is permitted to sell the business format on to local subfranchisees, in addition to operating its own units (Burton, Cross, and Rhodes 2000). Territorial exclusivity allows the franchisee to exploit a market without competing against other franchisees of the same brand (i.e., intrabrand competition). The lower number of franchisees in a given territory decreases transaction costs due to more efficient coordination and monitoring processes (Brouthers 2002; Ni and Alon 2010). Territorial exclusivity also bolsters the franchisee's role, as the development of the franchise business in a specific country, and thus the royalties a franchisor can extract from a market, critically depends on their efforts. Because franchisees with multiple units have a stronger interest in long-term profitability and preserving the brand name value than single-unit franchisees, they are less likely to engage in opportunism (Dant et al. 2013; Garg and Rasheed 2003). Therefore, the interests of a franchisee with territorial exclusivity rights tend to be better aligned with those of the franchisor, especially master franchisees, which assume extended strategic and operational control over and thus responsibility for the franchise's operations in the host market (Jell-Ojobor, Alon, and Windsperger 2022).
The strategic importance of exclusive contractors may cause the franchisor to accept lower royalty rates for two reasons. First, lower royalty rates increase local units’ profitability, providing the franchisee with an incentive to expand the business by opening more outlets (which is in the franchisor's interest). Second, ceteris paribus, lower royalty rates give the franchisee less incentive to engage in opportunistic behaviors, which could be more harmful to business outcomes because the risks are not diversified across different franchisees. Accordingly, we posit:
Contract duration
Contract duration refers to the agreed-on duration of a contract's validity (typically in years). In other words, this contract characteristic reflects the time scope for which the franchisor and franchisee commit to their relationship (even though premature contract terminations and contract extensions are possible; Lafontaine and Shaw 1999). Previous research on the role of contract duration in franchising partnerships suggests that a longer contract duration enhances performance in terms of growth and profitability through lowered coordination and control costs (Gorovaia 2019). In addition to lower transaction costs, long-lasting contracts promote commitment and interorganizational learning, enabling both parties to develop relational strategic assets and leverage knowledge more efficiently (Gorovaia and Windsperger 2018). Moreover, longer contracts increase both partners’ financial interdependence and help franchisees identify with the business interests of the franchisor more strongly (El Akremi, Perrigot, and Piot-Lepetit 2015).
From a moral hazard perspective, a longer contract duration, which reflects a willingness to commit to a long-term relationship, renders opportunistic behavior less likely. If both parties make a long-term commitment, franchisees have a stronger incentive to invest in the strategic development of the franchise (Brickley, Misra, and Van Horn 2006). Investments in the franchise and the protection of the brand are more valuable to franchisees, as they perceive the franchise as an asset to be leveraged for an extended period. By contrast, short contracts entail a significant risk of opportunism on the part of the franchisee, because the limited time scope creates pressure to amortize the initial investments and encourages profit maximization, potentially at the expense of the service level. Considering that the issue of moral hazard is particularly pronounced in the short run but becomes less concerning in the long run, we expect royalty rates to be negatively related to contract duration. This expectation echoes previous findings in a licensing context (Jayachandran et al. 2013; Quelch 1985). Thus:
Contextual Factors
To provide more nuanced insights into the role of each driver, we further consider the product-market profile and the type of service that is franchised as contextual factors. In terms of product-market profile, we adopt the widely used distinction between business-to-consumer (B2C) and business-to-business (B2B) markets (Dwyer and Tanner 2008; Srinivasan, Lilien, and Sridhar 2011). B2B markets fundamentally differ from B2C markets in several ways. Compared with B2C markets, B2B markets are characterized by a smaller number of customers who make fewer but far larger individual purchases, more complex and formalized purchasing processes that involve multiple stakeholders (e.g., managers, engineers, lawyers), and higher levels of buyer–seller loyalty due to customized offers and integrated processes (Dotzel and Shankar 2019; Lilien 2016).
The fundamental differences between B2B and B2C markets could potentially influence the relevance of the proposed determinants of royalty rate. For example, it is conceivable that, in B2B markets, deeper relationships between franchisees and their (fewer) customers as well as the lock-in effects due to customization (e.g., consulting services) and process integration (e.g., facility management) imply a generally lower risk of litigation, rendering legal rights protection a less important factor in the determination of royalty rates. Similarly, the impact of cultural distance on perceived moral hazard might be less pronounced in B2B settings, compared with B2C settings. Typically, business purchases involve formal evaluation processes that reduce the influence of subjective, culturally shaped preferences. Thus, we expect the uncertainty of a B2B franchise's acceptance in a foreign market to be less dependent on cultural aspects.
In terms of service type, we distinguish between services that require customers’ participation in that they involve their bodies or minds in some way (e.g., restaurants, health and fitness facilities, educational services; termed “people-processing services”) and services that involve customers’ tangible or nontangible possessions (e.g., household services, recycling and sanitation services, financial services; termed “possession-processing services”) (Wirtz and Lovelock 2016, p. 23). Distinguishing service types is useful because services vary in complexity and the amount of tacit knowledge (e.g., intangible system know-how) that needs to be codified and transferred to a franchisee (Lovelock 1983; Nonaka and Takeuchi 1995; Zander and Kogut 1995). We argue that people-processing services tend to involve more complex operating procedures and routines (partly because of more frequent and intense customer interactions) and thus are more difficult to codify and transfer than possession-processing services. For example, the provision of a standardized, high-quality output requires significantly more documentation and training for a restaurant than for a dry-cleaning service.
This distinction can have important implications for the role of factors such as legal rights protection. For example, the complexity of people-processing services may render legal rights protection less relevant, because contract violations (e.g., withheld efforts) are more difficult to detect and verify. In addition, the business format is more difficult for third parties to imitate, which reduces potential concerns about copycats that could harm the brand.
Given the limited previous research on these contextual effects (as well as the considerable number of relationships involved), we refrain from presenting and testing formal hypotheses but examine these effects in an exploratory manner.
Methodology
To test our hypotheses, we used a unique data set that combines proprietary contract data with publicly available country data. We obtained 125 international franchising contracts from RoyaltyStat, a commercial database that compiles agreements filed with the U.S. Securities and Exchange Commission. The data set contains all available international franchising contracts since 1994 (up to May 2019). Each contract includes information on the type of business (and trademark) that is franchised, the relevant territory (one or more countries) for which the rights are granted, the payable royalty rate, the duration of the contract, and any territorial exclusivity rights.
Overall, the data cover 125 unique franchisor–franchisee–country combinations, which constitute the units of analysis. The 119 unique franchisors involved in these contracts originate from 19 countries, including the United States, Canada, Australia, Brazil, the United Kingdom, Japan, and Germany. The 125 franchisees originate from eight countries: the United States, Canada, Mexico, Poland, Turkey, Japan, Hong Kong, and Thailand. 4 The data set is equally heterogeneous in terms of the industries it covers. The contracts pertain to 24 service industries, including restaurants, household services, retail trade, travel services, and business services. Table 2 provides an overview of key sample characteristics. We enriched the contractual data from RoyaltyStat with trademark- and country-related data from several secondary sources, such as the Trademark Electronic Search System (TESS) and the World Bank. We provide more details on all measures in the following subsections.
Sample Characteristics.
aWe list the continents in a consistent order for readability reasons (i.e., North America–Europe includes contracts closed between U.S. franchisors and European franchisees and between European franchisors and U.S. franchisees).
Notes: N = 125.
Measurements
Table 3 provides an overview of all constructs, their operationalization, and the corresponding data source. To ensure linearity, we log-transformed all continuous variables before using them in the main analysis.
Summary of Variable Operationalization.
Categorized by three independent coders (on the basis of contract information provided by RoyaltyStat).
Royalty rate
We measured royalty rate as a percentage of a franchisee's (net) sales revenues, as defined in the contracts we drew from the RoyaltyStat database. If royalty rates were tiered (e.g., 4%–8%), we used the average as a royalty rate (e.g., 6%). To ensure that predicted values fell between 0 and 1 (according to the notion of percentages), we applied a logit transformation to the variable (following Jayachandran et al. 2013).
Economic potential
To approximate the target market's economic potential, we used a country's GDP per capita in U.S. dollars (in line with Contractor and Kundu 1998a, 1998b; Melo, Borini, and Ogasavara 2019), which we obtained from the World Bank (2022a). We refined the operationalization in two ways. First, we used lagged values to ensure high correspondence between the data and the time of decision making. In other words, we used the target market's GDP per capita in 2014 if a contract was filed with the U.S. Securities and Exchange Commission in 2015. Second, because our data exclusively deal with service industries, we further weighted the measure according to the importance of services in a specific economy. Thus, we used the share of services’ added value (i.e., net output of services; in percentage of GDP) as a weighting variable (World Bank 2022b). If a contract spans multiple territories (e.g., Canada and the United States), we aggregated (weighted) market sizes.
Legal rights protection
We used the World Bank's (2022c) Strength of Legal Rights Index to operationalize the degree of legal rights protection in a specific target market. This index ranges from 0 (weak) to 12 (strong) and captures the degree to which a country's collateral and bankruptcy laws protect the financial interests of borrowers and lenders. Despite the focus on financial interests, we can reasonably assume that the strength of legal rights is largely harmonized across the same legal system's business-related domains (e.g., corporate, property, and labor law). Therefore, we consider this index a good proxy of a legal system's overall capability to protect and enforce businesses’ legal rights. We had index data available for 2013–2018. Thus, we recorded the value that was temporally closest to the year of filing for each target country. Because 94 contracts were filed before 2013, we needed to impute values for these cases. Multivariate imputation methods were not an option, as values are missing for all units. An examination of each country's index scores for the 2013–2018 period showed no change for all countries, with the exception of Turkey, whose score improved from 2017 onward. Given the limited variation over time, we assume that the 2013 values also reflect past levels of legal rights protection reasonably well.
However, the longer a contract dates back, the more problematic this assumption becomes; therefore, we conducted additional, in-depth research to identify whether any relevant legal changes occurred in any of six pertinent countries (i.e., countries mentioned in contracts before 2013). Specifically, we checked for potential changes in the national collateral and bankruptcy laws (in accordance with the World Bank's operationalization) of Australia, Canada, Chile, New Zealand, Poland, and the United States. We identified legislative changes in Canada (three revisions of the Personal Property Security Act since 1997), Poland (a new financial collateral directive and bankruptcy law in 2002 and 2003, respectively), and the United States (Bankruptcy Abuse Prevention and Consumer Protection Act in 2005). Because all these initiatives aimed to strengthen the protection of legal rights, we considered lower levels of legal rights protection (2013 index score reduced by one standard deviation; we report additional sensitivity analyses in the “Additional Analyses” section) for all years before the new legislations. For example, we use a Strength of Legal Rights Index score of 9.62 for U.S.-related contracts that were filed before 2005 and a score of 11 for all subsequent contracts. If a contract spanned multiple territories (e.g., Canada and the United States), we used the average of the countries’ index scores.
Cultural distance
We measured the cultural distance between the franchise's home country and the host market using a summary statistic based on Hofstede's (2001) four original dimensions of national culture (individualism/collectivism, uncertainty avoidance, power distance, and masculinity/femininity). We selected this framework because of its suitability for investigating phenomena related to managerial decision making (Griffith, Dean, and Hoppner 2021).
We calculated the statistics using Messner's (2021) geometrical measurement approach, which offers several important advantages over conventional arithmetic approaches. In summary, arithmetic approaches such as Kogut and Singh's (1988) index neglect the evident interdependence of some of the cultural dimensions, falsely treat a country’s position on Hofstede's dimensions as absolute, and result in different distances for identical country pairs depending on the considered sample of countries (Messner 2021). Geometrical approaches overcome these two shortcomings by (1) treating culture as a weight vector in an n-dimensional feature space and (2) using the angle of heterogeneity between two such vectors as an indication of the cultural distance between two countries (Desmet, Ortuño-Ortín, and Wacziarg 2017; Messner 2021). Importantly, this metric “does not necessitate independence of the cultural dimensions, does not call for the dimensions to have equivalent measurement scales, and produces identical results independent of the number of countries the dimensions are based on” (Messner 2021, p. 57). The formula we used to calculate the distance for each country pair is documented in Messner’s (2021) Mathematical Appendix (Equation 8). If rights were granted for multiple territories (7% of all cases; e.g., a Japanese franchisor granting rights to operate a franchise in the United States and Canada), we used the average cultural distance.
Territorial exclusivity
Territorial exclusivity indicates whether the franchisor grants a franchisee the exclusive rights to operate the franchise in the target market (i.e., without intrabrand competition). This binary variable indicates the nonexclusive (coded as 0) or exclusive (coded as 1) nature of the granted territorial rights. Only two contracts also included the right to sublicense the franchise, which means that the vast majority of exclusive contracts in our sample represent area development agreements rather than master franchising agreements.
Contract duration
The contract data from RoyaltyStat also include information on the duration of a contract (in years). In practice, the typical duration of franchise agreements is ten years (Brickley, Misra, and Van Horn 2006), which is also reflected by the average contract duration in our sample (M = 11.03, SD = 12.27). If a franchisor granted perpetual rights, we used 99 years (following Jayachandran et al. 2013).
Contextual factors
As mentioned previously, we distinguish between two different product-market profiles and two different types of services. The “product-market profile” variable indicates whether a service targets business customers (coded as 1) or consumers (and businesses) (coded as 0). The “service type” variable indicates whether a service involves customers’ bodies or minds in some way (coded as 1), such as restaurants, health and fitness facilities, and educational services, or processes customers’ tangible or nontangible possessions (coded as 0), such as household services, recycling and sanitation services, and financial services (Wirtz and Lovelock 2016). Three independent coders classified all contracts according to these categories and achieved initial intercoder reliabilities (Krippendorff's alpha) of .93 (product-market profile) and .88 (service type), respectively. We resolved the remaining four cases that were coded differently by at least one coder.
Brand strength
In line with previous research, we expect high brand strength to enable franchisors to command higher royalty rates (Lanchimba, Windsperger, and Fadairo 2018; Panda, Paswan, and Mishra 2018; Pénard, Raynaud, and Saussier 2003). Large, well-established franchises tend to be perceived as more prestigious and resourceful (Shane and Foo 1999). By signaling a franchise's attractiveness and wide acceptance, a strong brand raises franchisees’ expectations of future demand. At the same time, a strong brand also requires a franchisor to constantly invest, which in turn necessitates higher royalties, compared with weaker brands (Lanchimba, Windsperger, and Fadairo 2018).
To control for differences in brand strength, we used the age of the franchised trademark (in years) at the time of contract filing as a proxy (see Lanchimba, Windsperger, and Fadairo 2018). To obtain this information, we used the TESS database of the U.S. Patent and Trademark Office. We recorded each trademark's year of registration and calculated the corresponding age. For the few trademarks for which this information was not available, we identified a brand's year of founding using publicly available information (e.g., corporate websites).
Business sector
Because our limited sample size does not allow us to include industry-specific dummies, we captured some of the industry-related heterogeneity using two dummy variables. The dummies indicate whether a contract pertains to a retail business (coded as 1) or a restaurant (coded as 1), respectively. Three independent coders classified all contracts according to these business sectors. The initial intercoder reliabilities were high, ranging from .94 (retail sector) to .95 (gastronomy sector). Divergent codings were discussed and resolved.
Period
Again, because of the small sample size, the available data restrict the use of year dummies. To account for a proportion of the unobserved heterogeneity across time, we included two additional dummy variables that help distinguish three distinct periods. The first period includes all contracts that were closed before 2001, which marked the beginning of a global economic downturn following the September 11 terrorist attacks. The second period spans from 2002 to 2009 and marks the end of the Great Recession. The third period includes all remaining contracts (2010 and later) and serves as the reference category (coded as 0).
Table 4 reports correlations and basic descriptive statistics. To support interpretability, we report the correlations and descriptive statistics using the natural units of each (nontransformed) variable.
Correlation Table.
*Statistically significant (p < .10).
Notes: Values displayed refer to the variables before being transformed (except for dummy variables). Descriptive statistics associated with economic potential are reported in thousands of U.S. dollars.
Analytical Procedure
We employed ordinary least squares regression to test the hypothesized relationships. First, we estimated a model that includes only the control variables and contextual factors (Model 1). Second, we added the contract characteristics to the model (Model 2). Third, we included the country characteristics (Model 3). Finally, we ran a series of additional models to probe the moderating roles of product-market profile and service type. In each step, we also assessed all variables’ variance inflation factors (VIFs), because some of the explanatory variables, especially the country characteristics, may be correlated. After assessing the empirical evidence in support of our hypotheses, we conducted several robustness checks (for full documentation, see the Web Appendix). Owing to our small sample size, we lacked the statistical power to detect all but the strongest effects (Cohen 1988). Therefore, to avoid Type II errors, we relaxed the threshold for statistical significance to .10, following other studies (e.g., Magnusson, Westjohn, and Boggs 2009; Parboteeah, Hoegl, and Cullen 2008). We report the results of all analyses in the next section.
Results
Table 5 summarizes the main results. The results from Model 1 show a positive coefficient associated with product-market profile, indicating that royalty rates tend to be higher for B2B services (β = .309, p = .031). Furthermore, the negative coefficients associated with the time dummies (2002–2009: β = −.279, p = .010; 1994–2001: β = −.627, p = .008) suggest that royalty rates were generally lower in the past (compared with 2010 and later). Service type (β = .050, p = .665) and the other controls, brand strength (β = −.024, p = .614), retail sector (β = −.051, p = .725), and gastronomy sector (β = −.284, p = .137), are not significantly associated with the dependent variable. This basic model explains 15.9% of the variance in royalty rates. The VIF values indicate that multicollinearity is not a concern (VIF = 1.08–1.46). For parsimony reasons, we retain only the variables that are statistically significant (i.e., product-market profile and period).
Estimation Results.
*Statistically significant (p < .10).
Notes: The dependent variable is royalty rate (logit-transformed). Displayed coefficients are nonstandardized.
The results from Model 2 show a significant, negative effect of contract duration (β = −.185, p = .042) but no significant effect of territorial exclusivity (β = .031, p = .818). These results provide initial support for H5 but not for H4. The coefficients associated with product-market profile (β = .336, p = .013) and period (2002–2009: β = −.237, p = .032; 1994–2001: β = −.532, p = .007) remain largely the same. Model 2 explains 16.8% of the variance in royalty rates. Multicollinearity among the explanatory variables is not a concern (VIF = 1.04–1.10).
Model 3 represents the complete model we used to draw our conclusions. Regarding the effect of economic potential, the results provide evidence in support of the expected positive relationship to royalty rates (β = .324, p = .075); thus, H1 is supported. Likewise, stronger legal rights in the target market are associated with lower royalty rates (β = −1.606, p = .002); thus, H2 is supported. We also find the expected negative relationship of cultural distance with royalty rates (β = −.106, p = .047), in support of H3.
Regarding the effects of contract duration and territorial exclusivity, the coefficients are similar to those estimated in Model 2. Contract duration is associated with lower royalty rates (β = −.245, p = .006), and territorial exclusivity does not significantly affect royalty rates (β = .031, p = .821). Thus, we find support for H5 but not for H4. The effects of product-market profile (β = .389, p = .003) and period (2002–2009: β = −.279, p = .010; 1994–2001: β = −.627, p = .008) remain the same as in Models 1 and 2. The complete model explains 24.8% of the variance in royalty rates. We again paid explicit attention to possible multicollinearity: the VIF values associated with the explanatory variables range from 1.03 to 4.05, which is well below commonly suggested (more or less conservative) thresholds, such as 10 (Kleinbaum, Kupper, and Muller 1988) and 5 (Rogerson 2001).
Following the hypothesis testing, we explored the moderating roles of product-product profile (B2C vs. B2B) and service type (possession- vs. people-processing). For each contextual factor, we estimated a series of models, each including the respective factor's interaction with one of the focal determinants. With respect to product-market profile, as indicated by the previous main analysis (β = .389, p = .003), royalty rates tend to be higher in B2B settings (Mroyalty rate = 10.0%) than in B2C settings (Mroyalty rate = 6.3%). However, we find no significant interactions of product-market profile with any of the country or contract characteristics. These results suggest that the focal determinants play an equally important role in B2C and B2B markets (while the baseline rates may differ).
Regarding service type, we find that, in general, it is not significantly related to the level of royalty rates (β = .040, p = .703; i.e., royalty rates do not differ between people- and possession-processing services per se). The tested interactions suggest that service type contextualizes the role of certain country characteristics but has no influence on the impact of either of the two contract characteristics (βcontract exclusivity × service type = .026, p = .925; βcontract duration × service type = −.160, p = .378). Specifically, we find that for people-processing services, the positive effect of economic potential is more pronounced (βeconomic potential × service type = .493, p = .037) and the negative effect of legal rights protection is less pronounced (βlegal rights protection × service type = 1.212, p = .072). The negative relationship between cultural distance and royalty rates applies equally to both types of services (βcultural distance × service type = −.130, p = .225).
Additional Analyses
We carried out several additional checks to enhance confidence in our model's robustness. To assess whether our findings are robust to alternative measurement approaches, we estimated six additional models. We report the results of all six models in the Web Appendix. First, we tested the same model without applying the weighting (by the share of services’ added value in relation to GDP) and found a significant, positive relationship between economic potential and the level of royalty rates (β = .346, p = .076). All other effects remained largely the same (Model A-1). Next, we used total GDP instead of GDP per capita. We again observed a significant, positive relationship between economic potential and royalty rate (β = .224, p = .028), with no meaningful changes in all other estimates (Model A-2). Two additional models checked the robustness of our results to variations in our proxy of legal rights protection. We used two alternative rules for imputing missing values, one less conservative (−1 unit) and one more conservative (−2 units) than the initially applied rule (−1 SD). The results remain substantially unchanged (Models B-1 and B-2). Finally, we assessed the robustness of the effect of cultural distance. In line with recent best practice recommendations (Griffith, Dean, and Hoppner 2021), we ran the same model estimations using alternative measurements. First, we calculated the intercountry distances using Kogut and Singh's (1988) classic formula (for critiques, see Konara and Mohr [2019] and Messner [2021]). Using this alternative measurement approach, we still found a significant, negative effect of cultural distance on royalty rate (β = −.062, p = .050; Model C-1). Second, we used Schwartz’s (2012) cultural orientation scores instead of Hofstede’s (2001) national culture scores. With this alternative indicator, we still detected a significant, negative relationship between cultural distance and royalty rates (β = −.149, p = .063; Model C-2).
We also checked whether we could parse out the effects of individual cultural dimensions. For example, the differences in terms of power distance may be particularly consequential for contract negotiations and the perceived risk of moral hazard. However, we observed no significant effect when using a difference score that captures each franchisor–franchisee dyad's absolute difference in power distance. We conducted the same analysis for each cultural dimension and found the same null results in all but one case (uncertainty avoidance is negatively related to royalty rates; β = −.080, p = .081). These findings suggest that culture-driven moral hazard concerns result from global assessments rather than assessments of specific cultural attributes.
Furthermore, we explored whether the main effects potentially vary over time. To test this, we individually examined the interactions between each country and contract characteristics with both time dummies. We observed no significant variation in any of the effects, which leads to further confidence in the robustness of our findings.
Finally, we assessed whether our results might be biased due to the potential endogeneity of contract duration and territorial exclusivity. To address this issue, we chose a control function approach (Papies, Ebbes, and Van Heerde 2017; Petrin and Train 2010; Wooldridge 2015). First, we obtained the fitted residuals by regressing each contract variable on the country characteristics (contract duration: ordinary least squares regression; territorial exclusivity: probit regression) and added the fitted residuals to the second-stage equation. Next, we used the Hausman test to compare the endogeneity-corrected estimates against the original estimates. The nonsignificant test result (χ2 = .12, p = .726) suggests that endogeneity is not an issue and that the original model is more efficient (Papies, Ebbes, and Van Heerde 2017; Sande and Ghosh 2018).
Discussion
This research sheds light on the determinants of royalty rates in international franchising agreements. Franchising is an attractive business concept that enables franchisors to expand internationally quickly and with few resources, and it allows franchisees to benefit from the franchise's brand name value and proven operational procedures. However, compared with other market entry modes (e.g., wholly owned subsidiaries, joint ventures), international franchising has received less research attention (Robson et al. 2018). This research extends current knowledge about international franchising by focusing on a contractual provision that helps address the problem of double-sided moral hazard in such partnerships: the royalty rate. Specifically, we propose and test a set of country and contract characteristics that might affect the perceived risk of moral hazard and, in turn, lead to different royalty rates. Our findings contribute to the literature on international franchising and governance mechanisms in international interfirm relationships, and help managers understand which factors they should consider in international franchise contracting.
Theoretical Contributions
From a theoretical perspective, our study complements existing work by shifting the spotlight to the contracting stage. Previous work has illuminated why firms might choose franchising as a foreign market entry mode (e.g., Burton, Cross, and Rhodes 2000; Contractor and Kundu 1998a; Erramilli and Rao 1993), but the contractual conditions that franchisors and franchisees would willingly accept to enter such partnerships have received scant attention (Madanoglu, Alon, and Shoham 2017). The contracting stage is central to the process (Agrawal and Lal 1995) because neither franchisors’ benefits arising from entering a new market nor franchisees’ benefits arising from leveraging an established business concept materialize without a contract that aligns both partners’ interests through bilateral incentives and reduces their perceived risk of moral hazard.
The lack of literature on this issue is most likely due to the confidentiality of such business agreements, which poses a major obstacle for empirical research on international franchising (Lafontaine and Oxley 2004). As a consequence, previous attempts to explain contractual conditions in international franchising are either conceptual (Schmidt 1994) or restricted to a single country dyad with limited cross-national variation (Lafontaine and Oxley 2004). Therefore, our work represents an important extension of previous work toward generating empirically based generalizations that advance theory development regarding royalty rates in international franchising. Although Lafontaine and Oxley (2004) provided first insights into royalty rates across countries, their work offers limited insights into factors that explain the level of royalty rates.
Our findings offer empirical evidence supporting the relevance of a host market's economic, legal, and cultural characteristics. Whereas a host market's economic potential is associated with higher royalty rates, the strength of legal rights protection in that market and the cultural distance are associated with lower royalty rates. In line with our expectations, beneficial economic conditions in the host market, as manifested in, for example, high levels of disposable income, improve the business outlook. Higher royalty rates compensate for a franchisor's opportunity costs of not entering the market directly and are acceptable to the franchisee to ensure the franchisor's support in exploiting the market's economic potential. Strong legal rights protection decreases the risks and potential costs associated with contract violations and other events that might harm a franchisor's intellectual property, thus reducing its need to financially hedge against such risks through higher royalty rates. Finally, cultural distance is a barrier to interfirm trust and renders consumer acceptance of the foreign business format more uncertain. Paired with the need for significant initial investments, this uncertainty renders franchising in culturally distant markets risky for franchisees, which causes them to accept royalty rates that are lower than those for markets that are culturally closer to the franchise's country of origin.
We complement the macro perspective by also considering two major contract characteristics that might influence both parties’ perceived risk of moral hazard. First, related results suggest that contract duration, which can be interpreted as a sign of commitment (Brickley, Misra, and Van Horn 2006), leads to lower royalty rates. This finding is in line with Vásquez’s (2005) proposition that franchisors may alleviate franchisees’ concerns about withheld effort by offering longer contracts. By contrast, territorial exclusivity appears to have no significant impact on the level of royalty rates. One possible explanation for the lack of empirical support might be an opposing mechanism that cancels out the expected negative effect of territorial exclusivity rights. For example, it could be argued that granting a franchisee territorial exclusivity rights conflicts with a franchisor’s goal of revenue maximization, because it restricts the franchisor from authorizing new units of other franchisees. In this sense, the elimination of intrabrand competition may disproportionally benefit the franchisee (Schmidt 1994), prompting the franchisor to consider charging higher (not lower) royalty rates. However, the available data do not enable testing of this possible explanation. Our findings concerning the impact of contract duration and territorial exclusivity on royalty rates add to the body of research dealing with formal contracts as governing mechanisms of interfirm relationships in general (e.g., Griffith and Zhao 2015; Zheng et al. 2020) and franchising partnerships in particular (e.g., Kashyap, Antia, and Frazier 2012; Kashyap and Murtha 2017).
Furthermore, our results show that the relevance of some determinants depends on the type of service franchised. We introduce the distinction of people-processing versus possession-processing services to this literature stream. Our exploratory analyses reveal that the positive relationship between economic potential and royalty rates is more pronounced and the negative relationship between legal rights protection and royalty rates is less pronounced for services targeting people (e.g., hospitality, health, or entertainment services) than for services targeting their possessions (e.g., repair, storage, or financial services). The former contextual effect may be explained by how economic potential is typically measured: GDP-based measures might be appropriate indicators for people-processing services, as they often target mass markets, whereas possession-processing services might target more specialized, niche markets with specific needs, thus rendering broad economic indicators less accurate. The latter contextual effect might arise because services that involve high levels of human interactions tend to be more complex and thus are more difficult to codify and monitor. Therefore, contract violations, such as withheld efforts, are more difficult to detect and sanction in the first place, making enforcement of defined contractual rights generally more difficult, regardless of a host market's legal system, and renders the formal protection of legal rights relatively less important. Despite fundamental differences in many respects, we do not find significant differences in the importance of the studied determinants between B2C and B2B markets.
Managerial Implications
This research has direct implications for franchisors and franchisees intending to enter an international franchising agreement. In general, our conceptual model provides a basic framework for managers’ considerations of the problem of moral hazard in international franchising partnerships. It helps understand each party's perspective and possible concerns about the other partner's incentives (or lack thereof) to comply. Managers tasked with setting or accepting the terms of a contract should explicitly consider and assess these factors to support their decision in favor of or against a contractual offer.
Our model helps managers lay out the sources of moral hazard systematically and potentially underscores their assumptions with related data. For example, a prospective franchisee can assess the cultural distance between the host market and the franchisee's home market to garner more accurate expectations of consumer acceptance (which influences the royalty rate the franchisee is willing to accept). Likewise, a franchisor can gather intelligence on the host market's legal environments, such as information on the enforceability of contracts and the speed and costs of potential litigation. All these considerations can be supported with publicly available data.
As specific recommendations, our results suggest that franchisors can indeed charge higher royalty rates in economically promising markets, which tend to be of primary interest when expanding internationally (Walker and Etzel 1973). Although increased royalty rates are generally viewed as more acceptable for large markets with a strong service sector and high purchasing power, franchisor decisions should also encompass additional legal and cultural considerations. If meaningful differences in terms of legal rights strength exist, franchisors and franchisees should accommodate these differences by means of a risk-related markup or discount. For example, Sweden and the Netherlands are comparable in terms of economic potential (2021 GDP per capita: Sweden: U.S. $60,239; Netherlands: U.S. $58,061; World Bank 2022a) and cultural distance (Sweden: .460; Netherlands: .394; Messner 2021) relative to the United States, but they starkly differ in their level of legal rights protection (Sweden: 11; Netherlands: 2; World Bank 2022c). As a consequence, royalty rates should be lower in Sweden than in the Netherlands.
Likewise, the implications of the host market's cultural distance from the franchise's country of origin with respect to consumer acceptance of the business concept should be considered. Cultural difference can be considerable even within certain geographic regions. For example, the United Kingdom (.098) and Germany (.250) are relatively culturally close to the United States, whereas Portugal (.750) and Serbia (.738) are as distant from the United States as Malaysia (.746) and Indonesia (.758) are (Messner 2021). Similar levels of intraregional variance exist in Southeast Asia and Latin America, which highlights the need for country-specific assessments. For example, in two prospective foreign markets that are equally promising in economic terms and safe in legal terms—Canada (2021 GDP per capita = U.S. $52,051; 2019 Strength of Legal Rights Index score = 9; World Bank 2022a, 2022c) and Belgium (2021 GDP per capita: U.S. $51,768; 2019 Strength of Legal Rights Index score = 8; World Bank 2022a, 2022c)—prevalent cultural differences may still influence the royalty rates franchisors and franchisees deem appropriate. In this case, the larger cultural distance from Belgium (.392) than from Canada (.074) suggests the use of relatively lower royalty rates.
Finally, our findings related to the impact of contract characteristics on royalty rates suggest that longer contracts can lead to significantly reduced royalty rates, due to lower risks of opportunism and enhanced incentives to exert effort. Longer contracts foster interdependence and facilitate the strategic (co)development of the franchise and its brand equity, which benefits both the franchisor and the franchisee. However, the willingness and ability to commit to a long-term partnership require additional assessments, potentially taking the partner's reputation, interpersonal relationships, and competition-related factors into consideration.
Limitations and Future Research
This research has a few noteworthy limitations, mostly regarding data availability. The confidentiality of contractual agreements makes some data inherently difficult to obtain. Although the contractual data we carefully compiled and enriched with secondary data enabled us to demonstrate which country and contract characteristics drive royalty rates in international franchising, they suffer from two shortcomings that limited our ability to glean deeper insights.
First, the small number of observations limits the degrees of freedom available. More observations would have enabled us to control for and explore more of the unobserved heterogeneity. For example, we were able to distinguish several broad types of industries but could not include more specific industry dummies, which may have explained additional variance in royalty rates. Although assessing economic potential at the national level in terms of GDP per capita makes sense for many mass-market services, such as hospitality, health, and educational services, for some services that cater to other businesses rather than end consumers, industry-level characteristics, such as the development stage of the local industry and competitive intensity, may be more relevant to assess the economic potential of the endeavor. A finer-grained analysis would allow future research to disentangle these nuances.
Second, our data set is limited with respect to meso-level variables. Although we enriched the contractual data with additional information on the age of the franchised trademarks, as a proxy of brand strength, we found that it was unrelated to royalty rates. This result is somewhat surprising, given the intuitive notion that strong brands hold commercial value, thus allowing their owners to charge higher royalties (Jayachandran et al. 2013). Thus, it would be worthwhile to study other firm-level variables that might be more directly linked to the perceived risk of moral hazard, such as a partner's franchising expertise and reputation in an industry. To examine the role of firm characteristics in the contracting stage of international franchising, future research should consider trading breadth (single or few contracts with many companies) for depth (many contracts with a single or a few companies). Data sets with greater depth would enable researchers to study firms’ strategies in setting up royalty rates for different markets in a more granular way, and offer some potential to unravel the complex interdependencies among franchisors within the same industry. For example, the phenomenon of mimetic isomorphism (DiMaggio and Powell 1983) suggests that royalty rates might be influenced by the behavior of competitors, such that a franchisor might imitate existing “formulae.”
Third, our study focuses on territorial exclusivity and durations as two specific contractual provisions. However, previous research has also highlighted the importance of more global characteristics of contracts, such as their overall degree of specificity (Mooi and Ghosh 2010). Contract specificity refers to the level of “explicitness, specification and precision of a contract, such that a high level of contract specificity means that terms are more detailed and explicit relative to technical specifications of the product, implementation procedures, financial and legal considerations, and overall contract features” (Griffith and Zhao 2015, p. 23). Thus, contract specificity is closely related to other global contract features, such as contract completeness (Kashyap and Murtha 2017), complexity (Poppo and Zenger 2002), and flexibility (Wang, Ma, and Wang 2021). Investigating the influence of such global contract characteristics is worthwhile, as previous findings on interfirm relationships establish a link between contract specificity and the risk of contract violations (Griffith and Zhao 2015). Because the available data are limited to specific contractual terms, we cannot study the role of such global contract features. Due to the potential relevance of such features, we encourage future studies to extend our findings by also considering global contract characteristics as potential drivers of royalty rates in international franchising.
Finally, effective royalty rates can reduce the problem of moral hazard but not eliminate it. Therefore, other ex post governance mechanisms are indispensable. Interestingly, recent research investigating the interplay between ex ante contract design and ex post monitoring and enforcement (Kashyap and Murtha 2017) suggests that franchisees evaluate these mechanisms jointly as an “overarching governance constellation” (p. 146). Thus, future research should consider the potential impact of (ex ante) royalty rates on the effectiveness of complementary ex post governance mechanisms.
Supplemental Material
sj-pdf-1-jig-10.1177_1069031X221123265 - Supplemental material for What Drives Royalty Rates in International Franchising?
Supplemental material, sj-pdf-1-jig-10.1177_1069031X221123265 for What Drives Royalty Rates in International Franchising? by Jennifer Zeißler, Timo Mandler and Jeeyeon Kim in Journal of International Marketing
Footnotes
Special Issue Editors
Kelly Hewett, Cheryl Nakata, and Kay Peters
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
