Franchise Contracts
3.6 Some Final Thoughts and Comments
In this chapter, we have described the main monetary contract terms found in franchise contracts in the U.S. In particular, we have shown the type and level of fees collected by franchisors, the variance of these fees, and how the fees have evolved since 1980. Before moving on to discuss dual distribution and vertical restraints in Chapter4, a few final points should be made in relation to these fees. First, using our data but only up to 1992, Lafontaine and Shaw (1999) have shown that while franchise fees and royalty rates vary across fran-chisors, they remain quite stable over time within franchised chains. In fact, franchisors do not systematically increase or decrease their royalty rates or franchise fees as they become better established, whether this is measured
83A few franchise contract clauses, such as choice of law or mandatory arbitration clauses, relate to interpretation and enforcement rather than defining the franchise relationship itself or its economic underpinnings. Readers interested in learning more about such contract clauses, or about franchise law, would do well to consult a legal treatise such as Garner (2002). For an overview of laws that affect franchising in a number of different countries, see Zeidman (1989) and, more recently, Mendelsohn (2003).
in number of years in franchising or in terms of total outlets. The ten-dency instead is to keep the fees, especially the percentage of sales fees, rel-atively constant. Thus, the uniformity in fees described earlier across fran-chisees within a chain, and across borders, also applies over time within a chain.
Second, we have explored somewhat in this chapter the relationship be-tween royalty and advertising rates on the one hand, and ongoing fixed pay-ments on the other hand. Sometimes franchisors use both sales-based per-centage fees and ongoing fixed payments. But in other cases, franchisors rely on fixed payments in lieu of a percentage royalty or advertising fee, in which case the use of fixed payments and percentage fees is negatively correlated.
The theoretical literature on franchising implies that sales-based fees and ini-tial franchise fees should be negatively correlated everything else being the same and, in particular, if the value of the franchise is held constant. After all, if a franchisee is asked to pay more in initial fees, this must be made up by a reduction in ongoing payments later on and vice versa. Empirically, this negative relationship has been quite elusive. For example, Lafontaine (1992a) and Sen (1993) found no negative relationship between the sum of royalty and advertising rates and the initial franchise fee for the sets of U.S. franchisors they studied, even after controlling for variables that proxy the value of the franchise. More recently, Kaufmann and Dant (2001) even find a positive relationship between royalty rates and franchise fees after controlling for unit sales. One could argue that the lack of observed negative relationship arises from differences in contract terms that affect both fees in similar ways and are not controlled for in these analyses. Along these lines, Sen (1993) finds that both fees decline if the franchisor imposes input purchase requirements on its franchisees, a requirement that allows the franchisor to collect markup dollars from their franchisees as discussed earlier. Any correlation analysis for roy-alty rates and franchise fees that would not account for such a clause might yield a positive or zero correlation simply because differences in the input purchase requirements make franchisors adjust both fees similarly. If firms do not modify their other contracting practices frequently, however, one can account for them possibly by examining the relationship between changes in the fees of individual chains over time. Lafontaine and Shaw (1999) perform this type of test and again find no relationship between total percentage fees and initial franchise fees. Of course, it may be that not all relevant factors are held constant in these analyses either. At this point, however, the empirical evidence suggests that franchise fees are set quite independently from sales-based fees. The absence of a negative relationship between the two types of fees implies that the franchise fee may not be set at a level to extract all of
the economic rent that franchisees earn.84This would occur, for example, if franchisors mostly set their initial franchise fee to compensate for the costs they incur in setting up the outlet, as Bond (2001) suggests they do.
Finally, although fewer data are available on contract terms used by non-U.S. franchisors, studies of franchise practices outside the non-U.S. suggest that the basic types of fees, the extent to which they are used, and their levels are fairly similar to those described here for the U.S. Canadian franchisors included in our data, for example, use the same types of fees and set them at levels that are quite similar to those chosen by U.S. franchisors. P´enard, Raynaud and Saussier (2003) examined the fees used by more than 200 franchisors in France, and found that most of them also use percentage of sales royalties and initial franchise fees, though the latter appear to be slightly lower in France.85Frazer (1998) shows that while the same types and levels of fees are used by franchisors in Australia, there is a greater reliance on ongoing fixed payments as 17 percent of the firms in her sample of 262 franchisors rely on such payments compared to just 7 percent of franchisors doing the same in our data. Finally, Seaton’s (2003) data show that the fees of 161 franchisors in the UK in 1998 are very similar on average, and follow the same general distribution, as the ones found here for U.S. and Canadian franchisors.86 Moreover, the conclusions about the lack of correlation between the fees and their stability over time also apply to non-U.S. franchisors. Specifically, Gagn´e et al. (1998) establish a lack of negative correlation in fees for franchisors operating in Qu´ebec, Canada, while Seaton’s (2003) analyses indicate that the fees used by franchisors in the UK also are not negatively correlated and do not change much over time either.
84See Kaufmann and Lafontaine (1994) and Michael and Moore (1995) for evidence that franchisees earn economic profit at least in some franchised systems and for discus-sion of the reasons why franchisors many choose to leave profit with their franchisees.
Lafontaine and Raynaud (2002) argue that downstream profit (combined with moni-toring and termination rights) and residual claims are complementary incentive mech-anisms in franchised chains, i.e., they are used to reinforce one another.
85The authors do not mention advertising fees, but a report from the F´ed´eration Franc¸aise de la Franchise (French Federation of Franchising) (2003) contains data on such fees.
86Many other descriptive studies also have found similar fees in other countries.