Paper to be presented at theEMNet-Conference on
"Economics and Management of Franchising Networks"
Vienna, Austria, June 26 – 28, 2003
Successful Franchising Using the Plural Form
Thomas Ehrmann & Georg Spranger
Center for Entrepreneurship
University of Muenster
Leonardo-Campus 18
D - 48149 Muenster
Germany
Tel.: +49 (0) 251 83 38330
Fax : +49 (0) 251 83 38333
Web:
Mail:
Abstract:
Many successful franchise companies consist - other than their name might suggest - of more than just the system’s headquarter and its associated franchisees. Besides independent franchise outlets, the majority of chains operates a significant number of company-owned stores run by employed store managers. The evolution and constant existence of such a hybrid organizational form cannot be explained under a pure principal-agent-view and deserves further analysis. As this article explores, plurally organized franchisors may benefit from effects of cost reduction, quality enhancement, growth stimulation and optimized risk control in contrast to pure franchise chains. Within our sample of 240 franchise systems we found strong empirical support for the quality argument of the Plural Form. Compared to pure franchise companies plurally organized chains are much better in sending out positive signals to an external community containing internal information and thus reducing costly agent uncertainty. By running company owned stores as well as independent franchise outlets, chains force themselves successfully into cooperational and less opportunistic behavior towards their franchisees.
We gratefully acknowledge financial support from the Konrad-Adenauer-Foundation and the University of Muenster Graduate Fund for this research.
1
I. Introduction
In principle, the manufacturers of a product or the providers of a service (as principals) have the choice between numerous solutions regarding the organizational design of their distribution channels. The extreme positions are marked by a company-owned distribution system on the one hand and a distribution through independent external subjects on the other. Within this broad spectrum, franchising takes up a middle position as principals interact with vastly independent franchisees on a fixed contract basis. The choice between company-owned and independent distribution elements depends on the contractual appropriateness of the solution as well as on the visibility of certain in- and output variables. Thus it is of great importance for the franchisor as the principal, whether he is able to control the agent’s individual input and whether the agent himself is able to influence his output significantly or not by changing input variables such as his work effort. As the agent’s influence on personal output increases and the principal is able to observe the agent’s specific input, company-owned stores gain ground over franchisees (and vice-versa). Is it either easier for the franchisor or his only option to control the agents output, for instance through observing his revenues, franchising becomes the better distribution mode (Lafontaine, F./Slade, M. (1998)). Sometimes the agent’s influence on the store’s output by increasing individual input is very small, as for instance in supermarkets. In cases like this, franchisors will rarely decide to franchise outlets (Maness, R. (1986)) and put the control of distribution in the hands of company-owned outlets. However the specific situation, in a world of homogeneous markets there should exist just one contractual choice for all distribution outlets in the long run. Concerning franchise chains, one should, in the long run, therefore observe either a shift towards wholly franchised or wholly company-owned systems. Hence the coexistence of franchised and company-owned outlets within one chain – which is defined as dual or plural form ((Bradach (1997), Gallini/Lutz (1992), Lewin-Solomons (1998)) – and its use by a majority of systems cannot be explained from a pure principal-agent-view and deserves further analysis (Ehrmann, T. (2002)). Doing this our paper starts with an examination of the basic characteristics of the plural form and an overview over the organizational settings of 240 large franchise chains (II). Then we specify several theories that try to explain the coexistence we find in reality and, afterwards, derive hypotheses concerning the advantages of plurally organized over pure franchise firms (III). These hypotheses are then tested along our sample of 240 systems (IV). The paper closes with an interpretation of our empirical results (V).
II. The Plural Organizational Structure of Franchise Chains
A. The Origin of the Plural Form
Hybrid firm structures are a combination of distinct parameters of firm organization and are no new phenomena in organization science. Generally speaking, hybrid structures combine a mixture of the organizational methods “price” and “hierarchy” with the organizational institutions “markets” and “firms” (Ouchi, W. (1980)). The various ways of possible matches result in a broad spectrum of potential organizational constellations. We talk about the plural form as soon as we observe a side by side mix of price and hierarchy within one single firm (Hennart, J. (1993)). Cases where two different organizational methods exist within on institution have to be distinguished from firms that present one organizational form as a mixture of both prices and hierarchy. As we know, in reality many companies employ a sales force as part of their sales department and manage independent salesmen as freelancers simultaneously. Others, for example companies in the automotive sector, produce a pre- or final-product by themselves and, at the same time, buy the same part from external suppliers. The case that gained our research interest is where franchise chains employ fixed-paid managers to run company-owned outlets as well as vastly independent franchisees operating franchised stores. Whereas the first ones can be controlled by the franchisor through hierarchical relationships, the latter ones, bearing a rather great entrepreneurial risk by investing money in a chain, are more likely to be governed through monetary incentives than through authority. Within this complex organization the employed managers represent the hierarchical branch, whereas the independent and risk-bearing franchisees belong to the price-steered fraction of inter-firm subjects.
It is crucial for our further analysis to distinguish between three possible developments leading to such a plural form. First, plural forms can be the results of a continuous process of selecting the individually best method for the current situation of the firm. Thus we can speak of a case-by-case decision between the best fit of either the price or the hierarchy method for a given situation and environment.[1] Second, from the perspective of the entire firm, a mixture of both methods can be appropriate to create an organization that contains “the best of both worlds” in one structure.[2] We will come across this approach again when looking at the plural form’s function of actively managing risk exposures. And third, some researchers argue that the plural form arises due to the intention of realizing certain synergies that would not exist if just one method was used without the other.[3] All three options become more visible when we will discuss the arguments for the existence of the plural form in part III.
B. The Design of the Plural Form in Reality
As a basis to analyze the organization of franchise chains in reality, we used the data of 240 franchised chains, published in the yearly Franchise500 report by the Entrepreneur Magazine. A self-restriction to 240 systems minimizes - from our point of view - the downward distortion of empirical figures, which results from the rapidly decreasing system size of companies with lower rankings. This effect seems to be detrimental in earlier studies by Lafontaine/Shaw (2001)andby Pénard/Raynaud/Saussier (2002), where much larger samples of 4842 and 581 franchise chains are analyzed. The high degree of heterogeneity already inherent in our sample of 240 chains is shown in Table 1 below. Particularly we compare companies like McDonald’s, founded in 1955 and with 28.236 outlets in 2001 the largest chain of the sample, with systems like Purified-Water-to-Go, ranked number 190, founded 40 years later in 1994 and representing the smallest system of our sample with just 51 outlets total in 2001. In order to analyze a franchise chains organizational constitution, we compute the percentage of plural organization () as the number of franchised units out of the number of all outlets together. As the mean degree of plurality of our sample we calculate 91%, accompanied by a standard deviation of 17%. Whereas for instance McDonald’s is truly plurally organized with a share of 25% of company-owned outlets, Purified-Water-to-Go as another example does not have any company-owned outlets at all.
Parameter / Sample Size /Mean
/ Std. Deviation / Minimum / MaximumTotal Outlets / 240 / 1.297 / 3.205 / 51 / 28.236
Number of franchised units / 240 / 1.159 / 2.849 / 41 / 24.695
Number of company units / 240 / 138 / 639 / 0 / 6.918
Percent franchised / 240 / 91 % / 17 % / 15 % / 100 %
Royalty rate in % / 215 / 6 % / 3 % / 0 % / 20 %
Franchise fee in T$ / 235 / 27 / 17 / 0 / 130
Minimum investment in T$ / 240 / 525 / 2.084 / 1 / 22.000
Maximum investment in T$ / 240 / 1.202 / 4.551 / 13 / 44.800
Years in business / 240 / 28 / 19 / 5 / 137
Years of franchise experience / 240 / 20 / 12 / 4 / 69
Corresponding results are presented by the two studies already mentioned. Pénard/Raynaud/Saussier (2002) extract the following means and standard deviations (in brackets) out of a sample of 581 French franchise chains: number of all outlets 74 (158), percentage franchised: 67 (68), royalty rate in percent: 4,4 (3,9), franchise fee in T$: approx. 15 (13), years in business: 24 (29), years of franchise experience: 14 (12). Lafontaine/Shaw (2001) analyzed 4842 US-American franchise systems and present the following means and standard deviations: number of all outlets: 207 (788), percentage franchised: 78 (71), years in business: 17 (15), years of franchise experience: 11 (10). As we stated earlier we believe that the results of both studies are distorted downwards because of the rapid decrease of firm size in larger samples. Nevertheless the figures are sufficient to get an idea of franchise chain organizational design. According to the three studies one can expect franchise chains to be plurally organized in a range between 91% (Ehrmann/Spranger) over 78% (Lafontaine/Shaw) up to 67% (Pénard/Raynaud/Saussier). Thus we expect plural systems to have a realistic share of company-owned outlets of about 9-33%.
Completing this rather static analysis we also had a look at the organizational development of our sample over the period observed, thus adding a dynamic component to our plural form arguments. As displayed in Figure 1 below we find a relatively dense concentration of systems with a high , i.e. systems, that have a rather small share of company units against a large proportion of franchised outlets. In 1997 83% and in 2001 85% of our sample firms show a of at least 80%, which seems to serve as a lower limit concerning the organizational constellations. Both in 1997 as well as in 2001 about 37% of our sample represent pure franchise chains without any company-owned units. A significant majority of 63% is more or less plurally organized throughout the 5-year period. Additionally a clear trend towards the -limit of 80% is observable for those companies holding vastly more company-owned units prior to and in the beginning of 1997.
Summarizing the findings above we conclude that the franchise systems observed reflect a broad variety of different organizational solutions. Taking a similar business strategy for all systems into account – for instance the maximization of revenue and profit –, it becomes obvious that the motives for organizing the systems are manifold and ambiguous. In the ongoing paper we first want to explore the different theoretical arguments that are put forth by researchers to justify the existence of the plural form. As will be shown, plurally organized chains may realize advantages concerning costs, quality, growth and risk control over singular organized pure franchise chains. In a second step, we test these apparent benefits on our sample and attempt to explore the real reason(s) why the plural form is as widespread as it is.
III. Theoretical Arguments for the Existence of the Plural Form
A. Cost Benefits
When a principal (i.e. the franchisor) contracts with an agent (i.e. the franchisee or the company-unit manager), costs arise as soon as one partner feels that crucial information is distributed asymmetrical between the two parties. Depending on whether such an uncertainty based on limited information exists before, while or after the signing of a contract, one has to deal with uncertainty of the contract’s quality, with uncertainty concerning the other party’s behavior or uncertainty in regard of the expected outcome of the contract. While a franchisor controls the quality of his agent ex-ante via intense selection processes and the outcome of the contract by controlling the agent’s output ex-post (for instance by checking his revenues), it is difficult to observe and control the agent’s true behavior while he himself enforces the contract according to his own will. For reasons of keeping up system quality and protecting the value of the trademark, it is absolutely crucial for the franchisor to exert control over the agent’s behavior, insuring that he follows the rules set out in the contract. One efficient option for the franchisor to reduce this kind of costly uncertainty concerning the agent’s behavior, may lie in choosing the right organizational structure.
While doing this, the principal has to consider the costs for controlling the agent on the one hand and the costs arising through inefficient investment and through cheating against the system and free-riding on behalf of other system members on the other. Managers of company-owned units are generally employed on a fixed-wage basis and thus have little influence on raising their salary (i.e. their output) by increasing personal work input. Thus an opportunistic company-unit manager may tend to maximize his input-output-balance by reducing work input while maintaining his output i.e. his position and salary. To prevent this kind of shirking the franchisor will have to monitor the manager closely and punish misbehavior either by ending the contract or by reducing the agent’s salaries.
When investing into a franchise unit, franchisees on the other hand accept a significant risk for the chance to improve personal income by influencing the outlets revenues and costs through increased personal work effort. The franchisees therefore will rarely have to be controlled on shirking tendencies, which keeps monitoring costs relatively low. Risky and costly for the franchisor is the Franchisees’ tendencies to maximize their income-output-balance by improving financial output and not by minimizing their work input. Thus a franchisee tends to exert actions that may quickly improve his personal output but may hurt the franchisor or other franchisees and thus weaken the entire chain and it’s trademark value in the long run. Reducing costs by lowering the offered quality or offering discounts unapproved by the franchisor are just two possible actions that franchisees could exert without any control by the franchisor. In other words, the franchisee is able to maximize his income by free-riding on the system’s trademark and quality and thus by cheating on other system members and hurting the chain as a whole. Franchisees may also try to avoid certain measures ordered by the franchisor for the sake of the entire chain, that drive up the franchisee’s costs without paying him an instant reward, rather benefiting the entire chain and the common trademark. From the perspective of the franchisor, a resistance to the execution of measures benefiting the entire chain results in avoidable opportunity costs for necessary improvements not carried out. In contrast to the managers of company-units, franchisees cause costs for the franchisor because of free-riding on the trademark’s value and because of resistance to necessary investments. Whereas pure franchise chains depend solely on one organizational element and have to bear the costs coming along with it, plurally organized systems have the choice between franchised and company-owned units and thus between two different sources of costs. For every location they can consider the costs of monitoring against those of free-riding and inefficient investment and thus go for the individual best solution. This idea has been manly promoted by Brickley/Dark (1987). Thus it is discussable, whether franchisors use the plural form to minimize the agency costs arising from behavior uncertainty. As will be explained in part IV, in case this hypotheses should earn support, there should be empirical evidence of ownership redirection tendencies among the older and more matured franchise systems.
Besides agency costs, franchisors gather knowledge about local markets in order to maximize profits from their trademarks. Although the acquisition of such information is quite costly, the franchisor needs to know local market preferences in order to adjust his decentral production and distribution facilities under the restrictions of systemwide uniformity and standardization. As Minkler (1992) has stated, the need and costs to acquire such knowledge are likely to grow with increasing unfamiliarity, heterogeneity and volatility of local markets as well as with the decreasing half-life of the information collected.
The franchisee as the receiver of the residual income (i.e. the difference between his revenues and costs) will have a much greater motivation to collect and use local information in order to raise his personal output than the fixed-paid manager of a company-owned unit. As an alternative to the information gathering franchisee, the franchisor would first have to engage in costly market research himself and then, secondly, would have to force and to control his company-units in using the information and maximizing the chains profits. Besides the cost component, franchisors would have a problem identifying and collecting the kind of implicit knowledge that franchisees win and collect through experiments and experiences. Even if a franchisor could manage to do this, he would still lag behind the franchisees and loose information when implementing the knowledge within the company units. Thus for gathering and using local information, franchisees will always be one step ahead and will succeed over employed unit managers in maximizing profits out of the local knowledge.
Having this in mind Minkler (1992) has demonstrated that plurally organized franchise chains optimize their search costs for local information by setting up franchise units where the acquisition of local knowledge is necessary and costly and where their use has a great impact on the outlets output. At the same time, the costs conscious franchisor who sets up company-owned units in the direct neighborhood of these franchisees, is able to transfer this knowledge to his own outlets and benefits from the franchisees’ initiative and commitment. According to Minkler's (1992) analysis, franchisors should develop new business areas by first installing franchise units exploring the new local market.