Ansprechpartner: Dipl.-Kfm. Patrick Saßmannshausen, Research-Assistent: Arnd Hallbach & Christoph Flohr


SEAN PATRICK SASSMANNSHAUSEN, ARND HALLBACH & CHRISTOPH FLOHR
IGIF – Institut für Gründungs- und Innovationsforschung
University of Wuppertal, Germany

Contact to the authors: sassmannshausen@wiwi.uni-wuppertal.de

This paper deals with franchising. More precisely, it explores the process of becoming (or not becoming) a franchisee. It gives explanations and predictors relevant in this process. It also shows that, in the end, more potential franchisees decide to start up their own independent business than are willing to sign a franchise contract. Our findings offer strong theoretical and practical implications relevant to the franchise process and industry.



For empirical research, we used an internet based questionnaire. Persons who (in the past) had shown a clear interest in franchising were contacted by email, invited to visit a website, and to fill in a questionnaire. The form contained 33 questions. We questioned individuals who in the past had demanded time consuming advice from a consultancy specialised in franchising. First of all, the results show that only very few of them actually walk the walk after talking the talk. In our survey, we received n=800 answers out of approx. 18,500 individuals contacted via email. The sample disaggregates into three populations: 435 answers came from individuals who have not become franchisees yet, 156 individuals recently started a franchise outlet, and 208 started a new venture without being part of a franchise system. Therefore, according to our data, out of approx. 18,500 persons who were seriously interested in franchising only very few (extrapolated 1,396) finally took the plunge. In contrast, an extrapolated number of 5,200 started their own business outside the franchise industry. So why do potential franchisees prefer to remain employees or become new venture start-up entrepreneurs instead of becoming franchisees?



In a first step, the sample offers us the opportunity to explore differences among the three groups by using nonparametric tests (the three populations are: potential franchisees who never get started, actual franchisees who took the plunge, and new entrepreneurs who started their own venture independently). In addition to mere post descriptive rank sum calculations, we also conducted the Kruskal-Wallis Test, Rank Sum Wilcox Test for comparison, and Spearman’s Rank Correlations. All three methods apply for comparison of samples built by more than two groups, especially – but not only – if the normal distribution conditions are not met, so the t-test seems inappropriate. The Kruskal-Wallis Test detects if more than two populations are different. This is the case for the three groups examined in our paper. The Wilcox Rank Sum Test shows us, who – out of our three populations – are different and – even more important – how they are different. With Spearman’s Rank Correlation, we can relate their aptitude to their performance (performance in this case is understood as actively becoming franchisee, or “real” entrepreneur, or remain employee).



In a second step, our paper delivers an explanatory model. It answers the question why most individuals with an interest in franchising never take the plunge, whereas others become franchisees, or decide to become “real” entrepreneurs, i.e. start a new venture on their own. Therefore, we carry out three intermediate analyses. In a first step, after reducing the number of direct variables by factor analyses, we conduct a logistic regression model for the total sample of n=800 participants. The variable we want to predict is categorical (becoming a franchisee or not), and we cannot prove normal distribution for every single independent variable; moreover, our variables do not only assemble from binary coded scales but also include many different rank types of Likert scales. These three reasons make it necessary to use logistic regression instead of any other alternative. In a second step, we repeat the logistic regression model, but replace the predicted categorical variable by the two categories “starting a new venture” or “not starting a new venture”. In a third step, we build a structural equitation model (AMOS). The structural equitation model takes into account direct as well as latent variables. Financial capital available to the potential franchisee is an example of a directly measured variable, whereas risk taking attitude and social embeddedness are examples of latent variables. Both groups of variables concern and have been measured (respectively created) on three different levels of observation: individual, franchise-specifics, and resource level. Some variables prove to be of significant influence, but others do not. In each of the three steps taken, the results show that the outcome of franchise processes is not only driven by chance. There are some predictors that make it more likely for a person to become a franchisee or a “real” entrepreneur or to remain an employee respectively.



Our research results in some theoretical and practical implications. The theoretical implications point towards a revisiting of McClelland’s theory of motivation. By theoretical reasoning, franchisees – compared to entrepreneurs – can be expected to have approx. the same need for achievement, but less need for power and a higher need for affiliation. The theory perfectly adapts to our empirical findings. We therefore suggest it is relevant for franchising. It can help to explain why some individuals from our sample become franchisees, others real entrepreneurs, whereas some remain employees. Three practical implications can be derived from application of McClelland’s theory as well as from our empirical data. They concern franchise systems, franchise consultants and potential franchisees. Franchise systems spend a great portion of resources on the selection of new franchisees. Mistakes in the selection process can be costly. Our results can help to develop and to carry out a “franchisee applicant quick screen”. Thus, franchise systems can reduce their efforts in the selection process. Another implication is dedicated to business development in the area of consulting. Consulting firms specialising in franchise outlet start-ups, should consider offering entrepreneurial start-up consulting as an additional service, since in the long run, more of their clients become independent entrepreneurs than become franchisees. Keeping track of those customers (respectively business founders) and offering advice and advanced trainings to them (not only in the field of franchising but also in entrepreneurship and new venture creation) might be a business opportunity. The third lesson learned is for those who are interested in becoming a franchisee, the potential franchisees. The results give information on some of the influencing factors on the franchise decision. Potential franchisees can appraise more accurately what it takes to take the plunge, and therefore our model can lead to a better self-evaluation at the beginning of the search process. This can save time and efforts taken to find an appropriate franchise system. Nevertheless, our results do not only answer given questions, they also open up space for new questions. Thus, the paper ends with suggestions for further research in this particular area of interest.



Literature:

Kaufmann, Patrick J. (1999): Franchising and the Choice of Self-Employment. Journal of Business Venturing, Vol. 14, pp. 345-362.