An Extended Transaction Cost Model of Decision Rights Allocation The Case of Franchising

Nada Mumdziev Center for Business Studies University of Vienna Brunner Strasse 72 A-1210 Vienna, Austria Email: [email protected]

Josef Windsperger Center for Business Studies University of Vienna Brunner Strasse 72 A-1210 Vienna, Austria Tel: +43 1 4277 38180, Fax: +43 1 4277 38174 Email: [email protected]

1 Abstract The purpose of this paper is to develop an extended transaction cost approach of decision rights allocation in franchising networks. Specifically, this study demonstrates that when considering trust in transaction cost theory,

(TCT) supplements the explanation offered by the TCT on the allocation of decision rights in franchising networks.

We found that transaction-specific investments have a positive effect on the allocation of decision rights to franchisees. These findings imply that franchisors tend to delegate decision rights to franchisees when they are able to reduce formal control, due to the bonding effect of franchisees’ transaction-specific investments. Contrary to the traditional transaction cost view and compatible with the incentive view of delegation, we found that behavioural uncertainty has a positive effect on the allocation of decision rights to franchisees. These findings imply that franchisors tend to delegate decision rights to franchisees when they encounter difficulties in measuring franchisees’ performance and controlling their behaviour. Furthermore, we found that trust functions as moderator in the relationships between the transaction cost variables and franchisor’s propensity to delegate the decision making power to the franchisees. Overall, our study contributes to the literature by constructing and testing an extended transaction cost model to explain the structure of decision rights in franchising networks.

2 1. Introduction

Successful franchise firms govern their contractual relations by efficiently allocating decision rights between the franchisor and franchisees. Decision rights refer to the use of both the franchisor’s system-specific assets, such as knowledge and skills in site selection, store layout, product and brand development, buying and merchandising, and the franchisees’ local market assets, such as their local market know how in advertising, customer service, quality control, human resource management and product management (Windsperger 2004; Mumdziev and Windsperger 2011). This study presents an extended transaction cost analysis on the allocation of decision rights in franchising networks by considering trust as moderator on the relationship between the transaction costs variables and franchisee’s fraction of decision rights.

Trust can be defined as the expectation that an exchange partner will not engage in opportunistic behavior, even in the face of tempting short-term incentives (e.g. Bradach and

Eccles 1989; Mayer et al. 1995). Traditional transaction cost literature has neglected the impact of trust on inter organisational cooperation (e. g. Williamson 1975, 1985) as it primarily focuses on transaction cost effects of environmental uncertainty, behavioural uncertainty and transaction- specific investments. However, Williamson acknowledged in a later study (1991) that trust functions as a shift parameter by influencing the comparative cost of governance. Another justification for incorporating trust into the TC model comes from the alliance literature, which shows that trust influences the governance of inter-organisational relationships (e. g. Bradach and Eccles 1989; Gulati 1995; Zaheer and Venkatraman 1995; Noteboom et al. 1997; Alvarez et al. 2003; Lui and Ngo 2004; Poppo and Zenger 2002; Gulati and Nickerson 2008; Lazzarini et al. 2008). The results of this work indicate that trust lowers transaction costs in inter- organisational relationships, especially when the opportunism risk is high.

The role of trust as a moderator has received less attention in the inter-organisational network literature. Recently, Mellewigt et al. (2007) analyse the moderating role of trust in the

3 relationship between asset specificity and contractual complexity. They conclude that trust mitigates opportunism risk and reduces contractual complexity as a control device. Ryu et al.

(2008) test if trust plays a moderating role in the relationship between environmental uncertainty and the propensity for vertical control in the buyer-supplier relationship. The results confirm that firms tend to loosen vertical control when they trust their exchange partners. Furthermore,

Hoffmann et al. (2010) analyse the role of trust in the firm’s decisions to vertically integrate or cooperate. Results confirm both an opportunism-mitigating effect of trust that lowers transaction costs, and an opportunism independent effect that increases the transaction value of the cooperation.

Despite the large number of studies on trust in alliances, few studies investigate the role of trust within the franchising context (Dickey et al. 2007; Cochet et al. 2008; Lopez-Fernandez and Lopez-Bayon, 2011; Davies et al. 2011). Dickey et al (2007) and investigate the influence of trust on franchisees’ behavior and attitude toward franchisor. They argue that trust will be developed as a mechanism of reducing franchisees’ opportunistic behavior in areas that are not covered by the contract. Cochet et al. (2008) analyse franchisors’ reliance on relational governance mechanisms (such as trust) to attenuate agency problems which arise from franchisee autonomy. They found evidence that franchise firms use relational governance to counterbalance their loss of control associated with allocation of decision autonomy to individual franchisees. Lopez-Fernandez and Lopez-Bayon (2011) found that trust arising from past relationships has a positive effect on delegation of decision rights. Davies et al. (2011) construct and test a relational exchange model that demonstrates how trust affects franchisee compliance.

They show that trust reduces non-compliant opportunistic behaviour. Similar to Dickey et al.

(2007), they show that trust reduces the likelihood of franchisees’ non-compliance with franchisor’s operational guidelines.

4 In this study, we propose an extended transaction cost model of decision rights allocation in franchising. According to the transaction cost theory, the allocation of decision rights aims at reducing transaction costs, due to transactional uncertainty and transaction-specific investments.

We demonstrate that considering trust in the TCT model supplements the explanation offered by

TCT on the allocation of decision rights in franchising networks. Trust influences the impact of transaction cost variables (such as uncertainty and transaction-specific investments) on the delegation of decision rights, because it alleviates opportunism risk and increases information sharing between the franchisor and franchisees. Specifically, we argue that trust moderates the impact of behavioural uncertainty, environmental uncertainty and transaction-specific investments on the delegation of decision rights to franchisees. We found that behavioural uncertainty and transaction-specific investments have a positive effect on the allocation of decision rights to franchisees. These findings imply that franchisors tend to delegate decision rights to franchisees when they encounter difficulties in measuring franchisees’ performance and behaviour monitoring. This also is likely to occur when they are able to reduce formal control, due to the self-enforcing effect of transaction-specific investments. In addition, we found that trust increases the impact of environmental uncertainty and transaction-specific investments on the allocation of decision rights to franchisees and decreases the positive relationship between behavioral uncertainty and delegation of decision rights.

This study contributes to the literature in three important ways. Firstly, we extend the franchise literature that deals with decision rights in franchising (Arrunada et al. 2001; 2005;

Windsperger 2004; Azevedo 2009; Mumdziev and Windsperger 2011; Lopez-Fernandez, Lopez-

Bayon 2011) by developing a transaction cost explanation of the allocation of decision rights.

Secondly, we extend the literature which deals with trust in franchising (Dickey et al. 2007;

Cochet et al. 2008; Lopez-Fernandez and Lopez-Bayon, 2011; Davies et al. 2011) by analyzing the impact of trust on decision rights allocation between franchise partners. Thirdly, this study extends the transaction costs literature by testing the role of trust as a moderator between the

5 transaction costs variables (uncertainty and transaction-specific investments) and the franchisors’ propensity to allocate decision rights to the franchisees. Adding trust as a moderator has considerably increased the explanatory power of the TC model. Overall, we extend the TCT by investigating the moderating role of trust in the relationships between the transaction cost variables and the franchisor’s propensity to delegate decision rights to the franchisees. This provides new insight by demonstrating that considering trust in TCT supplements the explanation offered by TCT on the allocation of decision rights in franchising networks.

The paper is organised as follows: Section two presents the extended transaction cost model of decision rights allocation in franchising. Section three presents the results of the empirical analysis. In section four we summarise and discuss the results, and in section five we draw some conclusions.

2. Research Model and Hypotheses

Our research model consists of three transaction costs determinants: behavioural uncertainty, environmental uncertainty and transaction-specific investments. We test their effect on the decision rights allocation to franchisees as a dependent variable. Furthermore, we add trust in the model. Specifically, we hypothesise that trust moderates the impact of behavioural uncertainty, environmental uncertainty and transaction-specific investments with regard to the delegation of decision rights (see Figure 1).

Insert Figure 1

2.1 Behavioural uncertainty, trust and decision rights

6 Transaction cost theory views behavioural uncertainty as arising from the difficulties of accurately monitoring the contractual performance of exchange partners (Williamson 1985).

Anderson and Gatignon (1986) define this concept as internal uncertainty, which exists when the firm cannot accurately measure agent’s performance. In the application of a TC model to explain firms’ make-or-cooperate decision, Hoffmann et al. (2010) find that higher performance measurement difficulties significantly increase the firms’ tendency to vertically integrate.

Behavioural uncertainty increases the risk of opportunism that arises in the form of various dishonest and detrimental behaviour, such as cheating, shirking or distortion of information. In order to reduce the risk of franchisee’s opportunistic behaviour, the franchisor exercises greater control over the transaction partner’s activities (Williamson 1985; Gatignon and Anderson

1988). One way is to impose stronger control by centralising decision making. According to this traditional transaction cost view (Williamson 1975; Hennart 1991), we expect a negative relationship between behavioural uncertainty and franchisees’ portion of decision rights.

Therefore, we formulate the following hypothesis:

H1: Franchisees’ portion of residual decision rights is negatively related with franchisors’

perception of behavioural uncertainty.

If a franchisor perceives the franchisee as trustworthy, this will influence his perception of behavioural uncertainty and hence its propensity to use formal control mechanisms. Trustworthy franchisees need to be less monitored under high difficulties to measure the performance of the network partners. As a result, the higher trust as informal control device, the lower the franchisor’s necessity for hierarchical control by centralising decision making in the case of performance measurement difficulties becomes. Consequently, we can derive the following hypothesis:

7 H1a: The relationship between behavioural uncertainty and franchisee’s portion of

decision rights is less negative under high trust.

2.2 Environmental uncertainty, trust and decision rights

According to the control view of governance (Williamson 1975), firms will increase their information-processing capacity if the coordination requirements increase with environmental uncertainty. Stinchcombe (1990) asserts that organisations implement more elements of hierarchy when the degree of uncertainty increases. Similarly, Noordewier et al. (1990) show that environmental uncertainty is positively related with firm’s level of control. Therefore, when franchisors perceive increased environmental uncertainty (such as market volatility and low predictability of sales volume), they intend to increase control over operational decisions at the local markets. We subsequently formulate the following hypothesis:

H2: Franchisees’ portion of residual decision rights is negatively related with

environmental uncertainty.

In high-trust relationships that facilitate information exchange and reduce opportunistic behaviour between the partners, franchisors will rely more on delegation of decision rights to franchisees when environmental uncertainty increases. Therefore, the negative effect of environmental uncertainty on the allocation of decision rights is expected to be lower if franchisors trust their franchise partners. We formulate the following hypothesis:

H2b: The relationship between environmental uncertainty and the franchisee’s portion of

the decision rights is less negative under high trust.

2. 3 Transaction-specific investments, trust and decision rights

8 According to the transaction cost theory, transaction-specific investments increase the partners’ quasi-rents that can be expropriated by the less dependent partner (Williamson 1985;

Klein 2000). When the transaction-specific investments of the franchisees are high, their quasi- rents are likely to exceed the potential hold-up gains from opportunistic behaviour. This bonding effect increases the self-enforcing range of contracts (Klein 1995, 1996). In this situation, the hostage effect of transaction-specific investments motivates the franchisees to behave cooperatively in order to realise the relationship-specific quasi-rents (Williamson 1983; Katz

2008).

This incentive effect of transaction-specific investments may influence the allocation of residual decision rights in the following way: Based on the traditional TC-view, decision rights are control devices to mitigate opportunism risk. The higher the opportunism risk for the franchisor, the higher is the degree of control by the headquarters. On the other hand, high franchisees’ transaction-specific investments result in lower opportunism risk, thereby requiring a lower level of control by the franchisor. Therefore, the higher the franchisees’ specific investments, the lower their motivation to behave opportunistically becomes, and the lower the franchisor’s need to monitor and control the franchisees by centralisation of decision making becomes, and hence the more residual decision rights are transferred to the franchisees. We can derive the following hypothesis:

H3: Franchisees’ portion of residual decision rights is positively related with

franchisees’ transaction-specific investments.

Since transaction-specific investments result in high quasi-rents, which in turn increase the franchisee’s motivation to cooperate, higher trust increases the quasi-rent-generating effect of transaction-specific investments. Through this, the self-enforcing range of contract increases

(Hwang 2006). In this case, the franchisor reduces hierarchical control by transferring a higher

9 fraction of decision rights to the franchisees, compared to a situation with a lower level of trust.

Therefore, we expect that trust will positively interact with the effect of the transaction-specific investments on the allocation of decision rights to franchisees. In other words, the positive effect of transaction-specific investments on the allocation of decision rights to the franchisees will be stronger in a high-trust situation. We formulate following hypothesis:

H3a: The relationship between transaction-specific investments and franchisee’s portion

of the decision rights is more positive under high trust.

3. Empirical Analysis

The empirical data for this study was collected via questionnaire from German franchise systems. To obtain the list of all franchise systems active in Germany and their contacts, we used the directory of the German Franchise Federation (DFV) and “Franchise Wirtschaft” (a Bond’s

Franchise Guide type directory published in Germany). The questionnaire was developed in several steps, refined and discussed within in-depth interviews with franchise experts and practitioners. Finally, we conducted a pre-test with 20 franchisors in Austria. We used the key informant approach for the data collection (McKendall and Wagner III, 1997) to choose the respondents. Most of them were senior managers responsible for the franchise expansion. The questionnaires were sent to 491 relevant franchise systems. We received 137 completed questionnaires, which is a response rate of about 28%. The non-response bias was estimated by comparing early versus late respondents (Armstrong and Overton, 1977). Furthermore, the respondents were also compared regarding their age, size, advertising fee and royalties to determine whether non-response was a problem for the data, because these variables were available in the Franchise Wirtschaft.

10 3.1 Measurement

Dependent variable

Decision rights: Based on the measure developed by Windsperger (2004), residual decision rights are measured by asking franchisors to assess the influence of franchisee’s on the following decisions: selection of suppliers, procurement of equipment, procurement of inputs, price decision, product/service offering, advertising decision, recruitment and training on the outlet level, choice of investment projects and financing of investment projects on the outlet level, decisions on the introduction of new products in the local market, and the application of controlling system. The strength of franchisees’ influence on those decisions was assessed on a seven-point scale (1 = no influence, 7 = very high influence). By averaging the scale values we constructed a decision index varying between 1 and 7. The higher the index, the higher is the franchisee's influence on residual decision making, i.e. franchisees’ fraction of decision rights.

Independent variables

Trust: We measure partner-related trust with four items by asking the franchisor to assess the following items on a 7-point Likert scale: level of confidence between them and their franchise partners, atmosphere of openness and honesty with their partners, readiness of franchisees to cooperate when they are trusted and trustworthiness of their franchise partners.

Our indicators focus on the behavioural dimension by conceptualising trust as confidence the franchisors have on the reliability and integrity of the network partners. All factor loadings exceeded the threshold of 0.70. Reliability analysis was assessed by Cronbach’s alpha (0.87).

Behavioural uncertainty: Behavioural uncertainty arises from the inability to monitor and control the performance of the local partners (Rindfleisch and Heide, 1997; Williamson, 1991).

Therefore, we measure behavioural uncertainty by asking respondents to assess the following items on a 7-point Likert scale: difficulty to measure performance, control behaviour and difficulty to assess the capabilities and competencies of the local managers (franchisees). Factor

11 analysis confirmed the underlying construct, with all variables loading above the threshold of

0.70. Cronbach’s alpha is 0.758.

Environmental uncertainty: The measure of environmental uncertainty is based on Celly,

Frazer (1996) and John and Weitz (1988). To measure this construct, we asked respondents to assess the following items on a 7-point Likert scale: possibility to forecast the local market development and fluctuations of outlet sales at the local market. Cronbach’s alpha is 0.56 which is relatively low. However, recent research assigns more significance to maximise validity rather than internal consistency (John and Benet-Martinez 2000). According to Pedhazur and

Schmelkin (1991), reliabilities above 0.5 can be viewed as acceptable under the condition of construct validity.

Franchisees’ specific investments: According to the transaction cost theory, the governance form is influenced by transaction-specific investments of the franchisees (Klein

1995; Williamson 1983). The natural log of franchisees’ initial investments represents franchisees’ transaction-specific investments.

Control Variables

Sector: This is a dichotomous variable: 0 refers to services franchising and 1 to product franchising. Due to the different know-how intensity of product and services firm, their impact on the allocation of decision rights may also differ.

Size: The size of the network was operationalised by the total number of outlets. From the transaction costs perspective, larger firms have a greater ability to absorb risk and a higher coordination and control capacity (Erramilli and Rao 1993). Therefore, we expect that larger franchise firms can realise economies of coordination and monitoring, thereby increasing control over operational decisions.

3.2 Results

12 Descriptive statistics

Table 1a shows the descriptive statistics. The sample of 127 German franchise systems contains

81 “service franchising” and 46 “product franchising” systems. The decision rights (DR)-index has a minimum value of 2.25 and the maximum of 7. The closer the index value to 7, the stronger the franchisees’ influence on operational decisions. A mean of 5.13 indicates relatively high influence of franchisees on the analysed decisions.

Insert Table 1a

The descriptive statistics regarding the components of the dependent variable (decision rights- index) is specified in Table 1b. The values from 1 to 7 represent the extent to which franchisees have influence on a particular decision (1 = no influence on the decision; 7 = very high influence on the decision). Generally, we can divide them into values <5.00 and >5.00. The mean values which are <5.00 refer to the decisions regarding suppliers, product, equipment and inventory procurement, innovation of products/services and controlling system. These decisions are closely related to the business concept (e. g. which product to offer, which inventory or equipment to use to ensure the quality level, which suppliers will provide final or intermediary products and which controlling system to use). On the other hand, the decisions such as recruitment, employee training, advertising and price decisions, as well as investments and financing, are under stronger influence of the franchisees. This may indicate that franchisees’ local market knowledge and managerial capabilities play a more important role for these decisions. Table 2 presents the correlations between the variables used in the regression analysis. None of the correlations seem to be large enough to indicate concern about multicollinearity (Hair et al. 1995).

13 Insert Table 1b

Insert Table 2

Regression results

The hypotheses are tested by applying multiple regression analysis. Results are shown in Table

3. Model 1 includes only the control variables. Model 2 includes the TC variables

(environmental uncertainty, behavioural uncertainty and transaction-specific investments) and tests H1, H2, and H3. The moderating effect of trust, hypothesised by H1a, H2a, H3a, is tested in models 3-6.

Insert Table 3

Model 1 shows that the control variables have no significant effects on allocation of decision rights. Model 2 includes behavioural uncertainty, environmental uncertainty and franchisees’ transaction-specific investments, which then tests the first-order effects on the franchisee’s portion of decision rights. Behavioural uncertainty has a positive and strongly significant influence on the delegation of decision rights (ß= 0.486; p < .01). Although the coefficient is significant, this result does not confirm the transaction cost hypothesis H1. A positive sign of the coefficient implies that the franchisors actually delegate more decision rights when they encounter difficulties in measuring franchisees’ performance and controlling his/her behaviour.

This result is compatible with the incentive view of delegation (Aghion and Tirole 1997). Under higher behavioural uncertainty, the franchisors provide more incentives by delegation of operational decisions. This result is robust across all tested models. The negative coefficient of environmental uncertainty is slightly supportive of the hypothesis H2 (ß= -0.209; p <.10). This

14 result confirms that franchisors tend to implement more elements of hierarchy, i.e. centralised decision making, when the degree of external uncertainty increases. The coefficient of initial investments, representing the effect of franchisee’s transaction-specific investments on the allocation of decision rights, has the expected positive sign indicating that higher franchisees’ transaction-specific investments will result in more delegation of decision rights. However, the coefficient is not significant.

In Models 3-6, we use moderated regression analysis (MRA) to examine the moderating effects of trust on the allocation of decision rights (Sharma et al. 1981). For the analysis, the scores of the main effects are mean-centered to avoid problems of multicollinearity. The correlations between the interaction terms and the centered main effect variables do not indicate any problems with multicollinearity. In addition, all VIF values in our model were below the value of 2, indicating no problem with multicollinearity (Belsley et al. 1980).

Model 3 tests the moderating effect of trust on the relationship between behavioural uncertainty and decision rights allocated to franchisees. H1a hypothesised that the negative relationship between behavioural uncertainty and franchisee’s portion of decision rights will be subdued under high trust. The interaction coefficient is negative and significant (ß= -0.204, p < .

05) indicating that trust as implicit bond weakens the positive impact of behavioural uncertainty on the delegation of decision rights. Consequently, under high performance measurement difficulties, the franchisors have less need to delegate decision rights as formal incentives when they trust their partners. Furthermore, the results show that both the main effects - behavioural uncertainty and trust - and their interaction terms are statistically significant. This means that trust not only interacts with behavioural uncertainty, but also directly impacts the delegation of decision making to franchisees. Trust is therefore a quasi moderator (Sharma et al. 1981). Its moderator effect is thus not clear, as each of the interaction components (behavioural uncertainty and trust) could be interpreted as a moderator. However, TC-theory provides the theoretical

15 justification for treating trust as a moderator variable, as the traditional TC-theory neglects the existence of trust between transaction partners and its effect on the governance choice.1 The effect of this interaction term is also confirmed in Model 6, where all interaction terms are included in the model.

In Model 4 we test H2a, which predicts that trust will weaken the negative effect of market uncertainty on the allocation of decision rights to franchisees. The interaction coefficient is positive and strongly significant (ß= 0.272, p<0.01) indicating that - under high trust

-franchisors reduce control over operational decisions and rely more on franchisees’ decisions when environmental uncertainty increases. Therefore, the negative effect of environmental uncertainty becomes weaker by the trust variable. The main effect of the environmental uncertainty, although negative as expected, is not significant. Similar to Model 3, trust is a quasi moderator, as its main effect is statistically significant. This indicates that even if the value of environmental uncertainty were zero, trust would still have a positive effect on the dependent variable. This moderation effect remains statistically significant in the Model 6 as well.

Model 5 provides results for the hypothesis H3a, which predict a positive moderating effect of trust on the relationship between franchisees' initial investments and decision rights.

Although positive as expected, the coefficient of the interaction term is not statistically significant. Finally, Model 6 provides the regression results with all three interaction terms included in the regression analysis. In this model, all interaction effects are statistically significant. Contrary to the results in Model 5, the moderating effect of trust on the relationship between initial investments and decision rights is significant and positive as expected (b= 0.278, p<0.01). This provides support for hypothesis H3a. Since trust increases the quasi rent- generating effect of transaction-specific investments, the delegation-increasing effect of higher transaction-specific investments is higher under high trust. The coefficient of trust as the main

1According to Sharma et al. (1981), the search for moderator variables should be guided by theory rather than by strict empiricism and the definition of moderator variable need not be limited to the psychometric definition (which suggests only pure moderators). 16 effect is also positive and strongly significant (ß= 0.329, p<0.01). The adjusted R² reaches 0.440, indicating that the model explains 44 % of the variance. Overall, the empirical results provide some support of the transaction cost hypotheses. In addition, the results show that trust as a moderator has significantly improved the explanatory power of the transaction costs model.

4. Discussion and Implications

This paper presents an extended transaction cost model decision rights allocation in franchising. Specifically, it investigates the influence of the transaction cost variables

(behavioural uncertainty, environmental uncertainty and transaction-specific investments) as well as trust as a moderator on the allocation of decision rights to franchisees. The empirical results from the German franchise sector provide partial support of our TC hypotheses. First, we find that behavioural uncertainty has a positive effect on the allocation of decision rights to franchisees. In contrast to our transaction cost interpretation, these findings imply that franchisors tend to delegate decision rights to franchisees when they encounter difficulties in monitoring behaviour, and measuring performance of the franchisees. This result is compatible with the incentive view of delegation (e.g. Aghion and Tirole 1997; Baker et al. 1999). Aghion and Tirole argue that delegation of decision rights to the agents increases their incentives to search for and develop projects. Consequently, in a situation of high behavioural uncertainty, the franchisors provide more formal incentives through delegation instead of tightening control.

Secondly, we find some support for the hypothesis that franchisors transfer less residual decision rights to the franchisees when they perceive higher environmental uncertainty.

Furthermore, our results provide strong evidence of the moderating role of trust in the relationships between the transaction cost variables and the franchisees’ fraction of decision rights. We find that trust reduces the negative impact of environmental uncertainty on franchisor’s control over operational decisions. In a situation of high environmental uncertainty,

17 the franchisors will transfer more decision rights to the franchisees, if they are perceived to be trustworthy, i.e. believe that they will not take advantage of an uncertain business environment for short-term opportunistic gains. This result is similar to the findings of Ryu et al. (2008) who found that trust weakens positive relation between manufacturers’ perception of environmental uncertainty and their propensity for vertical integration. Furthermore, we find that trust as implicit bond weakens the impact of behavioural uncertainty on delegation of decision rights.

Therefore, in a high-trust situation, the franchisors have less need for delegating decision rights as formal incentives. Finally, we find that trust increases the bonding effect of transaction- specific investments, hence strengthening their positive impact on the delegation of decision rights to franchisees. Interestingly, the results show trust functions as a quasi-moderator, as it has statistically significant direct and indirect effects on the dependent variable. Although the moderating effect of trust is significant, a significant main effect of trust is not expected by the transaction costs theory. This result could be explained from a relational governance perspective.

This view sees trust as a principal mode of informal control (Ring and Van de Ven 1992) which implies that the franchisor’s reliance on the goodwill and integrity of the network partners (Ring

1996) has a direct impact on the governance choice. Contrary to this view, the role of trust in TC theory depends on its capacity to reduce the likelihood of opportunistic behaviour by an exchange partner. Consequently, trust has a moderating role on the impact of the transaction cost variables on the choice of formal governance (i.e. allocation of decision rights). In conclusion, our empirical results demonstrate that considering trust in the TC model supplements the explanation of the allocation of decision rights in franchising offered by the TC theory.

Implications and Limitations

This study has important implications for both researchers and franchisors.

Complementary to the agency-theoretical and property rights perspectives (Arrunada et al., 2001,

2005; Azevedo, 2009; Windsperger 2004), we develop an extended transaction cost explanation

18 of the decision rights allocation in franchising networks. Specifically, we extend the transaction cost theory by considering trust as moderating mechanism between transaction cost variables and decision rights as formal governance mechanism. We find some support that the transaction cost variables and trust (as moderating variable) impact the allocation of decision rights between the franchisor and franchisees. This analysis provides new insight regarding the allocation of residual decision rights in franchise relationships. Specifically, the moderating role of trust indicates that trust increases delegation of decision rights to the franchisees, when franchisees’ transaction-specific investments and environmental uncertainty are high. On the other hand, trust as an implicit bond reduces delegation as formal incentives, especially when it is difficult to measure franchisees’ performance. Furthermore, our study also contributes to the literature on the relationship between formal governance and trust. Consistent with the relational governance view (e. g. Poppo and Zenger, 2002; Lui and Ngo 2004; Yu et al. 2006; Mellewigt et al. 2007), the findings of this study show that trust is quasi-moderator by directly influencing the allocation of decision rights.

The results of our study yield practically relevant knowledge for franchisors seeking to allocate residual decision rights in the franchise system. Based on the incentive view of delegation, the franchisor should delegate a higher fraction of decision rights to the franchisees when behaviour control and performance measurement at the local outlet are difficult. On the other hand, the TC-view suggests that the franchisor should increase control over operational decisions when the environmental uncertainty is high. By contrast,, the franchisor or franchisor- manager has always to consider in their governance decisions that trust-based relationships with franchisees will increase the efficiency advantages through delegation

Finally, our study has some important limitations: Firstly in our study, the influence of the variables used in the regression analysis depends on measures based on the franchisors’ evaluation. However, franchisors’ assessment could to some extent deviate from franchisees’ evaluations. To include both perspectives would contribute to the reliability of the measures.

19 Secondly, since franchising networks evolve, the role of trust might vary with the relationship cycle. To grasp the dynamics of trust between the franchise partners a longitudinal design would be beneficial. A third limitation of this study is that we examined the impact of transaction cost variables and trust on decision rights, without investigating the performance implications of the allocation of decision rights for the franchisor. Future research has to investigate the relationship between the decision rights structure and the efficiency of the franchise systems.

5. Conclusion

Our study offers an extended transaction cost explanation of the allocation of decision rights in franchising. We demonstrate that considering trust in transaction cost theory supplements the explanation offered by the transaction cost theory on the allocation of decision rights in franchising networks. The results indicate that trust moderates the impact of transaction cost variables on the allocation of decision rights between the franchisor and franchisees.

Overall, we contribute to the franchising literature by constructing and testing an extended transaction cost model to explain the structure of decision rights in franchising networks.

20 References

Aghion, P., Tirole, J. (1997), Formal and real authority in organization, Journal of Political Economy, 105, 1-29

Alvarez, S.A., Barney, J.B., Bosse, D.A. (2003), Trust and its alternatives, Human Resource Management Journal, 42(4), 393-404

Anderson E, Gatignon H (1986) Modes of foreign entry: A transaction cost analysis and propositions. Journal of International Business Studies 17(3): 1-26

Armstrong, J.S., Overton, T.S. (1977), Estimating non-response bias in mail surveys, Journal of Marketing Research, 14, 396-402

Arrunada, B., Garicano, L., Vazquez, L. (2001), Contractual allocation of decision rights and incentives: The case of automobile distribution, Journal of Law, Economics & Organization, 7, 257-286

Arrunada, B., Garicano, L., Vazquez, L (2005), Completing Contracts Ex Post: How Car Manufacturers Manage Car Dealers. Review of Law & Economics, 1,149-173

Azevedo, P.F. (2009), Allocation of authority in franchise chains, International Studies of Management and Organization, 39 (4), 31-42

Baker, G., Gibbons, R., Murphy, K.J. (1999). Informal authority in organizations. Journal of Law Economics & Organization, 15(1), 56-73

Belsley, D. A., Kuh, E., Welsch, R.E. (1980). Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York: John Wiley.

Bradach, J.L., Eccles, R.G. (1989), Price, authority, and trust: From ideal types to plural form, Annual Review of Sociology, 15: 97-118

Celly, K.S., Frazier, G.L. (1996), Outcome-based and behavior-based coordination efforts in channel relationships, Journal of Marketing Research, 33(2), 200-210

Cochet, O., Dormann, J., Ehrmann, T. (2008), Capitalizing on franchisee autonomy: Relational forms of governance as controls in idiosyncratic franchise dyads. Journal of Small Business Management, 46(1), 50-72

Cronbach, L.J. (1951), Coefficient alpha and the internal structure of tests, Psychometrika, 16(3), 297-334

Davies, M.A.P., Lassar, W., Manolis, C., Prince, M., Winsor, R.D. (2011), An model of trust and compliance in franchise relationship, Journal of Business Venturing, 26, 321-340

Dickey, M.H.; Harrison, D., McKnigh, J.F.G. (2007), The role of trust in franchise organizations, International Journal of Organizational Analysis; 15, 3, 251-282

Erramilli, M.K., Rao C.P. (1993), Service firms' international entry-mode choice: A modified Transaction-Cost Analysis Approach. Journal of Marketing 57(3): 19-38

Gatignon, H., Anderson E. (1988), The multinational corporation’s degree of control over foreign subsidiaries: An empirical test of a transaction cost explanation. Journal of Law, Economics & Organization, 4, 305 – 336.

Gulati, R. (1995), Does familiarity breed trust?, Academy of Management Journal, 38, 85-112

Gulati, R., Nickerson, J.A. (2008), Interorganizational trust, governance choice and exchange performance, Organization Science, 19(5), 688-708

Hair, J.F., Anderson, R.E., Tathan, R.L., Black, W.C. (1998), Multivariate Data Analysis, Prentice Hall.

21 Hennart, J.F. (1991), The Transaction Costs Theory of Joint Ventures: An Empirical Study of Japanese Subsidiaries in the United States. Management Science, 37(4), 483–497

Hoffmann, W.H., Neumann, K., Speckbacher, G. (2010), The effect of interorganizational trust on make-or- cooperate decisions: Disentangling opportunism-dependent and opportunism-independent effects of trust. European Management Review, 7, 101–115

Hwang, P. (2006), Asset specificity and the fear of exploitation. Journal of Economic Behavior & Organization, 60, 423–438

John, O.P., Benet-Martinez, V. (2000), Measurement: Reliability, construct validation, and scale construction. In H. T. Reis & C. M. Judd (Eds.), Handbook of research methods in social psychology, pp. 339-369. New York: Cambridge University Press.

John, G., Weitz, B.A. (1988), Forward integration into distribution: an empirical test of transaction cost analysis, Journal of Law, Economics, and Organization, 4(2), 337-355

Katz, A.W. (2008), Contractual enforcement institutions and the structure of information, Journal of Institutional and Theoretical Economics, 164, 134-154

Klein, B. (1995), The economics of franchise contracts, Journal of Corporate Finance, 2(1-2), 9-37

Klein, B. (1996). Why Hold-ups occur: The self-enforcing range of contractual relationship. Economic Inquiry, 34(3), 464-484

Klein, B. (2000), The role of incomplete contracts in self-enforcing relationships, Revue D’economie Industrielle, 92, 67-80

Lazzarini, S.G., Miller, G.J., Zenger, T.R. (2008), Dealing with the paradox of embeddedness: The role of contracts and trust in facilitating movement of committed relationships. Organization Science, 19(5), 709 – 728

Lopez-Fernandez, B., Lopez-Bayon, S. (2011), Delegation of autonomy in franchising, In: Tuunanen et al. (Eds.), New developments in the theory of networks, Physica-Verlag, Heidelberg

Lui, S.S., Ngo, H.Y. (2004), The role of trust and contractual safeguards on cooperation in non-equity alliances. Journal of Management, 30(4), 471–86

Mayer, R.C., Davis, J. H., Schoorman, F.D. (1995), An integrative model of organizational trust. Academy of Management Review, 20, 709-734

McKendall, M.A., Wagner, J.A. III (1997). Motive, opportunity, choice, and corporate illegality. Organization Science, 8(6), 624-647

Mellewigt, T., Madhok, A., Weibel, A. (2007). Trust and formal contracts in interorganizational relationships - substitutes and complements. Managerial and Decision Economics, 28(8), 83-847

Mumdziev, N., Windsperger, J. (2011). The Structure of Decision Rights in Franchising Networks: A Property Rights Perspective, Entrepreneurial Theory and Practice, 35, 449 - 465

Noordewier, T.G., John, G., Nevin, J.R. (1990), Performance Outcomes of Purchasing Arrangements in Industrial Buyer-Vendor Relationships. The Journal of Marketing, 54(4), 80-93

Nooteboom, B., Berger, H., Noorderhaven, N.G. (1997), Effects of trust and governance on relational risk. Academy of Management Journal, 40, 308 – 338

Pedhazur, E. J., Schmelkin, L.P. (1991), Theories, problems and hypotheses. In: Pedhazur, E.J., Schmelkin, L.P. (Eds.), Measurement, design and analysis: An integrated approach. Hillsdale, NJ: Lawrence Earlbaum Associates.

Poppo. L., Zenger, T. (2002), Do formal contracts and relational Governance function as substitutes or complements, Strategic Management Journal, 23, 707-725

22 Rindfleisch, A., Heide, J.B. (1997), Transaction cost analysis: Past, present, and future applications, Journal of Marketing, 61(4), 30-54.

Ring, P.S. (1996), Networked organization: A resource based perspective Uppsala and Stockholm, Sweden: Uppsala Universitet.

Ring, P.S., Van De Ven, A.H. (1992), Structuring Cooperative Relationships between Organizations. Strategic Management Journal, 13(7), 483-498

Ryu, S. Soonhong, M., Nobuhide, Z. (2008), The moderating role of trust in manufacturer-supplier relationship, Journal of Business & industrial Marketing, 23 (1), 48-58

Sharma, S., Durand, R.M., Gur-Arie, O. (1981), Identification and Analysis of Moderator Variables. Journal of Marketing Research 18: 291-300.

Stinchcombe, A.L. (1990), Organizations and information. University of California Press, Berkeley, CA.

Williamson, O. E. (1975), Markets and Hierarchies: Analysis and Antitrust Implications New York: The Free Press.

Williamson, O. E. (1983). Credible commitments: Using hostages to support exchange. The American Economic Review, 73(4), 519-540.

Williamson, O.E (1985), The economic institutions of capitalism, New York: The Free Press

Williamson, O.E (1991), Comparative economic organization: the analysis of discrete structural alternatives, Administrative Science, 36, 269-296

Windsperger, J. (2004), Centralization of franchising networks: Evidence from the Austrian franchise sector, Journal of Business Research, 57 (12), 1361-1369

Windsperger, J. (2009), Allocation of Decision Rights in Joint Ventures, Managerial and Decision Economics, 30, 491- 501

Yu, C-M.J., Liao, T.-J., Lin, Z.-D. (2006). Formal governance mechanisms, relational governance mechanisms, and transaction-specific investments in supplier-manufacturer relationships. Industrial Marketing Management, 35, 128 – 139

Zaheer, A., Venkatraman, N. (1995), Relational Governance as an Interorganizational Strategy: An Empirical Test of the Role of Trust in Economic Exchange, Strategic Management Journal, 16, 373-392

23 APPENDIX

Figure 1 Research Model

24 Table 1: Descriptive statistics

VARIABLE N Min Max Mean Std. Dev.

Sector 127 Product franchising 46 Service franchising 81 Size 118 1 2,520 155.84 328,42

114 600.00 2,050,00 130.942. Franchisees’ specific investments 22.819.40 0 4 DR Index 127 2.25 7.00 5.13 0.10 Behavioral uncertainty Measurement of performance 127 1 7 3.33 0.140 Control of behavior 127 1 7 4.20 0.136 Assessment of capabilities 127 1 7 3.35 0.139 Environmental uncertainty Market development predictions 127 1 7 3.78 0.145 Outlet sales predictions 127 1 7 4.01 0.160 Trust Cooperation on partnership basis 127 1 7 6.10 0.094 Information exchange between partners 127 1 7 5.38 0.124 Existence of trust between partners 127 1 7 5.83 0.099 Openness and honesty between partners 127 1 7 5.87 0.097

25 Table 2: Correlations between regression variables

Behavior Environ Specific Decision al m. Sector Size investme Trust rights uncertain uncertain nts ty ty

Decision rights 1

Sector 0.038 1

Size 0.160 0.053 1

Behavioral uncertainty .370** -0.045 0.071 1

Environmental uncertainty -0.085 -0.102 0.014 0.129 1

Specific investments 0.021 -0.011 0.088 0.061 - .313** 1

Trust .357** 0.096 0.045 -0.056 - .365** 0.056 1

** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

26 Table 3: Regression results+

MODEL Dependent variable: 1 2 3 4 5 6 Decision rights index Control variables 5.190*** 5.141*** 5.131*** 5.205*** 5.168*** 5.212*** Constant (0.136) (0.133) (0.120) (0.133) (0.122) (0.114) + 0.076 + 0.191 + 0.217 + 0.267 + 0.150 + 0.266 Sector (0.220) (0.208) (0.189) (0.184) (0.193) (0.178) + 0.001 + 0.000 + 0.000 + 0.000 + 0.000 + 0.000 Size (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Transaction cost hypotheses H1 Behavioral uncertainty + 0.486*** + 0.568*** + + + 0.619*** (0.101) (0.096) 0.610*** 0.492*** (0.092) (0.094) (0.096) H2 Environmental uncertainty - 0.209* - 0.107 - 0.107 - 0.043 - 0.209** (0.115) (0.115) (0.108) (0.113) (0.110) H3 Franchisees’ specific + 0.142 - 0.098 - 0.178* - 0.095 - 0.088 investments (0.112) (0.101) (0.100) (0.108) (0.100) Moderator effects of Trust

Trust + 0.474*** + + + 0.329*** (0.099) 0.363*** 0.429*** (0.099) (0.097) (0.101) H1a Trust × Behavioral uncertainty - 0.204** - 0.166* (0.095) (0.095) H2a Trust × Environmental + + 0.345*** uncertainty 0.272*** (0.090) (0.081) H3a Trust × Specific investments + 0.054 + 0.278*** (0.094) (0.098) N 116 101 101 101 101 101 10.379* 7.907** F-test 1.518 6.055*** 8.874*** ** * 9.812*** R² 0.026 0.240 0.396 0.436 0.371 0.490 Adjusted R² 0.009 0.200 0.353 0.394 0.324 0.440

+ Standardized regression coefficients are reported. *** p < .01; ** p < .05; *p < .1;

27 APPENDIX: MEASURES OF VARIABLES

Variable Items

Trust Please assess the following statements on a scale 1 – 7: 1 = not true at all; 7 = true to the high extent (1-1) The cooperation is based on partnership basis (1-2) The exchange of information between us and the partners goes beyond the agreed scope (1-3) There is great trust between ourselves and the partners There is an atmosphere of openness and honesty between us and the (1-4) partners Behavioral Please assess the following statements on a scale 1 – 7: uncertainty 1 = I do not agree at all; 7 = I fully agree (2-1) It is difficult to measure performance of the franchisees. (2-2) It is very difficult to control the behavior of the franchisees. It is very difficult to assess the competencies and capabilities of (2-3) franchisees. Environmental Please assess the following statements on a scale 1 – 7: uncertainty 1 = I do not agree at all; 7 = I fully agree (3-1) It is very difficult to predict the market development at the outlet level (3-2) The sales at the outlet level are very fluctuating. Franchisees’ Natural log of initial investments (in EUR) specific investment DR Index To which extent do franchisees have influence on following decisions? Scale 1-7 (1 = no influence at all; 7 = to a very high extent)

(4-1) Investment decision (4-2) Financing decision (4-3) Supplier decision (4-4) Recruiting decision (4-5) Employees’ training decision (4-6) Product/service decision (4-7) Resale price decision (4-8) Advertising decision (4-9) Accounting system decision

Sector 0 = service franchising; 1 = product franchising

Size Number of franchised and company-owned outlets