Firms' relational model and joint value creation: the moderating role of leader-member exchange

(1)

Jeroen Jansma ( 10978992)

University of Amsterdam

Executive programme in management studies Master thesis Leadership and Management

Under the supervision of Dr. Claudia Buengeler

(2)

Statement of originality

This document is written by Jeroen Jansma who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

Using a experimental design (N = 472), with four different relational models ( communal sharing; equality matching; authority ranking; market pricing) and two levels of Leader-Member Exchange (LMX) quality (low; high), this study tested the main effects of firms relational models on the willingness to joint value creation, as well as the moderating effects of LMX quality. The main effect of the hypotheses about relational models was supported and authority ranking showed to have a more positive influence on the willingness to joint value creation in comparison to the relational model market pricing. Additionally, LMX did interact significantly within the relational model equality matching and this study found support for the expectation of a non-significant influence of LMX in the relational model market pricing. These results indicate that organizations and managers can influence social welfare of their stakeholder's through relational models and LMX, as a result increase the willingness to joint value creation and limit team production problems.

Keywords: Joint value creation, Relational models, Leader-member exchange, Team production problem, Stakeholders' theory, Social welfare.

(4)

1. Introduction.

Every day business owners, shareholders and chief executives strive for business success by creating a competitive advantage. The resource based view (RBV; Barney, 1991) adopts the perspective of a firm that creates a competitive advantage based on its (human) capital. In the research field of human resource management, the resource based view is a

theoretical perspective explaining the black box between human resources and performance. The firm’s perceived creation of value through human capital shows the need for cooperation among the levels of an organization (Smith, Carroll, & Ashford, 1995), and often requires employees to cooperate in performing highly interdependent tasks (Coff, 1997; Lado & Wilson, 1994). The role of cooperation and striving for joint value creation becomes even more important in the future. In particular, the success of emerging structural forms, such as the self-managed task team, the horizontal organization, the network organization, and the virtual corporation rest largely on effective cooperation (Smith et al.1995). Not only business owners, shareholders and chief executives but also other stakeholders benefit from the joint value creation. From a stakeholder's perspective the individuals who contribute to the joint value are seen as the core of the organization and the reason of the existence of the firm (SRI, 1963, as cited in Freeman & Reed, 1983). Multiple researchers have shown the importance of joint value creation from a stakeholder's perspective (Freeman, Harrison, Wicks, Parmar & De Colle, 2010; Harrison, Bosse & Phillips, 2010; Jones, 1995; Post, Preston & Sachs, 2002). Therefore, it is in the firm's best interest to create a sustainable competitive advantage by achieving joint value creation among the stakeholders within the firm.

However, in achieving joint value creation among the stakeholders, the firm usually face social dilemmas such as the public good dilemma. (Fehr, Fischbacher & Gächter, 2002; Ostrom, 2000): the maximization of individual employees’ material payoffs conflicts with the achievement of the collective goal of creating value for the firm (Olson, 1965). Under such circumstances one of the key challenges for the firm in order to enlarge the joint value creation is to provide an environment that helps overcome social dilemmas (Von Rueden & Van Vugt, 2016)

Stoelhorst and Bridoux (2013, 2016) have shown that in order to enlarge the joint value creation, one should encounter the framing of social dilemma's, at the individual level, group level as well as the firm level. Bridoux and Stoelhorst (2013, 2016) approach the enlargement of joint value creation from the perspective of Ostrom (2000) in which they let go of the classical homo economics theory and show the contribution of people in a social context. The foundations for framing the social dilemmas are the four relational models of

(6)

Fiske (Fiske, 1992, 2004, 2012; Haslam, 2004). That being the case a firm can perceive a relational model, through the appropriate relational model influence employees’ willingness to joint value creation and build social welfare as a potential source of competitive advantage (e.g., Collins & Smith, 2006; Evans & Davis, 2005; Mossholder, Richardson & Settoon, 2011; Stoelhorst & Bridoux, 2013)

This study builds upon the arguments put forward by Bridoux and Stoelhorst (2013, 2016) that joint value creation process is about creating social welfare through the

appropriate relational model, instead of just putting resources together and adding up person's individual contributions (Alchian & Demsetz, 1972). Additionally, this research emphasises the role of the manager in striving for joint value creation, by examining influences of the relationship between a leader and member. Both in stakeholders' theory and in the team production problem, the manager or leader plays an important role in accomplishing the desired result (Alchian & Demsetz, 1972; Donaldson & Preston 1995; Lado & Wilson, 1994). If the managers of today's firms want to create organizations that have a competitive advantage, they need to understand the causes and consequences of

cooperation (Smith et al. 1995). Based on the stakeholder's perspective, Donaldson & Preston (1995) concludes that corporate managers must induce constructive contributions from their stakeholders in order to accomplish their own desired results (e.g., perpetuation of the organization, profitability, stability, and growth). The relationship between the leader and member seems to have a crucial role in the establishment of joint value creation, because it has positively influencer of subordinate outcomes (Dulebohn, Bommer, Liden, Brouer, & Ferris, 2012; Hoegl & Gemuenden, 2001; Stewart & Johnson, 2009; Woolley, Chabris, Pentland, Hashmi & Malone, 2010)

Concluding, this study sets out to examine if the relationship quality between leader and member has a moderating effect on the relation between the firms relational model and the stakeholders willingness to contribute to joint value creation by answering the following question: Does leader-membership exchange influence the relationship between the firm's relational model and the stakeholders willingness for joint value creation? In so doing, this research provides a contribution to literature which is twofold: Firstly, it serves to replicate the research of Stoelhorst and Bridoux (2013) about the influence of the firms' relational model on the stakeholders' willingness to contribute in joint value creation. Doing so is relevant because, it stimulates future researches to let go of the transactional approach and use stakeholders theory as starting point. Secondly, this study shows the influence of high quality leader-membership exchange on the joint value creation process in different relational models. Doing so is relevant, because it extends the current knowledge about the influence of leader-member exchange and connects it to the literature of

(7)

stakeholders' theory. This study is also important practically, because it shows the firms' influence on the actions of their individual stakeholders as it reveals which relational modal a firm must perceive to increase social welfare. Additionally, this study gives practical insights for managers in how to limit team production problems.

2. Theoretical background and hypotheses

2.1 Joint value creation.

Bridoux and Stoelhorst (2016) define joint value creation as: "the value creation processes involving multiple parties, within and/or across the firm’s boundaries, which face high task and outcome interdependence in providing mutually supportive contributions to value creation" (Bridoux & Stoelhorst ,2016,p. 231). Research of Bridoux, Coeurderoy, & Durand, 2011; Dyer &Singh, 1998; Grant, 1996; Lindenberg & Foss, 2011 has shown the importance of joint value creating from the perspective of the resource based view. This means that, in order to create a competitive advantage a firm can create value through cooperation between its stakeholders.

2.2. Stakeholders

According to the Stanford Research Institutes (SRI) stakeholders are defined as: "those groups without whose support the organization would cease to exist" (SRI, 1963; as cited in Freeman & Reed, 1983). Donaldson & Preston (1995) show how the stakeholder model contrasts with the input-output model, as stakeholder's contribute in organization to obtain commonly shared benefits. Survey research has shown that managers deem it important to focus on the interest of the stakeholders instead of purely focussing on the interests of the shareholders. (Brenner & Molander, 1977; Posner & Schmidt, 1984). The studies by Clarkson (1991) and Halal (1990) distinguish firms that practice stakeholder management from those that do not, and both investigators found considerable numbers of firms in the first category. Managers may not make explicit reference to "stakeholder theory," but the vast majority of them apparently adhere in practice to one of the central tenets of stakeholder theory, namely, that their role is to satisfy a wider set of stakeholders, not simply the shareowners (Donaldson & Preston (1995).

Bringing stakeholders together, let them strive for common goals and fulfil these goals seems grounded in the stakeholders' perspective of the resourcebased view (achieving joint value creation). While seeking to achieve joint value creation, organizations can face social dilemmas like the public good dilemma (Olson, 1965), underinvestment in team-specific

(8)

assets (Blair & Stout, 1999) and shirking (Alchain, 1972) as a result of the assumption that human behavior is exclusively motivated by self-interest.(Bridoux & Stoelhorst, 2016). Based on this recognition, Ostrom (2000) shows that by letting go of the rational and self interested perspective and instead focusing on the social context, stakeholders become much better at solving social dilemmas.

2.3 Relational model

In order to show how stakeholders mental representations of relationships affect

contributions to joint value creation, the present research uses the relational models theory (Haslam, 1999,2004; Fiske 1992, 2004, 2012) to frame the organization's relational model. This is in line with the research of Stoelhorst and Bridoux (2013; 2016) Bridoux and

Stoelhorst (2013) have studied the impact of four elementary organization forms identified by Fiske on cooperation among employees. The four relational models, communal sharing, equality matching, authority ranked and market pricing, of Fiske (Fiske 1992, 2004, 2012; Haslam, 1999,2004) are the foundation for rich and complex relationships (Sheppard & Tuchinsky, 1996, Bridoux & Stoelhorst 2016) and operate at all levels of social interactions (Fiske & Haslam, 2005). Relational models theory states that the four elementary relational models are the building blocks of all complex social structures, including firms. Moreover, while the models are rarely found in ‘pure types’, often one model or another dominates in a given firm (Chacar & Hesterly, 2008; Stoelhorst & Bridoux, 2013).

2.3.1 Communal sharing

Communal Sharing (CS) is a relationship in which people treat some dyad or group as equivalent and undifferentiated with respect to the social domain in question. In this kind of relationship, the members of a group or dyad treat each other as all the same, focussing on commonalities and disregarding distinct individual identities. Examples of communal sharing groups are: groups that are based on trust and fair play, people who deal with conflict

through mediation and other methods that seek agreement from all, people intensely in love, kin group.

2.3.2 Equality matching

In Equality Matching (EM) relationships, people keep track of the balance or difference among participants and know what would be required to restore balance. The idea is that each person is entitled to the same amount as each other person in the relationship. Common manifestations are turn-taking, one-person one-vote elections and equal share distributions. Examples are sport team mates and work peers.

(9)

2.3.3. Authority ranking

In Authority Ranking (AR) people have asymmetric positions in a linear hierarchy in which subordinates defer, respect, and (perhaps) obey, while superiors take precedence and take pastoral responsibility for subordinates. The salient social factor in an AR relationship is whether a person is above or below someone else. The relational models theory posits that when people are thinking in terms of linearly ordered structures, they treat higher rank as superior. Examples are military hierarchies, social status systems such as class or ethnic rankings and rankings such as sports team standings/performance.

2.3.4. Market pricing

Market Pricing (MP) relationships are oriented to socially meaningful ratios or rates. People in market pricing usually reduce all the relevant features and components under

consideration to a single value or utility metric that allows the comparison of many diverse factors. In this kind of relationship, people tend to discuss value-relevant features of things propositionally, analyzing input and output. Examples are property that can be bought, sold, or treated as investment capital, bureaucratic cost-effectiveness standards, and

considerations of spending time efficiently.

2.4 Firms' relational model and joint value creation

Firm's can benefit from a well-defined relational model as it creates a distinct perception of the social interaction and generates a set of common expectations (Vodosek, 2003). In contrast, the use of dissimilar relational models is likely to create different perceptions of a situation among collaborators and may generate deviating expectations. Diverging

perceptions and expectations, in turn, hinder coordinated actions like joint value creation (Vodosek, 2003). As a consequence, it is expected that a clear relational model throughout the firm has a positive influence on joint value creation.

The relational models are not only cognitive frames but also consist of needs,

motives, evaluative attitudes, and judgments, as well as emotions (Fiske, 1992). The models trigger different behaviors in social interactions because they make different relational self-representations salient and these are associated with different motivations and, therefore, involve different rules of behavior (Stoelhorst & Bridoux, 2016). Because each relational model represents a distinct mental model for coordinating, organizing, and evaluating all aspects of a relationship, it is expected that all four relational models differ in stimulating when and how stakeholders are willing to contribute to joint value creation.

(10)

Joint value creation in communal sharing.

The organization form communal sharing is focused on forming a community. Employment relationship is likely to be long term and relational, requiring open-ended obligations on the part of both the organization and employees (Rousseau, 1995, 2004; Tsui, Pearce, Porter & Tripoli, 1997). A communal sharing climate focuses on increasing employees’ well-being and extending their organizational careers. In exchange, employees are expected to accept the organization’s interests as their own (Mossholder, Richardson & Settoon, 2011). The organization is looking to attract employees who can meet broad work demands and whose values support a willingness to work together with others (Hom et al., 2009). Training for employees is focused on interpersonal skills, team building as a way of increasing their understanding of others (e.g., Heaney, Price, & Rafferty, 1995). Incentives promote social behaviors like knowledge sharing, peer support, and helping (Mossholder et al., 2011). Stakeholders see themselves as members of this community. As a consequence, they are energized to exert themselves on the collective’s behalf, to direct and facilitate others’ effort toward the collective outcome, and to remain loyal to the collective, as they trust that other participants will behave in the same way (Ellemers, de Gilder, & Haslam, 2004; Fiske, 1992; Haslam & Ellemers, 2005). Feelings of commonality will motivate stakeholders to do

whatever is necessary to get the work done because they perceive the work to be done as “our work” (Fiske, 1992). Stakeholders personal welfare of others is considered significant above self-concerns (Fiske & Haslam, 2005) and the identification with a collective makes them more likely to cooperate and engage in joint value creation (Adler, 2001; Wenger, 1998). Based on the characteristics of the community organization and in line with the study of Mossholder et al. (2011) it is expected that the relational model communal sharing in organizations leads to higher willingness to contribute in joint value creation than do the three other models.

H1a. Willingness for joint value creation is higher when the firm's relational model is communal sharing rather than equity matching, authority ranking or market pricing.

Joint value creation in equality matching

Equality matching show similar benefits for cooperation as the relational model communal sharing (Fiske, 1999; Stoelhorst, 2013). The implied employment relationship (i.e.,

psychological contract) in equality matching perspective reflects a balanced psychological contract with both transactional and relational attributes, it requires the organization and employee to strive toward common interests (Mossholder et al., 2011; Rousseau, 1995). To

(11)

achieve balance, both parties must be open to exchanging information regarding employee input opportunities and outcome needs (Rousseau, 2004). Relationships among employees will be characterized by shared feelings of social obligation and expectations of reciprocity in exchanges, which can be seen as the most widely recognized form of social exchange (Fiske, 1992; Cropanzano & Mitchell, 2005). The focus on reciprocity emphasizes that one person’s actions are tied to another’s over time (Molm, 2003). As a consequence,

performance appraisal and feedback will recognize not only how employees perform their own assignments but also how well they cooperate with others’ work efforts and encourages both task- and person focused helping (Mossholder et al., 2011;Spreitzer, Sutcliffe, Dutton, Sonenshein, & Grant, 2005; Bolino, Turnley, & Bloodgood, 2002).

By establishing equality matching, an organization encourages interpersonal self representation of stakeholders, which means that stakeholders not only are concerned with their personal welfare but also have an honest concern for the welfare of the other

stakeholders. In line with the equity theory (Adams, 1963), the equality matching model triggers stakeholders to compare their own contribution to contributions of others and let them focus on maintaining balance in the relationship (Stoelhorst & Bridoux, 2016; Robinson, Kraatz, & Rousseau, 1994). As a result, the interpersonal perspective will lead to a bigger social network and seemingly increases the likelihood of joint value creation (Brewer & Gardner, 1996).

Joint value creation in authority ranking

As well as the relational model equality matching the level of self-representation in authority ranking is interpersonal, meaning that stakeholders are inclined to pursue the other

participants' welfare as well as their own (Stoelhorst & Bridoux, 2016). Employment relationship is likely to follow a “noblesse oblige” morality, which means that subordinates show respect and loyalty to higher ranked employees and provide security and protection to lower ranked employees (Giessner & van Quaquebeke, 2010). Self-interested behavior is inappropriate in an authority ranking model. In this model, parties should exhibit a concern for the other’s welfare: appropriate behavior is respect, deference, loyalty, and obedience in subordinates (Fiske, 1992). Organizations with an authority ranking climate are focused on the centralization of decision-making, concentrating powers and will judge employees on following the exact procedures, overly criticizing poor work, and ensuring that employees work up to their capacity (Bass & Bass, 2008; Jago, 1982).

Pleasing their boss and leading their subordinates, if they have any, will be

employees’ priority. This focus on hierarchy however has also shown downsides and can be damaging to joint value creation, as status differentials discourage social interactions

(12)

Stoelhorst & Bridoux, 2013). While the superior is in charge of coordination in an authority ranking model, relational partners have an equal say in equality matching and organize joint value creation together. In addition, the relational partners may be less willing to share knowledge needed to coordinate actions since they perceive the other party as an equal partner. So in an equality matching, organizations can pool more expertise to coordinate value creation than an organization with a authority ranking perspective (Stoelhorst & Bridoux, 2016).These arguments lead to the expectation that the willingness for joint value creation in the relational model equality matching will be higher than the willingness to joint value creation in the relational model authority ranking.

H1b. Willingness for joint value creation is higher when the firm's relational model is equality matching rather than authority ranking

Joint value creation in market pricing

Relationships occurring in a market pricing context are largely based on means-ends

considerations. Stakeholders are guided by their personal payoffs, motivated by self-interest, and a desire to optimize personal outcomes (Vohs, Mead, & Goode, 2006). This means, the appropriate behavior is to conduct a cost-benefit analysis to obtain the highest payoffs in return for the lowest contributions (Fiske, 1992).The implied employment relationship is transactional, involving short-term relationships marked by economic inducements for prescribed contributions (Rousseau, 1995). Organizations with a market pricing climate, will judge employee activity on individual accomplishments and emphasis results-based rewards, which will lead to toward interpersonal competition for rewards, a focus to own specific work responsibilities, individual task goals and diminish noninsturmental exchanges that might stimulate cooperation (Mossholder et al., 2011; Shaw, Gupta, & Delery, 2002). In settings characterized by high task and outcome interdependence, the market pricing characteristics implies that stakeholders are only induced to contribute if they are likely to be personally rewarded for doing so or sanctioned for failing to do so (Ellemers, De Gilder, & Haslam, 2004; Haslam & Ellemers, 2005). Helping behavior will occur on an occasional basis because work is designed to enable goal accomplishment through employees’ own efforts rather than jointly with others (Mossholder et al., 2011). Therefore, it is expected that the willingness to contribute to joint value creation in the relational model authority ranking will be higher than the willingness to joint value creation in the relational model market pricing.

H1c. Willingness for joint value creation is higher when the firm's relational model is authority ranking rather than market pricing

(13)

Both in stakeholders' theory and in the team production problem, the manager, owner or leader plays a role in accomplishing the desired result (Donaldson & Preston 1995; Alchian & Demsetz 1972). In team production theory the role of the manager is set in the essence of the classical firm (Alchian & Demsetz 1972). Donaldson & Preston (1995) detracts from the definition of stakeholders (SRI, 1963, as cited in Freeman & Reed, 1983) that corporate managers must induce constructive contributions from their stakeholders to accomplish their own desired results e.g., perpetuation of the organization, profitability, stability, growth.

2.5 Leader- member exchange

Leadership plays an important role in accomplishing the desired result of an organization (Donaldson & Preston 1995; Alchian 1972). When joint value creation is the desired result, the interdependence of multiple parties and therefore the reciprocity between the different stakeholders, emphasize the role of leadership even more (Bridoux & Stoelhorst, 2016). Leadership concepts can be classified into three domains within the construct of leadership namely: leader, follower, and relationship. The theory of leader-member exchange (LMX) clearly falls into a relationship-based approach to leadership (Graen & Uhl-Bien, 1991). Research has shown that the relationship between leader and member, positively influence subordinate outcomes (Dulebohn, Bommer, Liden, Brouer, & Ferris, 2012; Hoegl &

Gemuenden, 2001; Stewart & Johnson, 2009; Woolley, Chabris, Pentland, Hashmi &

Malone, 2010), therefore this study test if the relationship between leaders and followers also effects the relationship between firms' relational model and joint value creation.

The origin of LMX theory lays in social exchange theory (Dulebohn et al., 2012, Blau, 1964). Social exchange involves a series of interactions that generate obligations

(Emerson, 1976). Within social exchange theory, these interactions are usually seen as interdependent and contingent on the actions of another person (Blau, 1964). Social exchange theory shows that employees that receives resources from one relationship can pass the resource on to another relationship (Molm, Peterson, & Takahashi, 2001).This exchange can be visible in the relationship between a leader and member. Leader-membership theory posits that effective leadership processes occur when leaders and followers in a high quality relationship and the high-quality relationship results in many benefits (Graen & Uhl-Bien, 1991). LMX theory also proposes that differential social

exchange relationships exist between leaders and their subordinates. Leaders develop high-quality exchanges with some subordinates, and these relations are characterized by trust, liking, and respect (Bauer & Green, 1996). With other subordinates, however, leaders may form lower quality exchanges that are limited to the provisions of an employment contract (Erdogan & Enders, 2007).

(14)

2.6 The moderating role of the relationship between leader and member.

The quality of the exchange relationship, developed between leaders and subordinates, contributes to subordinate outcomes (Masterson, Lewis, Goldman, & Taylor, 2000; Stewart & Johnson, 2009). According to social exchange theory, when members notice the benefits from a good relationship with their leader, they develop an obligation to reciprocate

(Gouldner, 1960). Based on the obligation to reciprocate it is expected that employees in a high LMX contribute more in joint value creation as a repayment of their obligations to the leader. As relational model theory shows, each person will experience the relationship (LMX) from the perspective of the present relational model (Fiske,1992; Haslam & Fike, 2004). As a consequence, high LMX influences the specific combination of relational models that the members of a firm adopt to structure their social interactions. This means, difference in relationship between leader and member will trigger different relational models, and in doing so galvanize joint value creation in different ways (Stoelhorst & Van Vugt 2017).

Communal sharing and leader-member exchange.

Communal sharing coincides with a leadership style promoting trust and a collectively shared identity. Leaders in the relational model communal sharing treat team members as if they are part of the same family (Stoelhorst & Van Vugt 2017). The leadership style that

characterizes communal sharing is caring and considerate and there is a high degree of concern for the team as a whole, which is more important than the welfare of individual members. The concept of transformational leadership (Bass, 1990) fits management by communal sharing, as managers attempt to transform individual members’ interests to an overarching team or firm interest, by showing leadership-by-example and through displaying a servant, self-sacrificial leadership style. The communal sharing relational model seems to be a solid foundation for building high LMX as it links perfectly to the three building blocks of LMX. (Grien & Uhl-Bien,1995). As a result, LMX in communal sharing automatically

transforms to the partnership stage, where followers can count for needed support,

encouragement, and career investments. This transformation to partnerships is accompanied by a movement among members beyond their self-interest to focus more on larger mutual interests. (Grien & Uhl-Bien,1995). In addition to the strong relation of the partnership stage, communal sharing consists of transitivity (Fiske, 1992) what means that when the leader is kind to member A, member A should ideally be kind to member B. Based on the transitivity character, the moral dimension of an equivalence (CS) relationship (Fiske, 1992), and the focus on mutual interest between the different stakeholders due to high LMX (Griean & Uhl-Bien, 1995) it is expected that high LMX should lead to more willingness to joint value creation among the members in a communal sharing relational model.

(15)

H2a LMX moderates the relationship between firm’s relational model communal sharing and joint value creation such that the relationship between the relational model communal sharing and joint value creation is stronger when the LMX is high than when LMX is low.

Equality matching and leader-member exchange.

The relational model equality matching is strongly associated with the communal sharing factors (Fiske, 1999), therefore it is expected that the transitivity that is visible in communal sharing is also visible in equality matching. Further, stakeholders' perception in the relational model equality matching is based upon equality (Fiske 1992, 2012; Haslam, 1999, 2004) and the tit-for-tat mechanism. This will also be visible in the leadership style, thus leaders treat members that contribute more to the team by rewards, whereas individuals who are under-contributing can expect punishment (Stoelhorst & Van Vugt, 2017). In this mechanism fairness and balance in the exchange of resources plays a big part in the creation of a collaborative organization (Mosholders et al., 2011). Different justice perceptions by stakeholders, caused by the quality of LMX, are supported by equity theory (Adams, 1963) and referent cognitions theory (RCT; Cropanzano & Folger, 1989; Dulebohn et al., 2012). High-quality LMX relationships will lead to procedural and distributive justice perceptions (Erdogan & Bauer, 2010), while in contrast, low-quality LMX relationships may lead to perceptions of injustice in process and outcomes (Dulebohn et al., 2012). When employees determine that they have been treated well in specific helping exchanges with others, they develop fairness perceptions about particular partners that influence future exchanges at the entity level, under which effective helping (Bolino, Turnley & Bloodgood, 2002). In the same way, it is expected that the perception of justice, caused by the high LMX, will lead to members that have more willingness to contribute in joint value creation.

H2b LMX moderates the relationship between firm’s relational model equality matching and joint value creation such that the relationship between equality matching and joint value

creation will be stronger when the LMX is high than when LMX is low. Authority Ranking and leader-member exchange.

The relational model authority ranking consists of an attitude of respect deference, loyalty, and obedience by subordinates. In this relational model managers have power over team members, but it comes with responsibilities in terms of guidance and protection, and the obligation to protect the interests of the team (Fiske, 1992; Giessner & Quaquebeke, 2010). The relational model authority ranking is characterized by an authoritarian, paternalistic leadership style (Stoelhorst & Van Vugt 2017) which is linked to positive outcomes like increase in productivity (Shaw, 1955) and accurate solutions (Cammalleri et al., 1973) , but is also linked with leaders high in machiavellianism (Kiazad, Restubog, Zagenczyk, Kiewitz &

(16)

Tang, 2010) and associated with the subordinate perceptions of abusive supervision (Aryee, Chen, Sun & Debrah, 2007). However, the relational model authority ranking does mean that the leader should be a powerful authoritarian dictatorship. Instead, the leader must be seen as charismatic (cf. Weber, 1978) and as a positive figure (Giessner & Quaquebeke, 2010). This means that leaders should be role models and guide their subordinates (Brown, Treviño, & Harrison, 2005; Brown & Trevino, 2006).

Becoming the role model can be achieved by developing high LMX which contains the three dimensions, respect, trust, and obligation (Grien & Uhl-Bien, 1995). Trust for example, is one of the intangible benefits that a high LMX provides (Bauer & Green, 1996) and this results in subordinates that repay intangible benefits via high performance

(Masterson, Lewis, Goldman, & Taylor, 2000). Also, high-quality LMX relationships can expand safety citizenship role definitions (Hofmann, Morgeson & Gerras, 2003). This results in the expectation that when subordinates trust their leader and experience safety, they are more likely to follow, feel an obligation to reciprocate (Gouldner, 1960), and perform

according the expected output, in case of this research joint value creation.

H2c LMX moderates the relationship between firm’s relational model authority ranking and joint value creation such that the relationship between the relational model authority ranking

and joint value creation will be stronger when the LMX is high than when LMX is low.

Market Pricing and leader-member exchange.

Leadership in the relational model market pricing shows the most comparison to a

transactional leadership style, as it is characterized by a system of contingent rewards and punishment for individual worker performances (Bass, 1990). Here, leadership operates through a negotiated contract between the firm and its workers in which the tasks and roles of each team member are clearly laid out; as are the incentives they receive in return (Stoelhorst & Van Vugt 2017). Stakeholders' perspectives and motives are focused on the satisfaction of his/her own self- interest, without considering the common goal of the organization. The relational model is based around a transaction, where the parties

exchange substantive items (Fiske, 1992). Close social interactions occur less often because stakeholders are more concerned with their own work goals and responsibilities rather than those of others. As such, judgments about the willingness to joint value creation are likely to be tied to the event level and be focused on the concern of help-givers if they receive an adequate return on invested helping behavior (Mossholder et al., 2011). Interactions with the leader are also seen as strictly contractual which fits the description of a lower-quality LMX relationship (Colella & Varma, 2001; Dansereau, Graen, & Haga, 1975; Graen & Schiemann, 1978; Vecchio, 1982). In these situations, the leadership based process is essentially

(17)

non-existent (Graen & Uhl-Bien, 1995), therefore it is expected that the willingness to joint value creation within a firm that perceives the relational model market pricing, is not influenced by the relation between the leader and member.

H2d The relationship between firms’ relational model market pricing and joint value creation is independent of LMX.

(18)

3. Method

3.1 Procedure

In collection data I worked together with my fellow student Maris Keijser. Data were collected via the online survey provider Qualatrics. The goal was to obtain data on a convenience sampling method. Respondents were reached in two ways. The first group were found through personal contacts and acquaintances of the research team. The second group was acquired via the research bureau PanelClix. All respondents were recruited from the

Netherlands. Therefore, the study material was provided in Dutch. The data was collected within a time frame of six weeks.

To perform this research a vignette (fictive descriptions) was used (Rossi and Anderson, 1982). A vignette study allows manipulating the study factors of interest, in this case the relational models and LMX, and comparing the core relationships in these experimental conditions with the control condition. Using a vignette can be seen as a quasi-experimental research design and provides advantages in comparison with traditional social survey research, as for example reducing social desirability bias (Alexander & Becker, 1978; Walllander, 2009).

The study material contained an introduction in which the purpose of the study, the length of the study and the price that could be won. This introduced was followed by a measure capturing social value orientation (SVO). After the questions concerning the SVO each respondent had to read a randomly assigned company description capturing the firm's relational model. The four company descriptions, one description per relational model, were presented roughly equal number of times across all respondents. Participants were asked to imagine them working for the described company and answer all questions in the survey from the perspective of an employee working for the described company. After the description of the company, all the participants were presented with the same short story relating a request for cooperation form a colleague. This description was retrieved from the research of Stoelhorst & Bridoux (2013). The experimental group was then either assigned to a high or low LMX condition. The control group did not receive any information regarding the amount of LMX with their leader.

Participants were then asked to answer four questions regarding their willingness to contribute to joint value creation. At the end of the questionnaire there were a number of control questions. It took approximately 15 minutes to complete the survey in the

(19)

All the original items of the questionnaires used in this research were translated from English to Dutch and then backtranslated from Dutch to English following the widely used translation- back translation procedure by Brislin (1970).

3.2 Sample

The total sample size consisted of 472 respondents. 43,2 % of the respondents were

female, and 56,8 % were male. 77,5% of the respondents are currently employed. Data was collected in the Netherlands. Education levels varied, with 11,4% having received a high school education, 27,3.% with a lower vocational education (MBO), 42,4% with a higher vocational education (HBO), 8,3% with a university Bachelor degree, 10,4% with a university Master degree, and with the remaining 0,2% with other education level. The average age was 42,2 years (SD = 14,477); on average respondents had worked 19,365 years (SD = 13,767) of work experience. 10,7 % of the respondent are students.

The control group consisted of 194 respondents. Of these, 111 (57,2%) were males and 83 (42,8%) were females. 82% of the respondents in the control group are currently employed. Education levels varied, with 8,8% having received a high school education, 25,8% with a lower vocational education (MBO), 48,5% with a higher vocational education (HBO), 7,2% with a university Bachelor degree, 9,8% with a university Master degree. The average age was 43,27 years (SD = 14,139); on average respondents had 20,732 years (SD = 13,9545) of work experience. 8,3 % of the respondent are students. In table 1, the distribution of the participants over the different relational models is provided.

N %

Control group Communal sharing 52 26,8

Equality machting 49 25,3

Authorithy ranking 44 22,7

Market pricing 49 25,3

Total: 194 100

(20)

The experimental group consisted of 278 respondents. Of these, 157 (56,5%) were males and 121 (43,5%) were females. 75% of the respondents are currently employed. Education levels varied, with 13,3% having received a high school education, 28,4% with a lower vocational education (MBO), 38,1% with a higher vocational education (HBO), 9% with a university Bachelor degree, 10,8% with a university Master degree and with the remaining 0,4% with other education level. The average age was 41,05 years (SD = 14,940); on average respondents had 18,037 years (SD = 13,8724) of work experience. 12,3 % of the respondent are students. The experimental group consisted of two LMX conditions (high / low), and there are four priming conditions (communal sharing / equality matching / authority ranked / market pricing ). In table 2, the distribution of the participants over the different experimental conditions is provided.

N %

Experimental group Communal sharing high LMX 33 11,9

Communal sharing low LMX 40 14,4

Equality machting high LMX 32 11,5

Equality machting low LMX 36 12,9

Authorithy ranking high LMX 36 12,9

Authorithy ranking low LMX 37 13,3

Market pricing high LMX 33 11,9

Market pricing low LMX 31 11,2

Total: 278 100

Table 2. Descriptives: experimental conditions

3.3 Measurements Firm’s relational model.

To describe the firms' relational model, this research used the four different relational models of Fiske (1992). The company descriptions that were used come from empirical research conducted by Stoelhorst & Bridoux (2013) and the addition of the items 1 (CS, EM, AR, MP), 3 (CS, AR), 4 (EM), and 5 (AR, EM) used in the factor analyses of Haslem and Fiske's (1999).

To check whether the respondents were able to distinguish the four different relational models the respondents got one control question in which respondents were asked to select one out of four descriptions of the firm's relational model that most fitted the earlier described organisation. There were four answers in which each relational model was described through the used items of Haslem and Fiske's (1999) factor analyses.

(21)

Joint value creation.

Participants' willingness to contribute to joint value creation was measured through their response to the request to help a colleague as presented earlier. After the description the test subject were asked to answer four questions, which were retrieved from the research of Stoelhorst & Bridoux (2013), about their willingness to help their colleague. ‘How likely would you be to help John?’, ‘How motivated would you be to help John?’ ‘How justified would you feel in refusing to help John?’ (reverse-coded) ‘How likely would you be to look for an excuse not to help John?’ (reverse-coded) Respondents were asked to answer each question on a seven-point Likert scale ranging from 1("strongly agree") to 7 ("strongly disagree").

Cronbach’s α was 0,589. Deleting the third item How justified would you feel that refusing to help John would be?’ Cronbach’s α would increase to 0,668. In the analyse of the data the third item is deleted to increase the internal consistency.

Leader-Member exchange

Manipulation of experimental variable LMX was done through a vignette consisting of two levels of LMX (high; low). Through vignettes, the leader-member relationship can

successfully be manipulated with a high degree of experimental control as earlier

researchers have shown (e.g., Ansari & Kapoor, 1987; Fu & Yukl, 2000). Of all the scales that were available to measure the quality of LMX, two have been used frequently. One is the LMX-7 (Graen, Novak, & Sommerkamp, 1982) and another is LMX-MDM (Liden & Maslyn, 1998). This study used a description of high and low quality on the base of the combination of both scales as it used the description of Hunter, Ansari and Jayasingam (2013) that contained the LMX-MDM, and added the items of the LMX-7.

Only the respondents in the experimental group got the LMX manipulation. To control whether the participants in the experimental groups distinguished the quality of LMX

correctly, they were asked to answer the revised 7-items scale of Griean and Uhl-Bien (1995). This scale is quite similar to the LMX-7 (Graen, Novak, & Sommerkamp, 1982) with the major difference being a change in response categories (from four to five categories and revised scale anchors). The seven items characterize various aspects of the leader-follower workplace relationship (Law, Hui, & Chun, 2010). For six questions, responses were ranging from 1 (totally disagree) to 5 (totally agree) on a 5-point scale. The seventh item addressed the quality of the relationship with the question “How would you characterize your working relationship with your leader?” The response categories for this question were ranging from “Extremely ineffective” to “Extremely effective” on a 5-point scale. Cronbach’s α of the scale was 0,96.

(22)

Control variables.

The first control question was controlled for person's social value orientation (SVO). SVO was measured using the method developed by Murphy, Ackermann and Handgraaf (2011). This method involves presenting the test subjects with six hypothetical situations in which they have to divide an imaginary sum of money up between them and another person. SVO is included as it has been argued that pro-socials have a default setting that is focused on willingness to cooperate and in contrary; pro-selfs have a default setting that is focused on pursuing self-interest (Bogaert, Boone & Declerck, 2008). The other control variables included were gender (1 = ‘Male’, 2 = ‘Female’), age, education level, currently studying, current employment, and work experience. Gender was added as it has been proved that gender matters when it comes to cooperation and the motivation to cooperate (Eagly & Crowley, 1986; Vugt, Cremer, & Janssen, 2007). Age can also be seen as a determining factor for cooperation, as older men tend to be more cooperative then younger men (Van den Assem, Van Dolder & Thaler, 2012; Wagner, 1995). Additionally it has been shown that both age and gender may influence the LMX relationship (Bernerth et al., 2007; McColl-Kennedy & Anderson, 2005; Turban & Jones, 1988). The fourth and fifth control variables used were educational attainment education level (1 = ‘High School’, 2 = ‘Lower vocational education (MBO), 3 = ‘Higher vocational education’ (HBO), 4 = ‘University bachelor degree’ (WO), 5 = University master degree or higher’, 6 = ‘Other’) and currently studying (1 = ‘Yes’, 2 = ‘No’) because, Wagner (1995) found a significant effect between American students’ GPA scores and levels of cooperation. At last, current employment (1 = ‘Yes’, 2 = ‘No’), and work

experience were added as control variables as they can play an important role in the quality of leader–member exchanges and related work outcomes (Epitropaki & Martin,1999; Tsui, Egan & O’Reily; 1992). An overview of the study material sections used in this research can be found in Appendix A.

3.4 Manipulation checks

In order to check whether the respondents recognized the different relational models the questionnaire contained a control question. Table 3 shows the results of the comparison between the relational model that was presented and the control question. The majority of the respondents in the control group recognised the relational models communal sharing and authority ranked. However, respondents had difficulty to withdraw the relational models equality matching and market pricing. Correlated relational models could tend to be confused which is recognized by Haslam & Fiske (1999) as they state that: "This problem is pertinent to the distinction between the Communal Sharing and Equality Matching models. These models are often conflated so that egalitarian reciprocity is identified with close communal relationships and both are treated as aspects of a single dimension" (p. 247).

(23)

The results of the experimental group are in line with the control group, whereas the majority recognized the relational model communal sharing. In the relational models equality

matching and authority ranking, high LMX causes a shift towards communal sharing. In contrary, low LMX in the relational model authority ranking causes high recognition with the presented relational model. Considering the recognition of Haslam & Fiske (1999) about the comparison between items of the relational models, and in order to prevent group differences by maintaining comparable groups, all of the respondents are taking into account in

analysing the results.

In order to check whether the respondents could recognize and experience the description of the relationship between the leader and member, the questionnaire contained a manipulation check for the LMX. Participants in high LMX condition reported a better relationship with the leader (M = 3,793; SD = 0.53; N = 127) compared to those with low LMX (M = 2,245; SD = 0.88; N = 141). The results indicated that the experimental manipulation of LMX was successful. Control Question CS EM AR MP No answer % Correct Control group Communal sharing 34 5 7 6 0 65,0 Equality machting 22 15 2 4 6 30,6 Authorithy ranking 15 1 23 4 1 52,3 Market pricing 10 10 7 21 1 42,9 Experimental group

Communal sharing high LMX 21 4 5 3 0 63,6

Communal sharing low LMX 26 4 5 2 3 65,0

Equality machting high LMX 13 11 6 2 0 34,4

Equality machting low LMX 7 18 4 3 4 50,0

Authorithy ranking high LMX 11 7 13 3 2 36,1

Authorithy ranking low LMX 5 5 23 4 0 62,2

Market pricing high LMX 12 1 4 14 2 42,4

Market pricing low LMX 3 3 7 17 1 54,8

Table 3. Results control question relational models

(24)

4. Results

The data obtained from the questionnaires were analyzed using IBM SPSS Statistics version 23. Means, standard deviations, and correlations are reported in Table 4.

There are small indications for a correlation between LMX and joint value creation which is significant and can be described as a small-sized relationship (.147*).

4.1 Hypothesis test

Hypothesis 1a-c states that the relational models of a firm influence the willingness to value creation of its employees. It was expected that there is a difference of joint value creation between the relational models. The hypotheses 1a-1c were tested using one way ANOVA. The results show there was a statistically significant effect of relational models on levels of joint value creation, F(3, 189) = 3.415, p < .05.

In hypothesis 1a, it was expected that when the firm adopts communal sharing, the

willingness to contribute to joint value creation is higher than when the firm adopts one of the other three relational models, equality matching, authority ranking or market pricing. Tukey post-hoc tests revealed that there was no statistically significant difference in willingness to joint value creation in the comparison of relational model communal sharing with the relational model EM (p = 1.000) , or the relational model authority ranking (p = .998). In comparison to the relational model market pricing there was a significant difference (p = .047). Based on these results, hypothesis 1a was not supported.

In hypothesis 1b, it was expected that when the firm adopts equality matching, the willingness to contribute to joint value creation is higher than when the firm adopts the relational model authority ranking. Tukey post-hoc tests revealed that there was no statistically significant difference in willingness to joint value creation in the comparison of relational model equality matching with the relational model authority ranking (p = .996) or the relational model market pricing (p = .059). Hereby, hypothesis 1b was also not supported

In hypothesis 1c, it was expected that when the firm adopts authority ranking, the willingness to contribute to joint value creation is higher than when the firm adopts the relational model market pricing. Tukey post-hoc tests revealed that the joint value creation was significantly higher in the relational model authority ranking compared to the relational model market pricing (p = .042) Therefore, hypothesis 1c was supported.

(25)

Variables M SD 1 2 3 4 5 6 7 8 9 1. Age 41.96 14.642 (-) 2. Gender 1.43 .496 -.076 (-) 3. Working experience 19.145 13.954 .913** -.124** (-) 4. Student 1.89 .309 .400** -.075 .317** (-) 5. Employment 1.22 .416 .309** -.020 .204** -.071 (-)

6. Social value orientation 25.4497 14.939 -.069 .085 -.069 -.036 -.054 (-)

7. Relational model 2.46 1.122 -.026 .020 -.011 -.005 -.028 -.053 (-)

8. Leader-member exchange (manipulation) 2.9815 1.091 .009 -.038 .002 .008 .031 .007 -.030 (-)

9. Joint value creation 5.0936 1.165 .194** -.007 .212** .169** .014 .131** -.080 .147* (-)

Table 4. Means, standard deviations, and first order correlations of the study variables Statistical significance: *p<.05; ** p<.01

(26)

4.2 Moderation effect

Hypotheses 2a-d were tested using factorial ANOVA. Within the experimental group there was a non-significant main effect of the relational models on willingness to joint value creation, F (3, 269) = 1,854, p = .138, η² = .02. The Tukey post-hoc tests revealed that the willingness to joint value creation did not show significantly differences when the relational models were compared with each other. There was a significant main effect of LMX on the willingness for joint value creation F (1, 269) = 6,397, p < .05, η² = .023.

There was a non-significant interaction effect between relational models and LMX on the willingness to joint value creation F (3, 269) = 0,725, p =.538, η² = .008. Table 5 contains the means and standard deviations of the willingness to contribute in joint value creation of all groups.

Willingness to contribute in joint value creation

Relational model M SD Control group Communal sharing 5,237 1,291 Equality machting 5,224 1,253 Authorithy ranking 5,279 1,138 Market pricing 4,619 1,032

Experimental group - High LMX

Communal sharing 5,354 0,195

Equality machting 5,240 0,198

Authorithy ranking 5,352 0,187

Market pricing 5,131 0,195

Experimental group - Low LMX

Communal sharing 4,933 0,177

Equality machting 4,676 0,190

Authorithy ranking 5,324 0,184

Market pricing 4,774 0,201

Table 5. Means and standard deviation of willingness to contribute in joint value creation.

In hypothesis 2a, it was expected that LMX will moderate the relationship between firm’s relational model communal sharing and joint value creation such that the relationship

between the relational model and joint value creation will be stronger when LMX is high than when LMX is low. The mean difference within the relational model communal sharing is 0,420 which indicate a positive effect of high LMX. This difference however is not significant (p =.112) Therefore hypothesis 2a was not supported.

(27)

In hypothesis 2b, it was expected that LMX will moderate the relationship between firm’s relational model equality matching and joint value creation such that relationship between the relational model and joint value creation will be stronger when the LMX is high than when LMX is low. The mean difference within the model of equality matching is 0,563 which indicate that the willingness to joint value creation is higher when LMX is high. This difference is significant (p =.041) Therefore hypothesis 2b is supported.

In hypothesis 2c, it was expected that LMX will moderate the relationship between firm’s relational model authority ranking and joint value creation such that relationship between the relational model and joint value creation will be stronger when the LMX is high than when LMX is low. The mean difference within the relational model authority ranking is 0,028 which indicate that the willingness to joint value creation slightly higher when LMX is high. This difference however is not significant (p =.917) Therefore hypothesis 2c is not supported. Finally, in hypothesis 2d, it was expected that LMX has no influence on the relationship between firm’s relational model market pricing and joint value creation. The mean difference within the model of market pricing is 0,357 and indicates that there is a difference in the willingness to joint value creation when LMX is high. This difference however is not significant (p =.204) Therefore hypothesis 2d is supported.

Figure 2. Means joint value creation - experimental group. 4,600 4,700 4,800 4,900 5,000 5,100 5,200 5,300 5,400 Communal sharing Equality machting Authorithy ranking Market pricing Jo in t va lu e c re at io n Relational Model High LMX Low LMX

(28)

5. Discussion

Fiske's (1992) relational model theory was used to examine the relationships between firms' relational model and joint value creation, and LMX theory was used to examine the

moderation effect between the firms' relational model and joint value creation. Partial support for the hypothesized relationships was found, with authority ranking having a more positive influence on the willingness to joint value creation in comparison to the relational model market pricing, and LMX moderating the relationship between firms relational model equality matching and joint value creation. An analysis of result implications follows, along with a discussion of study limitations and proposed future directions.

5.1 Theoretical Implications

This research contributes to the literature in two ways. First, it supports the theory of Bridoux and Stoelhorst (2013, 2016) which in line with stakeholders theory (Freeman & Reed, 1983) and the perspective of Ostrom (2000) let go of the traditional assumption of economic theory that all human behavior is exclusively motivated by self-interest and replaces this assumption with a social welfare perspective. Bridoux and Stoelhorst (2013, 2016) claim that only a minority of stakeholders are predisposed to focus exclusively on their self-interest. The present research indeed shows that the relational models communal sharing, equality matching and authority ranking all show higher willingness to contribute to joint value

creation in comparison to the relational model market pricing. Bridoux and Stoelhorst (2013, 2016) also nuance the traditional view of stakeholder management in terms of two broadly defined “transactional” and “relational” approaches and replace these two approaches with the four relational model theory of Fiske (1992). Specifically, the relational model market pricing triggers behaviors that transcend the self-interest of the stakeholders (Stoelhorst & Bridoux 2016) and therefore shows a relatively straightforward line that refers to the traditional assumption of economic theory and it negative consequences, under which the willingness to help colleagues (Ghoshal & Moran, 1996). The other three relational models bring new insights and do not show a direct connection with the traditional economic perspective. The present study support the perspective of Bridoux and Stoelhorst (2013, 2016) that each of the four relational models in organization differ in their influence on joint value creation as the results show a significant effect of relational models on levels of joint value creation. In addition to the support for the theory this study also expected that there would be a sequence between the four relational models, were communal sharing relates to the highest willingness to joint value creation, followed by equality matching, authority ranking and market pricing. However, the results of this research did not support the

expected sequence. This is unexpected given that each relational model represents a distinct mental model (Fiske, 1999). Nonetheless, this research found support for the expectation

(29)

that the willingness to joint value creation is higher in the firms relational model authority ranking than in the firms relational model market pricing. This is in line with the research of Mossholder et al. (2011) who show that helping behavior will occur on an occasional level in market pricing organization.

Secondly, this research contributes to the literature by linking the relationship based approach of leadership (Graen & Uhl-Bien, 1991) to joint value creation, in light of the four relational models (Fiske, 1999). In line with former research (Stewart & Johnson, 2009; Gouldner 1960) this study supported the positive influence of LMX, as the results show that a good dyadic relationship between leader and member can have benefits for the cooperation between others outside the dyadic relationship. The manipulation check of this study shows that respondents who are presented a good relationship with their leader, tend to see the relational model as communal sharing, while the description of the organization perceives the relational model equality matching or authority ranking. This can indicate that high LMX has comparison with the relational communal sharing in the perspective of stakeholders. In contrary, low LMX in the relational model authority ranking causes high recognition with the presented relational model, which points to the relationship between authority and low LMX (Connelly & Kelloway, 2003; Ouchi, 1980). In addition to the main effect of LMX this research tested the moderation effect of LMX in each of the relational models. First of all, it was

expected that the good relationship between leader and member would positively influence the relationship between the firms relational model communal sharing and willingness to joint value creation. The results showed however no significant impact of LMX on the relationship between communal sharing and joint value creation. A possible explanation for the lack of significant impact of LMX in the relational model communal sharing can be found in the foundation of communal sharing. People who have a communal sharing relationship are expected to show a high degree of commitment, even altruism, towards each other and the firm. This high degree of commitment could lead to voluntary contributions to the

organization beyond the members' contractual tasks (OCBs) (Stoelhorst & Van Vugt 2017). The common expectation of voluntary contributions, under which joint value creation, can be an explanation of the non-significant impact of LMX in the relational model communal

sharing, consequently the effect of LMX losses it power because people within the model communal sharing are willing to contribute to joint value creation despite the relationship with their leader.

As expected, the effect of a good relationship between leader and member made a significant difference in the relational model equality matching. As the results reveal, LMX shows the most influence in the relational model equality matching. This is in line with the research of Bauer and Green (1996) and Masterson et al. (2000) as they show that a trust

(30)

based relationship one of the intangible benefits of high-quality LMX is, and this trust stimulates members to repay the benefits via high performance.

Another interesting result is the lack of influence of LMX on the relationship between the relational model authority ranking and joint value creation. A possible explanation can be found in the statement of Stoelhorst and Van Vugt 2017 (cf. Henrich & Gil-White, 2001) that authority ranking is characterized by an authoritarian, paternalistic leadership style, but one that is based on prestige rather than coercion. In this reasoning the results confirm that the willingness to contribute is high in the relational modal authority ranking but is not influenced by the relationship with the leader because the joint value creation is assumingly on the basis of on obligation towards other stakeholder. The lack of significance impact of LMX on the relationship between the relational model market pricing and joint value creation was expected seeing the similarities between the relational modal market pricing and the "transactional" behavior in the classical economic theory. The results of this study confirm this expectation as the LMX has no significant impact on the willingness to joint value creation in the relational model market pricing.

To sum up, this study adds value to the literature of LMX by showing that the effect of LMX is different in each relational model. It shows that a contingency perspective is needed to understand the effects of LMX.

Both contributions have an impact on stakeholder' theory and the literature of team production. By perceiving a relational model other than market pricing, organizations can increase social welfare and avoid the public good dilemma. As a result of avoiding the relational model market pricing, stakeholders seem to become more open for joint value creation. Whereas the firms' influence through the perceived relational model can be seen as prevention measures for the problem of team production (Bridoux and Stoelhorst, 2016) the addition of LMX theory can be seen as a possible solution for overcoming the self-interested behavior. This study shows in which cases organizations can overcome team production problem by LMX (EM) and in which cases the solutions lay outside LMX (CS, AR and MP) Both conclusions add value to the literature of team production problem, as it can change the perspective about team production and provide a new starting point in avoiding and

overcoming team production related problems like shirking.

5.2 Practical Implications

The results of this study suggests that firms can increase joint value creation not by awards or financial benefits (e.g.,Bowles, 2008), but increase social welfare by avoiding the

stakeholder's self-interest that is salient in the relational model market pricing and, instead, consistently apply the interpersonal identities that are salient in authority ranking and equality matching or the collective identity salient in communal sharing. Relational models can be

(31)

influenced by organizational structures and culture (Connelley & Folger, 2004; Fiske & Tetlock, 1997). For instance, in an organization with distinct structure, authority ranking relationships are likely to arise (Giessner & van Quaquebeke, 2010).To foster authority ranking relationships, managerial practices should aim at building and preserving the legitimacy of the superior(s) in stakeholders’ eyes. The social psychology literature has stressed that superiors are viewed as more legitimate if they exercise authority through procedures that people experience as fair (Tyler, 2006). In an organization in which leaders structure tasks via explicit rules of task goal and accomplishment, market pricing

relationships are likely to develop (Giessner & van Quaquebeke, 2010). The research field of social identity provides examples to help create an communal sharing organisation

(Stoelhorst & Bridoux, 2016) , like providing permanent employment, socializing newcomers with an emphasis on the common identity, emphasizing members’ commonalities, and using words like “we” and “us” (rather than “you” and “I”) and phrases like “we are a family”

(Ashforth & Johnson, as cited in Stoelhorst & Bridoux, 2016 ). Collaborative HR system (Lepak & Snell ,1999) can help to create an organization that perceives equality matching relationships. (Mossholder et al., 2011)

Aoki (As cited in Donaldson & Preston, 1995) recognized only investors and employees as significant stakeholders and saw managers as essentially "referees" between these two stakeholder groups. In line with Williamson (1985) this study emphasizes the role of managers in creating social welfare and underpins the responsibility of the manager to account for the interest of all stakeholders. This study show that in attempting to enlarge productivity by joint value creation, a leader should consider which relational model is present in the firm. The results show that leaders in organizations with an equality matching relational model should invest in the relationship with the member. This is in line with the arguments of Graen and Uhl-Bien (1995) who state that, managers should be supported and trained to build high-quality relationships with all their subordinates as it is beneficial for all stakeholders. Moreover, when a high-quality relationship in an equality matching

organization is met, it can reduce team production problems. This positive effect of LMX is particularly important in those jobs that involve strong elements of team production, where output reflects the contribution of individuals, and individual contributions cannot be easily identified (Alchain & Demtez, 1972).

The results of this study only show signs of a significant impact of LMX on the relation between equality matching and joint value creation, this provides managers a reason to become aware of the relational model of the firm. When the leader becomes aware of the relational model of the firm, he/she can consistently apply the right focus on leadership perspective and prevent a mismatch, as for example by making a special request to team

Firms' relational model and joint value creation: the moderating role of leader-member exchange

Statement of originality

Abstract

Table of contents

1. Introduction.

2. Theoretical background and hypotheses

3. Method

4. Results

5. Discussion