The Influence of Generalized Trust on
Customer–Seller Relationships
University of Amsterdam
Faculty of Economics and Business Master Thesis MSc Business Studies Academic year: 2013-2014
Name: Henk Gerrits
Student number: 10150757 1st Supervisor: Dr. K. Venetis
2nd Supervisor: Drs. Ing. A.C.J. Meulemans
Preface
This thesis reflects my last work as a student at the University of Amsterdam. First of all, I would like to thank my supervisor Dr. Karin Venetis for her guidance, enthusiasm and motivation during the entire research process. Furthermore, my special thanks go to my wife and children for their perseverance and their constant support. Moreover, I would like to thank Maarten Egberink for his enormous drive, patience and enthusiasm. Finally, many thanks to the people who participated in my survey!
Hardenberg, April 4th, 2014
Table of content
Preface ... 1
Abstract ... 4
1. Introduction ... 5
1.1 Contribution to the literature ... 6
1.2 Structure of the research ... 6
2. Literature review ... 8
2.1 Loyalty ... 8
2.1.1 What is loyalty? ... 8
2.1.2 Why is loyalty important? ... 9
2.1.3 Consequences of loyalty ... 10
2.1.4 Drivers of loyalty ... 11
2.1.5 The relationship between satisfaction and loyalty ... 11
2.1.6 The relationship between trust and loyalty ... 12
2.2 Trust ... 13
2.2.1 What is trust? ... 13
2.2.2 Dimensions of trust ... 15
2.2.3 The multi levelled nature of trust ... 15
2.3 Firm Trust ... 18
2.2.1 The importance of firm trust ... 18
2.3.2 The relationship between firm trust and loyalty ... 19
2.4 Interpersonal Trust ... 20
2.4.1 The importance of interpersonal trust ... 20
2.4.2 The relationship between interpersonal trust and loyalty ... 21
2.5 The relationship between interpersonal trust and firm trust ... 22
2.5.1 Interdependency of interpersonal trust and firm trust ... 22
2.5.2 Transferability of trust ... 23
2.5.3 The contribution of firm trust and interpersonal trust on loyalty ... 24
2.6 Generalized Trust ... 26
2.6.1 System trust and generalized trust ... 26
2.7 The moderating effects of generalized trust ... 29
3. Research model ... 31
4. Research Methods ... 32
4.1. Overview of data collection ... 32
4.2. Procedure ... 33
4.2.1 Translation- back translation procedure ... 34
4.2.2 Pilot testing ... 34 4.3 Data analysis ... 35 4.4. Participants ... 36 5. Measurement ... 38 6. Results ... 40 6.1.1 Descriptive statistics ... 40 6.1.2 Correlation analysis ... 42
6.1.3. Results regression analysis ... 43
7 Discussion ... 48
8. Implications for practice ... 52
9. Limitations and directions for future research ... 54
References ... 56
Appendix 1. Survey (Dutch version) ... 65
Abstract
This study aimed to examine the influence of generalized trust on the relationship between firm trust and loyalty and between interpersonal trust and loyalty separately. It is found that consumers perceive a higher trust when the nature of trust becomes narrower, from industry to firm to person. In the distrusted financial industry this effect is more evident than in the more trusted hospital industry. With respect to loyalty it is found that generalized trust has an indirect influence on loyalty via firm trust and interpersonal trust. This influence does not demonstrate an evident divergence between the importance of firm trust and interpersonal trust on loyalty in general. However, zooming in different industries the influence of
generalized trust appears to be different. This finding contributes to recover the low trusted environment some industries (e.g. the financial industry) are in. This study concludes that achieving a higher level of loyalty in a distrusted environment starts with improving the corporate reputation at the level of the firm. This does not absolve companies to invest in human capital simultaneously in the background. Managers should emphasize the role of the interpersonal trust in their educating programs.
Keywords Trust, Generalized Trust, Firm Trust, Interpersonal Trust, Loyalty
“Building trust requires talking and thinking about trust” (Robert C. Solomon)
1. Introduction
For service related firms it is important to know to what extent loyalty is formed by firm trust or by interpersonal trust. A firm that disposes of limited resources and managers constantly has to make tradeoffs in investing in reputation, brand equity or in human resources. What level of trust is most important to acquire the loyalty of your customer? Loyalty is identified as a strong driver of profitability and competitiveness and contributes directly to this most important firm goal. A higher level of loyalty leads to repurchase intentions,
recommendations to other potential customers and a reduced price sensitivity. Loyal customers are also less sensitive to a negative word of mouth about the firm. These
behaviours can be achieved by developing long term mutually profitable relationships with customers. Trust is along with satisfaction a key ingredient for a good functioning relationship between customer and seller.
Since the outbreak of the financial crisis in 2008 the image of the financial industry has radically changed. The level of trust in the financial industry as a whole decreased
substantially in a short time. The level of trust concerning the environment of an industry is called generalized trust. Beyond the crisis of confidence at firm and personal level the financial crisis influenced trust in the financial industry at the level of generalized trust. An industry that is based on trust will require a huge effort to recover the trust that it has lost. Recovering trust requires more understanding of the influence of generalized trust on the firm
1.1 Contribution to the literature
The relationships between trust, firm trust, interpersonal trust and loyalty have been extensively studied. The influence of generalized trust on loyalty has not been much
investigated yet. The extremely negative impact on and consequences for generalized trust in the financial industry as a result of the financial crisis provide a motive to investigate the role of generalized trust on loyalty. Hansen (2012) found that the influence of narrow-scope trust on loyalty is negatively moderated by generalized trust. In a low trusted environment the importance of narrow scope trust on loyalty increases compared to high trusted environments. Hansen (2012) does not determine what the effect is of generalized trust on firm trust and interpersonal trust separately. This study contributes to literature by investigating what trust level (firm or interpersonal trust) has most influence. Secondly, this research is designed to investigate if there are different effects between industries acting in high and low trusted environments. Therefore the research questions of this study are as follows:
What are the effects of firm trust on loyalty when the level of generalized trust changes.
What are the effects of interpersonal trust on loyalty when the level of generalized trust changes.
How does the relative importance of both levels of trust react when the level of generalized trust changes.
1.2 Structure of the research
In the remainder of this paper, the theoretical background to loyalty, the multi levelled trust and the effects of generalized trust are given in the section “Literature review”. Based on that, hypotheses are formulated and the model employed in this study is visually exposed. The following section includes the “Research Methods” and afterwards the “Measurements are
presented. The outcomes of the research are provided in the section “Results” where the hypotheses are tested. This part is followed by the section “Discussion” where the results will be interpreted. Finally, the “Implications for practice”, the “Limitations of this study and the directions for future research" will be pointed out.
2. Literature review
In the literature review the theoretical context of this study is elaborated. The first 2.1
demonstrates the definitions, the importance, consequences and drivers of loyalty. Moreover it exposes the relationship between loyalty and trust. It is followed by the definitions,
dimensions and the different levels of trust in section 2.2. The importance of firm trust and interpersonal trust and their relationships with loyalty are to be found in section 2.3 and 2.4. Section 2.5 contains the complicated relationship between interpersonal trust and firm trust. The third level of trust, generalized trust and its relationships with firm trust and interpersonal trust are elaborated in section 2.6. Last but probably most important for this study the
moderating effect of generalized trust on interpersonal trust and firm trust is elaborated in section 2.7. There are formulated four hypotheses to test.
2.1 Loyalty
This study is about the effects of different levels of trust on loyalty. In this section an insight is given in the definition, the importance, the drivers and consequences of loyalty.
2.1.1 What is loyalty?
In academic research customer loyalty has been a subject of many studies. There is not a universal definition of customer loyalty. A common definition is: “a held commitment to re-buy or re-patronize a preferred product consistently in the future” (Ganesh, Arnold, and Reynolds, 2000; Jamal and Anastasiadou, 2009; Kim et al., 2007; Oliver, 1999; Stank et al., 1999, 2003). In literature there is evolved a distinction in two different dimensions of loyalty: the behavioural and the intentional dimension. The behavioural dimension is expressed in the repeated repurchase of services and products in past and present (Baldinger and Rubinson, 1996). The intentional dimension is expressed in the intention to repurchase, the intention to
recommend a firm to a friend or to increase the influence of competitive players in a firm’s industry. (Wong and Sohal, 2003).
Customer loyalty cannot be earned in one single transaction and will mostly develop
incrementally over time. Customer loyalty becomes possible through the development of long term, mutually beneficial relationships with customers (Athanasopoulou, 2009). The
relationship must be favourable for both parties to be successful. It is important to align this long term character of loyalty with the long term goals of a firm. A customer can be loyal on a personal level or a firm level. Both levels can be perceived on a different level by the customer. Although in prior research is demonstrated that customers interpersonal loyalty is often more important than loyalty towards brands, companies or points of sale (Oliver 1997; Guenzi and Pelloni, 2004) this study is limited to the loyalty on firm level. Without
underestimating the interpersonal influence, customers have a relationship with a firm and loyalty is measured by the intention to stay as a customer of a firm. Relationships mostly are legally formalized by contracts between customer and firm and not between customer and salesperson.
2.1.2 Why is loyalty important?
In most firms there is an exchange relationship between the provider of services or products and the customer. This can be on a business- to- business level or on a business- to- consumer level. However, the transactions are based on the principal of an exchange of services or products on one hand and financial earnings on the other hand. Loyalty is identified as a
service industries there is not a physical product to sell to the customer. An organization is judged on its present behaviour and on the expectations of future behaviour. For this reason loyalty is top priority in service industries (N’Goala, 2007). This study is directed to loyalty in service industries. In next section the consequences of loyalty are more specified.
2.1.3 Consequences of loyalty
Achieving a certain level of loyalty requires effort to influence the behaviour of a customer for future actions. But what behaviours are favourable for an organization? The consequences of loyalty are divided in four different directions.
First of all, loyal customers tend to have a greater intention to re-purchase (Bloemer and Odekerken-Schroder, 2002; Bowen and Shoemaker, 1998; Chi, Yeh, and Yang, 2009; Hennig-Thurau, Gwinner, and Gremler, 2002; Kim et al., 2007; Singh and Sirdeshmukh, 2000; Stank et al., 1999, 2003). This is the base of loyalty and also mentioned in prior research as in- role behaviour. Repurchase behaviour is part of the basic exchange relationship between a buyer and a seller.
The second type of behaviour is recommending an organization to another potential customer. (Bloemer and Odekerken-Schroder, 2002). By recommending an organization the customer demonstrates a certain extent of personal identification with the organization to another customer. The customer will do this this only if he is reasonably sure about the behaviour of the recommended organization. Recommendation is not a basic action in the role of
exchanging services or products with financial transactions. It can be characterized as an example of extra role behaviour.
Thirdly, prior research has demonstrated that loyal customers are less price sensitive (Stank et al., 2003) and this leads to a direct increase of profitability. (Verhoef, 2003). A loyal customer
will be more sensitive for other factors than just the price, for example the quality of the relationship with the organization.
The fourth influence of loyalty is that there is a decreased risk of defection due to competitors’ promotional activities (Stank et al., 2003) and loyal customers demonstrate greater resistance to counter persuasion and negative word-of-mouth. (Bloemer and
Odekerken-Schroder, 2002). The influence of a negative word- of- mouth about a supplier is more limited in case of a loyal customer than in case of a non-loyal customer.
These favourable behaviours can only be achieved incrementally by a developing relationship over time and can be valuable for an organization especially in case of tough external
circumstances. For this reason it is important to examine the exact drivers of loyalty. In the next section the main ingredients of loyalty are discovered.
2.1.4 Drivers of loyalty
When analysing the definition of loyalty there is an extent of commitment between the customer and the organization he is dealing with. Researchers seem to agree that this commitment is based on relationship quality and contains two dimensions, satisfaction and trust (Ou et al, 2012). Relationship quality can be achieved through organization’s ability to reduce perceived uncertainty. (Crosby, Evans and Cowles, 1990). This holds for organizations and persons as well. In the following sections this particular relationship between loyalty with satisfaction and trust is elaborated.
unpleasant. Hence, satisfaction can be characterized as a global effect evaluation or feeling state that is influenced by perceived performance as the cognitive component of the
evaluation (Oliver, 1997). This suggests that customers compare their expectations to the experiences with a particular offering, and that expectancy confirmation leads to satisfaction (Bitner, 1990). Various studies in particularly marketing science have determined the drivers of customer satisfaction. (E.g. Westbrook, 1981; Zeithaml et al., 1988). Although the
customer satisfaction can depend on the context a general insight has developed as to the most important drivers of customer satisfaction. The drivers are constantly found to be (1) service quality provided by the seller organization, (2) product quality, and (3) value-for-money (Evanschitzky, Sharma and Prykop, 2012). Customers are satisfied if the company meets the expectations of customers on these three factors. The three drivers can be interpreted very broadly but they have one factor in common, the extent of meeting expectations.
2.1.6 The relationship between trust and loyalty
Prior research has demonstrated that there is a strong positive direct relationship between trust and loyalty. (Eiseringerich and Bell, 2007; Morgan and Hunt, 1994; Ouyang, 2010; Hansen, 2012). A higher level of trust in an organization or a person leads directly to a higher level of loyalty. Given the emphasis placed on trust in commercial exchange relationships, it is also important to understand how trust influences specific aspects of customer behaviour. (Doney and Cannon, 1997). In the next section there is given an extensive insight in the definition of trust, the different levels of trust and the importance of trust in doing business.
2.2 Trust
In business it is important to trust the supplier you are doing business with. There are several studies about the importance of trust in explicit economic exchanges (Arrow, 1974;
Granovetter 1985; Macauley 1963). This holds on business- to- consumer relationships and on business- to- business relationships as well. About trust is easily spoken but it is a personal expectation and feeling and this implicates that it should be complicated to define trust. There are various views or perspectives about trust. To create an insight of trust in the next sections it is elaborated what actually trust is, what dimensions trust determine and what levels of trust are to be distinguished.
2.2.1 What is trust?
Rotter (1967) defines trust as “a generalized expectancy held by an individual toward the others”. This expectancy is further developed by Moorman et al. (1992) as “willingness to rely on an exchange partner in whom one has confidence”. Drawing on literature in social psychology (Larzelere and Huston 1980) and marketing trust is defined as the perceived credibility and benevolence of a target of trust (Ganesan 1994; Kumar, Scheer, and
Steenkamp 1995). Morgan and Hunt (1994) and Arnott (2007) stated that ” trust exists when one party has confidence in an exchange partner’s reliability and integrity”. Doney and Cannon (1997) define trust as “the overall perception of an individual with regards to the credibility and benevolence of the group of people representing the partner firm in the relationship”. Doney et al. (1998) define trust more sharply. According to these researchers
necessary for trust to be present. To reduce uncertainty in doing business it is important to have a real expectation in the manner organizations behave in different situations. Therefore in this study the definition of trust is used as in Zaheer, McEvily and Perrone (1998): “Trust is the expectation that (1) an actor can be relied on to fulfil obligations (Anderson and Weitz, (1989), (2) will behave in a predictable manner and (3) will act and negotiate fairly when the possibility for opportunism is present” (Anderson and Narus, 1990, Bromiley and Cummings, 1995).
It contains the exchange relationship (1) the supplier can be relied upon to fulfil the
obligations that are agreed upon with both parties. In most definitions a term of exchanges is included. This is a relevant component as “trust is the most universally accepted variable as a basis of any human interaction or exchange” (Gundlach and Murphy, 1993). The second component is the expectation of future behaviour (2). To develop trust it is important to reduce uncertainties in future and therefore a predictable behaviour of the supplier
organization or its representative is crucial. The component of fairness (3) suggests that the supplier is really intended to make the best offer for the customer independent of the customer’s extent of vulnerability.
Trust can be achieved as an incremental process over time and previous studies show that trust is a critical element in the building of relationships between customers and salespersons (Achrol, 1991; Johnson et al., 2003; Swan et al., 1999). The high levels of trust characteristic of relational exchange enable parties to focus on the long-term benefits of the relationship (Ganesan 1994).
2.2.2 Dimensions of trust
The first dimension of trust focuses on the objective credibility of an exchange partner, an expectancy that the partner's word or written statement can be relied on (Lindskold 1978). Based on the definition of Zaheer et al (1998) this it to summarize as (1) reliability.
The next dimension of trust is predictability (2) and it simply refers to the degree of consistency of intended behaviour.
The third dimension of trust, fairness (3), is the extent to which one partner is genuinely interested in the other partner's welfare and motivated to seek joint gain. It refers to the sincerity of actions of the person or organization doing business with. Even in cases of vulnerability a customer must rely on a fair behaviour of the supplier.
2.2.3 The multi levelled nature of trust
A large stream of previous research underlines the importance of trust in interpersonal relationships (e.g., Rotter 1967; Schlenker, Helm, and Tedeschi 1973). Although some researchers disagreed about whether organizations can be targets of trust, more and more of literature emphasizes that people can develop trust in public institutions (Lewis and Weigert 1985) or organizations (Morgan and Hunt 1994), as well as individuals. Therefore, the trust literature suggests that in business context, customers can trust the supplier firm, its
salesperson, or both. Even the Oxford English Dictionary confirms this line of thinking and defines trust as: “confidence in or reliance on some quality or attribute of a person or thing, or
scope trust, interpersonal trust and firm-specific trust (Grayson, Johnson and Chen, 2008). In this study the term firm- specific trust is replaced by firm trust. This level of trust is narrow in scope because it affects only the relationship in which it has developed and thus has a
relatively limited scope of influence. It is based on the customer’s direct experiences within the relationship with his supplier. Trust in both a person and an organization is fostered by a process of collecting partner-specific information, but people use different types of
information to develop trust in each (Kramer, Brewer, and Hanna 1996).
Contrary to narrow scope trust we use the term “broad scope trust” to refer to a customer’s trust in the broader social context in which a relationship might develop. This trust is broad in scope because it affects a customer’s perceived behaviours and perceptions regarding not a specific relationship but rather a whole range of existing and potential relationships. This information often comes through first-hand interactions, but second-hand data about the partner, such as reputational information, can give this sort of trust a head start (Bouty 2000; Kollock 1994).
Broad scope is divided in system trust and generalized trust. System trust is defined by Lewis and Weigert (1985) as “trust in the functioning of bureaucratic sanctions and safeguards.” System trust is context specific because it refers to a customer’s views regarding the regulation of a particular activity system. System trust is the hard side of broad scope trust. The softer side of broad scope trust is generalized trust, defined as the trust in people in general. Zucker (1986) calls this the “background expectations” of trust, the soft side of broad scope trust. Since this study is directed to trust in an exchange context only the generalized trust is investigated.
In section 2.3 to 2.6 interpersonal, firm and generalized trust are extensively elaborated separately. The aforementioned division above is presented in figure 1.
2.3 Firm Trust
In section 2.2 the multi levelled nature of trust is exposed. In previous research the terms of firm trust and corporate reputation are used to address the same constructs. Firm trust is defined as a relatively stable, long term collective judgement by outsiders of an organization’s actions and performance. It implies a long lasting cumulative assessment rendered over a long time period. (Gioia, Schultz and Corley, 2000; Sirdeshmukh, Singh and Sabol, 2002; Ou, Abratt, and Dion, 2006). Firm trust can be measured by company assessment, such as
performance and strong record of profitability, and emotional appeal, for example, respect and trust (Fombrun, Gardberg, and Sever, 2000; Fombrun and Shanley, 1990; Ou et al., 2006).
2.2.1 The importance of firm trust
There is a steady stream of literature about trust in a firm or organization. Most firms are recognizable as a brand and this enables organizations to obtain a clear identification. The role of a firm’s reputation is comparable to the brand equity of a firm. But why is this recognition as a brand so important? Customers often do business with a firm because they can identify themselves with a firm or a brand (Bhattacharya and Sen (2003). This can be on a social or an emotional level. A second driver of doing business with a firm is based on a practical and economic level by saving time and searching costs. A particular firm or brand name can prevent consumers for doing research in every case of making a buying decision. It saves time and costs. But, the most important driver of the firm choice in our study is the reduction of risk. If consumers trust the firm’s identity they are likely to perceive lower risk. (Grewal, Gottlieb and Marmostein, 1994).
2.3.2 The relationship between firm trust and loyalty
Firm trust has positive impacts on a customer’s intention to cooperate (Yener, 2009). There is a link between firm trust and the customer’s future purchase intentions when customers can rely on the integrity of firm they trust. The supplier will not knowingly act contrary to the customer's best interests. According to Ganesan (1994), trust is a necessary ingredient for long term orientation because it shifts the focus to future conditions. Empirical evidence supports the notion that firm trust is central to a customer's intention to continue the exchange relationship. Anderson and Weitz (1989) find evidence that trust is the key to maintaining continuity in conventional channel relationships. Morgan and Hunt (1994) demonstrate a negative relationship between trust and propensity to leave, which is defined as the perceived likelihood that a customer will terminate the relationship in the near future. The relations are all directed to continue the relationship for a long time period and this is the essence of loyalty. Hansen (2012), Eisingerich and Bell (2007), Morgan and Hunt (1994) stated the positive direct relationship between firm trust and loyalty. As a result the following hypothesis is proposed:
H1: Firm trust has a positive effect on loyalty
In the next section an insight is provided in the definition of interpersonal trust, its importance and its relationship with loyalty.
2.4 Interpersonal Trust
In service industries a customer maintains mostly a relationship with a firm and a person or both. Although the customer is the same person, trust in the firm and in the salesperson will be judged on totally different types of information. (Kramer, Brewer, and Hanna 1996). The firm is judged by for example a strong record of profitability and emotional appeal.
(Fombrun, Gardberg, and Sever, 2000; Fombrun and Shanley, 1990; Ou, 2007; Ou et al., 2006). The salesperson on the other hand is judged by totally different attributes like customer orientation, expertise and likeability. (Guenzi and Georges, 2010). The definition of
interpersonal trust does not differ from the general definition of trust used in previous sections: Trust is the expectation that an actor (salesperson) can be relied on to fulfil obligations (Anderson and Weitz, (1989), will behave in a predictable manner and will act and negotiate fairly when the possibility for opportunism is present (Anderson and Narus, 1990, Bromiley and Cummings, 1995). Equal with firm trust, interpersonal trust develops incrementally over time, and contributes to development of long term relationships.
In service literature the role of interpersonal trust has been extensively studied but why is interpersonal trust important?
2.4.1 The importance of interpersonal trust
The topic of the contribution of interpersonal relationship in building customer relationships and loyalty with service providers has attracted increasing attention in the last few years (Bove and Johnson, 2001). In many service environments the firm’s relational intent and ability are to a great extent personified and expressed in practice by the representatives. (Price and Arnould, 1999). This personal factor in the relationship can be considered a key element of the offering (Czepiel, 1990) because they provide social and confidence/trust benefits (Goodwin and Gremler, 1996). With regard to service literature Gummesson (1987)
underlined that in service companies the quality of the relationships between customers and front-line employees, which incorporates both a professional and a social dimension, can strongly contribute to the customer’s overall perception of quality of the service provider.
2.4.2 The relationship between interpersonal trust and loyalty
The interpersonal relationship between the salesperson and the customer can have a
substantial impact on important relational outcomes for the selling firm. It fosters customer satisfaction, commitment and trust in the firm as well as repurchase intentions, willingness to recommend the provider to other potential customers, and to provide referrals (Foster and Cadogan, 2000; Johnson et al. 2003; Kennedy et al., 2001). Guenzi and Georges (2010) find that the selling orientation, expertise and likeability of a salesperson lead to trust and trust lead to customer’s behaviour as repurchase, recommendation, and a reducing intention to switch. Therefore the following hypothesis is formulated:
2.5 The relationship between interpersonal trust and firm trust
The relationship between interpersonal trust and firm trust can be characterized as complex. On one hand there is a strong interdependency between both levels. As stated in previous sections the different levels of trust are judged by customers in a totally different way and therefore both constructs can be measured differently. On the other hand there is a strong process of transferring interpersonal trust to firm trust and vice versa. In the next section this complex relationship between both levels is further elaborated.
2.5.1 Interdependency of interpersonal trust and firm trust
When doing business with a supplier firm and its representative there are factually two levels of relationship. Although the formal and contractual relationship mostly exists between customer and firm the representative often plays a key role in this relationship. Many authors find that customer trust and loyalty to the salesperson on the one side and to the firm on the other are distinct but interrelated constructs. (Doney and Cannon, 1997;Swan et al., 1999). Trust in both a person and an organization is developed on partner-specific information collecting, but customers use different types of information to develop trust in each (Kramer, Brewer, and Hanna 1996).
Both levels play different roles in affecting negotiation processes and exchange performances (Guenzi 2004). A long term relationship with a trusted firm could come under pressure by a salesperson who acts dishonestly and unreliably ( Kelly and Schine, 1992). Conversely, a highly trusted salesperson can preserve customer commitment during difficult times created by management policies that appear contrary to the customer's best interests (Schiller, 1992). Thus, the relationship between both levels of trust is reinforcing each other.
In this study the relationships firm trust- loyalty and interpersonal trust- loyalty are investigated.
In line with the aforementioned interdependency it is investigated that trust can be transferred from a firm to a salesperson and conversely. (Chen and Mau, 2009; Doney and Cannon, 1997). An interesting issue is how the transfer process between firm trust and interpersonal trust works. This issue is elaborated in the next section.
2.5.2 Transferability of trust
Although the levels of trust are perceived independently, both are conversely strongly related. In personal selling literature it is widely recognized that the trust in the salesperson transfers to trust towards the firm, which the salesperson belongs to (Chen and Mau, 2009; Doney and Cannon, 1997). Customers assume that the salesperson's behaviour reflects the firm’s values and attitudes (Doney and Cannon, 1997).
According to Strub and Priest (1976), trust can be transferred from a customer’s well known level of the relationship to a closely associated but less well-known group or individual. This suggests that previous interactions with the better- known party (e.g. the firm) provide a basis for inferring the extent to which a new or less well-known party (e.g. salesperson) can be trusted. There is a reciprocal causal relationship between the two targets of trust, whereby trust in the salesperson leads to trust in the firm and vice versa. (Doney and Cannon, 1997).
from the customer’s past experience with the supplier firm. Conversely, distrust also can be transferred if a person lacks other information.
A favourable reputation is also easily transferrable across firms and enhances the credibility of the supplier firm (Ganesan 1994). If a customer assumes that the firm's reputation is well deserved, trust will be granted on the basis of the firm's history in relationships with other firms. In other words, customers infer the trustworthiness of a supplier through the words and actions of other persons and firms. The transferring process can be used to predict a positive relationship between firm reputation and trust of the supplier firm.
2.5.3 The contribution of firm trust and interpersonal trust on loyalty
For service related firms it is important to know to what extent loyalty is formed by firm trust or by interpersonal trust (Pritchard et al., 1999). A firm disposes of limited resources and managers constantly have to make tradeoffs in investing in reputation, brand equity or in human resources. What level of trust is most important to acquire the loyalty of your customer? Taking into account the strong interdependence between organizational and interpersonal relations (Doney and Cannon, 1997; Iacobucci and Ostrom, 1996; Zemanek, 1997), most academic research emphasises the importance of the development of social bonds among the subjects involved (Iacobucci and Hibbard, 1999). Oliver (1997) concluded that interpersonal loyalty is often more important than loyalty towards brands or firms. The firm’s relational intent and ability are to a great extent personified and propagated by sales persons. Their characteristics and behaviours have a considerable impact on the results of relations with customers (Price and Arnould, 1999).
Also in business-to-business relationships, it has been pointed out that relationships require strong interpersonal elements rather than between organized corporate groups” (Yau et al.,
1999; Mavondo and Rodrigo (2001) find that in business markets the “social bonding” (reciprocal friendship, personal liking) is the most important ingredient of relational outputs for the selling firm, such as customer commitment, trust and co-operation. Based on these foundations the hypothesis is formulated as follows:
2.6 Generalized Trust
In section 2.2 the different levels of trust are divided into narrow scope trust and broad scope trust. The previous sections contained narrow scope trust consisting of firm trust and
interpersonal trust. In addition to narrow scope trust customers develop broad scope trust which is “trust in the social context in which the relationship is taking place”. (Driscoll 1978). Narrow scope trust is extensively studied in relationship marketing literature, though broad scope trust has been clearly under-exposed in literature (Hansen, 2012). The definition of broad scope trust proposed by Sirdeshmukh, Singh, and Sabol (2002), is “the expectation held by the customer that companies within a certain business type are generally dependable and can be relied on to deliver on their promises”. Other researchers define it as “a generalized expectancy that the promise of a group can be relied upon. (Rotter, 1980, Siegrist, Gutscher and Earle, 2005). Broad-scope trust applies to all organizations and individuals operating within a particular context. In line with the perceived levels of trust in previous sections, broad scope trust refers not to trust that is held broadly by all customers in a marketplace but rather to the trust that a particular customer has in the business context in which a set of organizations and individuals operate. Customers within the same marketplace may have similar levels of broad scope trust (Doney, Cannon, and Mullen 1998), but there will also be variance. In other words: broad scope trust can be defined as “a customer’s trust in an industry as a whole”. A customer’s broad scope trust is to divide in two kinds of trust formal (system trust) and informal (generalized trust).
2.6.1 System trust and generalized trust
System trust is based on the customer’s belief that there are third parties who are protecting a customer in a particular activity system. Responsible parties will publicize information about those who break trust (Milgrom, North and Weingast 1990) and will impose punishments for
untrustworthy behaviour (Hardin 1996). This is system trust, which Lewis and Weigert (1985) define as “trust in the functioning of bureaucratic sanctions and safeguards.” System trust is context specific because it refers to a customer’s views regarding the regulation of a particular activity system.
In additions, there is identified a second type of broad scope trust: trust in people in general, generalized trust. Zucker (1986) calls this the “background expectations” component of trust (Driscoll 1978), and it is a tendency to trust all members of a particular social system,
regardless of sector or context. Trust is not based on regulations but it reflects on a person’s beliefs about appropriate relationship norms, which are learned through multiple interactions over time. It is relevant to know if the target of a customer’s trust is an individual entity (narrow scope trust) or an entire group of entities (broad scope trust).
As a result of the definition “a generalized expectancy that the promise of a group can be relied upon” (Rotter, 1980, Siegrist, Gutscher and Earle, 2005) the terms of broad scope trust and generalized trust are used interchangeably. In the remainder of this study the term of broad scope trust is substituted by “generalized trust”. In line with the research of Hansen (2012) generalized trust is considered because informal trust is more directly related to the behaviour of firms than system trust.
This study examines how generalized trust influences interpersonal and firm trust. A perspective regarding the nature of this influence is presented in the next section.
2.6.2 The effect of generalized trust on firm trust and interpersonal trust
Previous research points out that a climate of trust in an industry is needed to enhance a customer’s willingness to do business with salespersons (Starkman 2005). So there IS a relationship between generalized trust and firm/ interpersonal trust. As stated in section 2.6 the effects of generalized trust have been seldom investigated. There are only two previous studies which investigate the role of generalized trust on the customer- firm relationships. The first study is performed by Grayson, Johnson and Chen (2008) and tests the influence of generalized trust on customer perceptions and behaviours on two different perspectives, the functionalist theory perspective and the institutional theory perspective. The second study is performed by Hansen (2012) and measures the moderating influence of generalized trust on customer- firm relationships. The moderating effect of generalized trust will be elaborated in section 2.7.
Researchers seem to agree that generalized trust helps reduce complexity and uncertainty for customers and that it plays an important role in relation to narrow-scope trust. Grayson, Johnson and Chen (2008) compare two differing views. First, the functionalist perspective which assumes a negative relationship between generalized trust and customer trust in firms and their representatives (Luhmann 1979). It suggests that generalized trust is a substitute for narrow scope trust. In fact narrow-scope trust is formed where it is needed (Luhmann, 1979). Secondly, the institutional perspective which assumes a positive relationship between
generalized trust and customer trust in firms and their representatives (Bachmann 2004). It suggests a reinforcing effect on different levels of trust. Grayson, Johnson and Chen (2008) find that generalized trust is not a substitute for narrow scope trust (functionalist’s
perspective) but rather generalized trust and narrow scope trust reinforce each other. It also suggests that the influence of generalized trust on customer attitudes and behaviours is
indirect and that firm trust and interpersonal trust plays a key role with regard to this
influence. Although many factors can influence narrow-scope trust, the results suggest that a potentially important factor is the level of trust that customers have in the business context in which a firm operates. Hansen (2012) underlines the positive influence of generalized trust on narrow scope trust.
2.7 The moderating effects of generalized trust
When exploring the moderating effects of generalized trust on the relationship between narrow scope trust and loyalty, the base theory is the attribution theory. Kelley (1967) has conceptualized this as the “process by which an individual interprets events as being caused by a particular part of an environment”. The attribution theory is commonly used in
explaining how consumers investigating experiences within customer–seller relationships. Consumer’s perceived causes can be threefold: internal, external and situational. The internal causes consist of the consumer himself or in his own decisions. The external causes consist of the company that offers the service; a relevant example in this study is narrow-scope trust. The situational causes consist of environmental effects, for example generalized trust (Oliver, 1993; Ryu, Park, and Feick, 2006). Negative experiences tend to be ascribed to situational factors rather than internal or external factors.
In a low trusted environment customers are more likely to attribute negative experiences to situational causes (broad scope trust) and less likely to external causes (a poor firm
narrow-scope trust and loyalty. In aforementioned examples a low trusted environment is assumed. In a high trusted environment the effects will be the opposite. Thus, it is expected that generalized trust would negatively moderate the relationships between narrow-scope trust and loyalty. This is because when generalized trust is low, the consumer would be expected to be more likely to attribute trust to the customer–seller relationship than to a situational cause; and vice versa when generalized trust is high. Thus, the consumer would probably be more likely to convert narrow-scope trust into relationship loyalty under conditions of low
generalized trust than under conditions of high generalized trust. Hansen (2012) demonstrated the expected negative moderating effect of generalized trust on the relationship narrow scope trust on loyalty.
H4: The influence of firm trust on loyalty is negatively moderated by generalized trust such that firm trust has a greater positive effect on loyalty when generalized trust is low compared to high
H5: The influence of interpersonal trust on loyalty is negatively moderated by
generalized trust such that interpersonal trust has a greater positive effect on loyalty when generalized trust is low compared to high
3. Research model
To provide a visual insight in this study the research model in figure 2 contains the measured constructs and the five hypotheses.
Figure 2. Research Model
Hypotheses
H1: Firm trust has a positive effect on loyalty
H2: Interpersonal trust has a positive effect on loyalty
H3: Interpersonal trust has a larger effect on loyalty than firm trust.
H4: The influence of firm trust on loyalty is negatively moderated by generalized trust such that firm trust has a greater positive effect on loyalty when generalized trust is low compared
4. Research Methods
In the following the research method is presented and it will serve as an approach of how to find an answer to the research questions.
4.1. Overview of data collection
The first goal of this explanatory study is to explain the relationship between the independent variables, firm trust and interpersonal trust on one hand and the dependant variable, loyalty on the other. The second goal is to demonstrate moderating effects on the aforementioned
relationships. This study provides a cross-sectional insight and does not provide any insight over time. A commonly used and appropriate method to demonstrate relationships and effects is distributing a survey to collect quantitative data. The survey is distributed on line to reach a large audience inexpensively with a short time to respond. Data were collected from
connected people in the researcher’s personal network on LinkedIn. This group consists of professional connected employees, entrepreneurs and managers. They represent diversified types of industries, different departments and levels of hierarchy from a range of
organizations in mainly The Netherlands. The survey is designed by the application of Qualtrics.com, an online tool providing functions of creating, editing, distributing and
monitoring an online survey. The survey was distributed online to 897 LinkedIn connections and the respondents were requested to take 8 to 10 minutes of their time to complete it. The survey was distributed on line on November 27th 2013. After one week a reminder was sent and the survey closed on December 11th after exactly two weeks. The complete survey was formulated in Dutch with a guarantee of anonymity from the respondents. All participants were informed that information obtained from the surveys would be treated as confidential
and would not be used or released for purposes other than this study. If desirable the respondents could give feedback after answering the survey.
4.2. Procedure
The survey was adopted from the study of Hansen (2012) and the items were literally adopted from the appendix of this study. As the design of this study is comparable with the study of Hansen (2012) it is appropriate to use the available survey. The items measuring loyalty (3) and generalized trust (3) are used in line with Hansen (2012). The items measuring narrow scope trust (3) are used to measure firm trust and interpersonal trust both. As stated in section 2.2 narrow scope trust consists of both firm trust and interpersonal trust.
All respondents were asked to give their opinion and experiences with two different industries in which trust is an important factor, financial industry and hospital sector. In science
literature there is not a universal standard of trust levels among different industries. The Edelman Trust Barometer is frequently used and cited in academic research. The Edelman Trust Barometer 2012 and 2013 conclude that the financial services industry and the banking industry are worldwide least trusted (both 50%). Another point of view is the valuation of trust at a profession level. In 2013 Reader’s Digest European Trusted Brand Survey it is concluded that the financial advisor is seen as one of the least trusted professions by 18.000 respondents. The profession of a doctor on the other hand is considered as highly trusted. There is a founded reason to belief that there is a substantial difference in trust level between
important factor. In both industries trust can be perceived in the different levels of generalized, firm and interpersonal trust.
The respondents were asked for their opinion about both industries (financial and health), their experiences with firms (banks and hospitals) and their representatives (advisors and specialists). It was expected that a large group of respondents would have experiences with banks and/ or hospitals and/ or financial advisors, and/or specialists. Replying to all questions completely per industry contributes to the testing of the model.
4.2.1 Translation- back translation procedure
All items used in the questionnaire were derived from English studies. Since mostly all
respondents to the survey have Dutch as their first language, the original items were translated into Dutch by an independent third person speaking native Dutch. In order to ensure that the content of the items remains unchanged, the translated Dutch items were back translated into English by a native English speaker. A small number of discrepancies between the back-translated and the original items were corrected in the final Dutch version of the
questionnaire. The complete survey in the Dutch version is enclosed in Appendix 1.
4.2.2 Pilot testing
Prior to the administration of the questionnaire and collection of data, the questionnaire needed to be pilot tested to assess the validity of questions (Saunders, Lewis and Thornhill, 2009). In this study the questionnaire was pilot tested on five more or less concerned people. By means of a pilot test, some face validity of the questionnaire could be assessed.
Respondents were given the opportunity to comment on the questionnaire and provide suggestions and recommendations for any adjustments. Participants of the pilot were asked questions regarding the time needed to complete the survey, clarity of instructions and
questions, layout and additional comments. The pilot testing was also executed to ensure that there are no technical complications.
4.3 Data analysis
The analysis of data has been executed in steps. First of all, the entire data base has been collected by the survey online survey software Qualtrics. Secondly, the obtained data were transferred to IBM Statistical Package for Social Sciences, SPSS 22.0. With this statistics program the data has been cleaned and prepared for the final data analysis. A part of the items were recoded reversely. In appendix 1 and 2 the reversely coded items were marked. After recoding the level of the Cronbach’s alpha was calculated in order to examine the internal consistency of the constructs. The outcomes of these results are reported in the Measurement section 5. The data were cleaned to leave over the 395 appropriate observations of the
ultimate responses. The incomplete responses were removed. Also the responses not including relevant experiences with either financial companies and their representatives or hospitals and their representatives were removed.
In SPSS two different analyses were run. Firstly, a correlation analysis was run between loyalty, firm trust, and interpersonal trust and generalized trust. Secondly, the model presented in figure 2 was tested and the influence of the moderator “generalized trust” was measured by a linear regression analysis.
4.4. Participants
The sample consisted of an amount of 897 LinkedIn connections in researcher’s personal network. 450 respondents started the survey and 324 filled in the survey completely. The response rate 36,1% which is a higher response rate than average percentage for this type of distribution (Saunders, Lewis and Thornhill, 2009). The reason for this might be that the respondents were part of a personal network which supposed to be concerned at a higher level. Respondents filled in three questions to obtain control variables. First control variable was gender and the sample consisted for the most part of male, n =277 (85,5%) and for a minority of female, n= 47 (14,5%). The difference can be explained by the demographics of the personal network. The network contained mainly entrepreneurs and high level managers and apparently this composition yields this distribution. The second control variable was the age of the respondent. The age structure is as follows: 18- 24 years, n= 5 (1,5%), 25-34 years, n= 38 (11,7%), 35- 44 years, n= 134 (41,4%), 45-54 years, n= 113 (34,9%), 55-64 years, n= 33 (10,2%) and 65+, n= 1 (0,3%). The third and last control variable was the level of
education. The majority of respondents earned at least a bachelor’s degree, n= 199 (61,2%) and a substantial part earned at least a master’s degree, n= 87 (26,8%). This implies a relatively highly educated sample.
From the total group of 324 respondents the responses were just valid if (1) there was an experience with each level of trust (industry, firm and person) and (2) all questions all were replied completely within an industry. A lack of experience or unfinished blocks of questions was interpreted as missing data and consequently invalid. An amount of 188 respondents had experiences on each level of trust in financial industry. They expressed their opinion about generalized trust in the financial industry, trust in their own specified bank, and trust in their
specified financial advisor and expressed their loyalty intentions to their bank. An amount of 207 respondents gave their opinion about trust in health industry, trust in their own specified hospital and in their own specified medical specialist and expressed their loyalty intentions to their hospital. This provided a total amount of 395 appropriate responses. Within this both groups there was an overlap of 115 respondents who have experience with both industries. The level of match between the demographics and the three groups describes the goodness of fit. Three requirements have to be met for the goodness of fit test: (1) a random sample, (2) the variables must be mutually exclusive and (3) there must be a minimum of 5 occurrences in each category. The third requirement was not initially met and therefore categories containing less than 5 occurrences were merged with the connecting categories.
Between the three groups (just financial, both and just hospital) the chi square test indicates that there is a slightly lower than expected amount of females having experience with financial suppliers and a higher than expected amount of females having experience with hospitals. This effect is just significant (Pearson Chi- square value 4,932 and p = 0.085, 2- sided). The higher hospital experiences from female respondents may be explained by the experiences concerning care of specific female events as being pregnant. Concerning the education level the results express a higher than expected experience of master graded respondents with financial experiences. Conversely, the bachelors graded respondents have more experiences with both industries (Pearson Chi- square value 13,145 and p = 0.011, 2- sided). There is not a valid explanation to find for this difference. The age structure is
5. Measurement
Items were administered in Dutch adopted from the English original used in the study of Hansen (2012). The original items are enclosed in Appendix 2. The survey provided a short introduction to the respondents to explain the definitions of trust and the firms/ persons to be measured. In line with Hansen (2012) all items were rated on a seven -point Likert scale ranging from strongly disagree (1) to strongly agree (7). In some items a choice
“inapplicable” was given to exclude respondents who did not have experiences with, for example, a bank advisor. Overall scale scores were computed by averaging item means. All measured items were based on prior research and are identical across industries (Hansen, 2012).
Loyalty was measured with the use of a 3-item measure and a rating scale, precisely a seven
point Likert scale ranging from strongly disagree (1) to strongly agree (7) adopted from Sirohi, Mc Laughlin and Wittink (1998) and Tax, Brown and Chandrashekaran (1998). An example of an item is: “I’m considering changing bank/ hospital within the next twelve months”. Analyzing the reliability of this scale yields a Cronbach’s α of .805.
Generalized trust was measured with the use of a 3-item measure and a rating scale, precisely
a seven point Likert scale ranging from strongly disagree (1) to strongly agree (7) adopted from Tax, Brown and Chandrashekaran (1998). An example of an item is: “In general, I believe that banks/ hospitals cannot be relied upon to keep their promises”. Analyzing the reliability of this scale yields a Cronbach’s α of .849.
Firm trust was measured with the use of a 3-item measure and a rating scale, precisely a
seven point Likert scale ranging from strongly disagree (1) to strongly agree (7) adopted from Tax, Brown and Chandrashekaran (1998). An example of an item is: “I believe that my bank/ hospital is trustworthy”. Analyzing the reliability of this scale yields a Cronbach’s α of .876.
Interpersonal trust was measured with the use of a 3-item measure and a rating scale,
precisely a seven point Likert scale ranging from strongly disagree (1) to strongly agree (7) adopted from Tax, Brown and Chandrashekaran (1998). An example of an item is: “Overall, I believe my financial advisor/ medical specialist is honest”. Analyzing the reliability of this scale yields a Cronbach’s α of .913.
Measuring the scale of reliability of the separate constructs demonstrates a constant reflection of the construct (Field, 2009). The reliability scores are based on Cronbach’s α which express the extent of intern consistency between the items. When α is equal or higher then 0.8,
reliability of the scales can be assumed. The results meet the requirements of a reliable measurement. The reliability scores are presented in the correlation analysis in table 2.
6. Results
In this section the results of the stepwise process will be reported in different sections. In the section 5. Measurement the reliability scores are described yet. In section 6.1.1 the descriptive statistics are presented. The descriptive statistics about the constructs in total and on industry level are presented in Table 1. In 6.1.2 the correlation analysis is elaborated. Table 2. presents the results of this correlations analysis including the correlation and reliability scores. Section 6.1.3 presents the testing of the model as a whole and the effects among the measured
constructs by the results of the regression analysis. These results are presented in Table 3 and this table also includes multicollinearity scores. Section 6.1.3 also includes the testing of hypotheses. A similar regression analysis is used to zoom in on the effects within the two industries and these results are presented in Table 4.
6.1.1 Descriptive statistics
The means of the four constructs are measured and on a seven point Likert scale. The
respondents’ total scores result in the highest mean at interpersonal trust (M= 5,592), a lower mean in firm trust (M= 5,275) and the lowest mean in generalized trust (M= 4,859). This result shows that interpersonal trust scores slightly higher than firm trust. Compared to this there is a relative low generalized trust in the both industries. Respondents apparently have a stronger trust in their own firm and its representative than in the industry as a whole.
Although there are slight differences the SD of loyalty is lowest (1,101) and the SD of generalized trust is highest (1,200).
The scores in both industries hardly differ from each other on the level of loyalty and interpersonal trust. Although the differences are not as large as expected the trust in respondent´s hospitals (M= 5,446) achieved higher scores than in respondent´s banks (M=
5,087). Based on the Edelman Trust Barometer mentioned in section 4.2 it is expected that there is a large difference between generalized trust (lower) in financial industry and the generalized trust in hospital industry (higher). The results in this research confirm this
difference. The Edelman Trust Barometer presents these differences as two extremes on both sides. The results for the financial industry (M= 4,617) do not differ in such extreme extent with the hospital industry (M= 5,079) as suggested in The Edelman research.
More remarkable is the importance within both industries. In hospital industry the means of firm trust and interpersonal trust are almost equal. (M= 5,446 versus M= 5,565). It suggests that respondents perceive trust in the doctors in the same extent as in their hospital. In the financial industry the difference between firm trust and interpersonal trust (M= 5,087 versus M= 5,622) is larger than in the hospital industry. It suggests that in case of a lack of
generalized trust in an industry respondents fall back on the interpersonal trust of an advisor more than on a bank’s firm reputation. The standard deviation of Generalized trust in financial industry is largest (SD= 1,250) and there is least variance on hospital loyalty (SD= 1,043)
In both industries the means increase as the level of trust becomes narrower. The lowest scores are measured on the industry level, followed by firm level and finally followed by interpersonal trust. It indicates that respondents give higher scores on trust as the context becomes narrower. In financial industry this distinction is more evident (Industry = 4,617, Firm = 5,087, Interpersonal = 5,622) than in hospital industry (Industry = 5,079, Firm =
6.1.2 Correlation analysis
The correlation analysis is used to explore the strength of the direct relationships between the variables. Pearson correlations are used to investigate the relationships between the variables. These correlations give an indication for the confirmation of hypotheses. In Table 2. an insight is provided in the means, standard deviations, and the Pearson correlations, reliability of the constructs loyalty, firm trust, interpersonal trust and generalized trust.
In this table average scale means are reported. All items were measured on 7 point Likert scales. All correlations are significant at the 0.01 level (using a 5% α level) and the
the found literature. The strength of all correlations is moderate (0.30 <r < 0.70, p < 0.01) except of the relationship between generalized trust and firm trust. With a score (r = .789, p< 0.01) there is a strong correlation between both constructs. A strong correlation may increase the probability of multi-collinearity in regression analyses. In the regression analyses in section 6.1.3 it will be taken into account. All correlations are positive and the results indicate that the three levels of trust separately contribute to a higher level of loyalty. There is also indicated that they enforce each other. There is a moderate positive relationship between firm trust and loyalty (r = .532, p< 0.01). Moreover there is a moderate positive relationship
between interpersonal trust and loyalty (r = .522, p < 0.01). Both firm trust and interpersonal trust have a positive effect on loyalty. The correlations of both levels of trust on loyalty are almost equal. Although not a part of the model there is a not significant direct relationship found between generalized trust and loyalty (r = .415, p = .838). However, the correlations do not take into account the proposed effects of the four constructs in the research model. This will be described in the next section 6.1.3 Results Regression Analysis.
6.1.3. Results regression analysis
The regression analysis was performed to test the Research model visualized in section 3. Research Model. The model consists of four constructs, loyalty (dependant variable), firm trust, interpersonal trust (both independent variables) and generalized trust (moderator). The moderating effects are expressed as “generalized trust x firm trust” and as “generalized trust x
At first it is measured if the multi-collinearity requirements are met. If these are not met there is a too strong presence of correlation among the independent variables. Thereby it could be too difficult to disentangle the contribution of each individual independent variable. To ensure that there is not a multi-collinearity issue the score of the tolerance must exceed the level of .20. For all measured constructs this requirement is met and the results are presented in the last column of Table 3. To eliminate (non-essential) forms of multi-collinearity, the variables firm trust, interpersonal trust and generalized trust were first grand mean centered. The interaction term was based on these mean centered variables. The results of the regression analysis are presented in Table 3.
The main goal of this research is to determine the effects of generalized trust on customer seller relationships in general. The hypotheses will be tested based on these assumptions. Moreover, it is relevant to zoom in the results per industry. As a result two identical
regression analyses are run in the financial industry and hospital separately. In these analyses the multi-collinearity requirements are also met. Both industry analyses are executed by an
identical approach as the overall analysis. In table 4. the results are presented overall and divided per industry.
Hypothesis 1 suggests a positive effect of firm trust on loyalty. Regression results presented in Table 3 indicate that firm trust significantly has a positive effect on loyalty indeed (β = .397, p < .01). Thus, hypothesis 1 is supported.
In the same line of thinking hypothesis 2 suggests a similar effect between interpersonal trust and loyalty. This research indicate that there is a significant positive effect of interpersonal trust on loyalty (β = .309, p < .01). Hypothesis 2 is also supported.
The distinction of the results by sector provides more detailed insights (Table 4.). The direct effects of firm trust and interpersonal trust are closer together in hospital industry (firm trust (β = .282, p<.01), interpersonal trust (β = .322, p<.01). In financial industry there is a larger difference between both levels, firm trust (β = .579, p<.01) and interpersonal trust (β = .278,
p<.01).
Furthermore the contribution of firm trust and interpersonal trust is reversed in both industries. In hospital industry the contribution of interpersonal trust (β = .322, p<.01) on loyalty is larger than the contribution of firm trust on loyalty (β = .282, p<.01). Conversely, in financial industry the contribution of interpersonal trust (β = .278, p<.01) on loyalty is lower than the contribution of firm trust on loyalty (β = .579, p<.01). It suggests that loyalty is importantly determined by the role of corporate reputation of financial firms and less by the interpersonal contact with salespersons. In hospital industry loyalty seems to be more
determined by trust in doctors than by trust in hospitals. At sector level hypothesis H3 would be supported in just hospital industry and the strong opposite effect in financial sector would reject H3.
Hypothesis H4 proposed that generalized trust will act as a negative moderator on the relationship between firm trust and loyalty (β: .081, p<.05). Although there is a significant moderating effect on this relationship the effect is in the opposite direction as expected. In case of an increasing generalized trust the contribution of firm trust on loyalty also increases compared to cases of decreasing generalized trust. Thus, there is no support for Hypothesis H4.
In financial industry there is no significant moderating effect of generalized trust on firm trust on loyalty (β: .087, p= .149). In hospital industry there is a positive moderating effect of generalized trust on the relationship between firm trust and loyalty (β: .100, p<.05). Although
there is a significant moderating effect on this relationship the effect is in the opposite direction as expected. In case of an increasing generalized trust in hospital industry the contribution of firm trust on loyalty also increases compared to cases of decreasing generalized trust.
Hypothesis H5 proposed a moderating effect of generalized on the relationship between interpersonal trust and loyalty. This hypothesis is not supported in this research on account of a too high p- value of .309 which does not lead to a significant score.
In financial industry there is not found a significant moderating effect of generalized trust on the relationship of interpersonal trust on loyalty (β: .015, p= .789). In the hospital industry the expected negative moderating effect of generalized trust on the relationship between
interpersonal trust and loyalty (β: -.112, p<.05) is found. In case of an increasing generalized trust in hospital industry the contribution of interpersonal trust on loyalty decreases compared to cases of decreasing generalized trust.
Overall, there is no convincing evidence for the expected moderating effects. The nature of the industry seems to influence the results of the effects within the research model.
