Managing Rewards to Enhance Customer Value: Empirical Study of Frequent Flyer Program

. The goal of relational program is to retain customers who are profitable to the organization. Reward point is a form of customer loyalty widely used by many industries including airline industry. Airline loyalty program, notorious as Frequent Flyer Program (FFP), is the most sophisticated marketing strategic used by airline industries nowadays. However most airlines have very little understanding of their FFP members yet have a little knowledge about their most valuable customers. Most airlines have inaccurately determined the customer values by only considering business worth of nominal profit generated by FFP members. The value of customers beyond purchasing behavior - called relational worth - has not been commonly captured yet. This non-financial value is predicted as a driver in retaining customers, hence becomes one of crucial factors in preserving the profitability of the organization. For this reason, this paper aims to examine the customer non-financial valuations of FFP members to the airline. The hypotheses are empirically tested with a sample of FFP members of respected airline conducted through online survey (n=475). The data were statistically analyzed using structural equation modeling (SEM). Results indicate that perceived rewards generate higher value of relational worth through indirect relationship by intervening variable of relationship quality than that of generated by a direct relationship.


Introduction
In the context of customer equity, consumers generate values to the organization by means of financial and non-financial contributions. Based on the Customer Lifetime Value (CLV) approach, financial value to the organization is typically defined as business worth of nominal profit generated by customer (Blattberg & Deighton, 1996). However, traditional model of CLV only captures the financial value of the customer to the organization (Reinartz & Kumar, 2000). The other value, non-financial or relational value, which refers to relationship worth in the business-to-customer context, has not been commonly captured yet (Berry, 1995). The current CLV scheme has not yet accounted for the value of customers beyond purchasing behavior and this could mislead organization in loosing valuable customers from targeting efforts (Bolton, Lemon & Verhoef, 2004).
The value of customer to a firm should not only be determined based on profit related matters but also social interactions which can generate significant future profits for the firm (Hogan, Lemon & Libai, 2003). Customer can also value to the firm through indirect means in which customers voluntary contribute in performing organization' responsibilities, such as: helping the firm to attract other customers, to retain some current customers, and also to provide guidance to the firms (Phillips, 2007). The non-financial behaviors include voluntary partnership activities by customer by means of spreading word-of-mouth (WOM), making business referrals, giving references & publicity, and offering information and feedback to the organization (Bowen & Shoemaker, 2003). These behaviors may have an indirect effect to the profitability of an organization.
In order to retain customers who are profitable to organization, company develop a relational program. A relational program is a tool for organization to retain customers who are profitable and to build customer loyalty by offering incentives (Rust, Zeithaml & Lemon, 2000). This concept has been widely implemented in many business areas including airline industry. Initiated by American Airlines with its frequent flyer program (FFP) known as "AAdvantage" in 1981, FFPs become the largest membership of loyalty program with more than 120 million members enrolled in one or more of the 200 FFPs globally (McCaughey & Behrens, 2011). FFP awards generally reward loyal and frequent customers in the form of "loyalty currency" which can be used for free and upgrade flights, shop products, and other services.
The primary goal of the FFP is to retain a base of committed customers who are most likely to contribute to the profitability of an airline (Suzuki, 2003). However most airlines have very little understanding of their FFP members and have a little knowledge about their most valuable customers (O'Connell, 2009). This program has been associated with the difficulty and restriction of the redemption policies. There were only around 28% of frequent flyer points being redeemed (Greenberg, 2008). This fact leads to the question whether FFP is indeed beneficial for the members as well as to the airline issuer considering high cost involved in managing this program (Yang & Liu, 2003). The research on the influence of rewards and the redemption policy on the attachment to the organization have not been examined yet (Melancon, Noble & Noble, 2011).
While FFPs have attracted a great deal of attention in the transportation and marketing literatures, there has been no study on the effect of frequent flyer program on the nonfinancial worth generated by the members to the airline. As far as the context of this nonfinancial customer behavior to organization, the first and only study on the effect of loyalty reward on the non-financial value of customer to organization (business-to-customer exchanges) ever conducted is study of indirect influence on perceived reward to relational worth by Melancon et al. (2011). The study was carried out in a specific geographic area (US) and on a specific type of relational program (paid and non-accumulated type of rewards) in the context of a professional sports team and a fictional hotel reward.
The effect of loyalty reward onto the nonfinancial value in the context of non-paid and accumulated type of rewards, such as airline frequent flyers, has never been studied yet. Hence, the non-financial value of FFP's members to the airline related to the loyalty prog ram has not been known yet. Furthermore, the culture distinction between Indonesia and US might lead to the different outcomes on the social value of customers. High and low context cultures were cited as important role in determining an individual behavior (Hall, 1976). Indonesian with a highcontext culture has careful enculturation, such as: the meaning of individual behavior and speech changes depending on the situation or context, and non-verbal messages are full of important and intended meanings. On the other hand, American with a low-context culture usually expressed intensions in verbal, such as: a person's meaning should be explicit and not taken for granted, and propositions have to be justified and opinions defended openly (Johansson, 2009). Moreover, the study of Melancon et al. (2011) has not yet covered the impact of direct versus indirect relationship on perceived reward to relational worth.
For this reason, this paper is intended to analyze the effect of loyalty reward on the nonfinancial worth of the members to the airline by examining the relational worth outcomes from direct versus indirect relationship. Scope of the study covers analyzing of a non-paid and accumulated reward program in the context of FFP offered by airline in Indonesia. Research is referred to FFP of one of the established airlines in Indonesia. The analysis could not be examined to loyalty program on the other Indonesian carriers due to insignificant number of membership (Globalflight, 2013). The study examined how perceived rewards offered through FFP develop the expected relationship outcomes. This study becomes unique by accessing direct FFP relationship between the airline and the members to examine the financial value of FFP members. Research with access to actual FFP data from an airline is still uncommon (McCaughey & Behrens, 2011). Meanwhile, to our knowledge, there has been no research on frequent flyer program of Indonesian airlines concerning customer valuation. Hence, the identification of the value of FFP members to the airline has not been thoroughly observed yet. This result does contribute to the literature on relational reward programs and suggestion to the industry in practice about how frequent flyer program influences non-financial value of customers to organization.The rest of the paper will be organized as follows, after the introduction, we describe the hypotheses development and the method in Section 2 & 3, followed by the result and discussion in Section 4 & 5, accordingly. Finally, Section 6 gives the conclusion remarks.

Hypotheses Development
The research is referring to Customer Relationship Management (CRM) as the grand theory. Blattberg, Malthouse & Neslin (2009) emphasis CRM context in enhancing customer relationship, involving: acquisition, retention and relationship development. CRM enables organization to increase the value of the customer base through better relationships with the customers by providing customized services based on the customer database (Kotler & Keller, 2006). The building customer relationship is implemented by means of relationship marketing strategy.
The research explores the effectiveness of Frequent Flyer Program (FFP) and how the FFP transforms into non-financial value outcomes. Relational Worth is defined as the non-financial behaviors that contribute to the source of a customer's value. The hypotheses are proposed based the approaches on relational benefit, relationship quality, and relational marketing outcome.

Influence of Perceived Rewards on
Relationship Quality

Perceived Rewards
Perceived rewards are defined as the reward type and structure customer perceived from a particular organization that can influence commitment of the customer to organization. A relational benefit is perceived reward benefit -other than the core service itself -obtained by customers as a result of having a long-term relationship with the organization (Hennig-Thurau, Gwinner & Gremler, 2002). Perceived rewards in this study are measured through three variables: economic reward benefit, social reward benefit, and controlling reward policy.
Economic reward benefit is monetary related reward in exchange for desired behavior. Economic benefit includes price discounts, frequency discount, volume discount, price reduction, and special rates (Gwinner, Gremler & Bitner, 1998;Berry, 1995). It is measured by a proxy of any financial incentives offered by organization to the customers in relationship marketing program (Hennig-Thorou et al., 2002). Economic reward benefits weaken the relational exchange between the two parties by developing emotional feeling (Deci & Ryan, 1985). Referring to Cognitive Evaluation Theory, economic reward benefits are expected to weaken intrinsic motivation because customers attribute their behavior focusing on getting the reward. Therefore, it is expected limited emotional bounding occurred in the relationship. Hence customers can be easily switched to competitor.
Social reward benefit is non-material rewards which increases internal enjoyment of behavior and internal reason of maintaining behavior. Social benefits have been considered to involve feelings of familiarity, personal recognition, friendship, personalized & customized services, bond, social treatments, and high status relative to the average customer (Berry, 1995;Phillips, 2007) and special treatments to customers, such as participating on exclusive events, better service, and friendship-like relationship (Berry, 1995;Gwinner et al., 1998). Social rewards strengthen the relational exchange between the two parties by developing emotional feeling. The rewards lead to engagement of customer to the organization through affective commitment and enhance the value of transactional behavior (Price and Arnould, 1999;Rust et al., 2000). It is measured by a proxy of the recognition by the employees, familiarity by the customers and friendship creation amongst them (Hennig-Thorou et al., 2002).
The reward policy is the context in which reward is structured. The contexts of reward policy aspects of economic and social benefits have an impact on the strength and nature of motivation (Decy & Ryan, 1985). Decy and Ryan posit that reward policy captured the easiness and flexibility of customers in getting the reward from the organization. The availability of reward options for the customers, limited or restricted option, is likely reduce individual autonomy while allowing choices likely increase feelings of competence (Decy & Ryan, 1985). Based on Cognitive Evaluation Theory' approach, the context and structure in which the reward is offered or redeemed influences consumer behavior as "controlling" as it is believed to reduce c o n s u m e r ' s s e l f -d e t e r m i n a t i o n a n d competence, leading to a reduced intrinsic motivation of the consumer to the offered rewards.
It is measured by a proxy of the easiness and flexibility of customer for the rewards (Melancon et al., 2011). This research adopted the above literature review by conceptualizing perceived rewards of frequent flyer program involving variables of social reward benefits, economic reward benefits, and controlling reward policy.

Relationship Quality
Relationship quality in the context of service industry is regarded as the nature of relationships between organization and consumers whom are rely on the organization because its employees' integrity; consumers feel confidence because the past historical performance has been consistently satisfactory (Crosby, Evans & Cowles, 1990). The conceptual of relationship quality has been shifted from traditional manufacturing quality control into relationship quality that stands out as a key role in relationship marketing (Gronroos, 1994). The expectation and interaction of organization with its customers has been identified as the main factors contributing to the development of relationship quality (Hennig-Thurau & Klee, 1997). Existing literatures consider various conceptualizations on relationship quality but amongst the most common constructs encompass customer satisfaction, trust, and commitment to the relationship (Crosby et al., 1990;Morgan & Hunt, 1994;Hennig-Thurau et al., 2002). Relationship quality in this study are measured through a proxy of three variables: affective commitment, normative commitment, and satisfaction.
Only individuals whom commonly interact with an airline will join the airline' Frequent Flyer Program (Chin, 2002). Therefore, this study will examine the relationship between relational benefit and relationship quality in a stable and continuing relationship (phase of maintenance). In the phase of maintenance, satisfaction and commitment are actively involved constructs of relationship quality (Dwyer, Schurr & Oh, 1987;Parvatiyar & Sheth, 2000;Wetzels, Ruyter & Birgelen, 1998). A trust is indicated only be active in the early process of relationship development but becoming latent in the maintenance of the relationship phase (Wilson, 1995). Hence, this research adopted the above literature streams by conceptualizing relationship quality during maintenance phase of frequent flyer program encompasses components of satisfaction and commitment.
The consumer motivation to maintain a relationship with an organization involves affective (desire-based) and or normative (obligation-based) commitments (Allen & Meyer, 1990).
Satisfaction is emotional or feeling reactions of consumers to the perceived difference between perfor mance appraisal and expectations (Oliver, 1980). On the contextual customer-organization relationship, nonfinancial relational benefits are significantly related to satisfaction with the service provider organization (Gremler and Gwinner, 2000;Price and Arnould, 1999). The benefits received from special treatments such as economic savings or customized services are expected to positively influence customer satisfaction with the service provider (Hennig- Thorau at al., 2002).
According to Hennig-Thorau and Klee (1997), satisfaction is related to the realization of customer social needs which then leads to emotional bond of the customer to the service provider. Therefore, either social benefits or economic benefits are expected to influence customer satisfaction. The introduction of variable satisfaction in the model of this study fills in the scientific gap of how customer efforts in obtaining rewards could influence the relational outcome in terms of relational worth between customer and organization. If expectations exceed performance, negative disconfirmation or dissatisfaction will occur. On the other hand, if performance is perceived to meet or exceed expectations, customer is satisfied. Satisfaction is measured by a proxy of the perceived difference between performance appraisal and customer expectation (Oliver, 1980).
Affective Commitment is defined as commitment of individu to stay with the providers because of emotional bonds with the organization and it is indentified with his/her involvement with the organization (Allen & Meyer, 1996). Customer with a strong affective commitment will keep retaining the relationship with the organization.
Affective commitment has also significant positive relationship to advocacy intention which is desirable promoting the organization to friends and family (Fullerton, 2003;Bowen and Shoemaker, 2003).
Harisson-Walker (2001) found a positive effect of affective commitment on WOM activity and WOM praise whereas the same relationship had not been found on sacrifice commitment. Affective commitment represents an intrinsic commitment of customers to maintain their relationship with the organization because their feeling of satisfaction and enjoyment interacting with the organization (Melancon et al., 2011). Consequently, those customers who are intrinsically motivated to engage in the relational program will likely develop emotional bond with the organization. Affective commitment is measured by a proxy of customer engagement based on emotional bonds with the organization (Allen & Meyer, 1990).
Normative Commitment is commitment of individual to stay with the providers because of obligation feeling, and it is indentified with feeling guilty or uneasy leaving the organization (Deci & Ryan, 1985;Allen and Meyer, 1991). Antecedent of normative commitments includes social norms or expectations of commitment to organization, cost of training employee, and rewards in advance, and member interdependence (Allen and Meyer, 1990;Gruen, Summers & Acito, 2000). Affective commitment is also believed to be antecedent for normative commitment because one requires emotional feeling first before having feeling of obligation. Normative commitment is measured by a proxy of customer engagement based on guilty or uneasy feeling if leaving the organization (Allen & Meyer, 1990).
The following hypothesis is proposed to test the relationship between Perceived Rewards and Relationship Quality. H1: Perceived rewards offered through FFP will have positive relationship with relationship quality

Relational Worth
Relational worth is defined as desirable social behaviors of customer toward the organization. This construct was also termed as "individual's extra role" in the relationship representing customer worth to the organization compared with other customers (Phillips, 2007). Relational worth is considered as social behaviors in non-monetary concept toward the organization as an outcome from valued relational exchanges with customers. This value, in addition to CLV, was believed to be a key component for organization to indentify customers who provide maximum profitability and equity (Bolton et al., 2004;Hogan et al., 2003). The first scholar developed the constructs of "relational worth" for examining customers' non-financial behavior was Melancon et al. (2011), amongst of them including word-of-mouth (WOM), immunity, openness, and acquiescence. The analysis of relational worth in terms of customer exertions in getting the reward is filling the gap of the previous research by Melancon et al. (2011).
Word-of-mouth is defined as a willingness to spread positive word-of-mouth about the organization (Anderson, 1998). WOM as social behavior is very valuable to organization whenever customers taking organization' duties by spreading positive word-of-mouth to the others outside the organization. WOM communication is a powerful tool in influencing other customers for future buying decision (Hennig-Thurau et al., 2002). Other scholar had even suggested that WOM is nine times more effective than advertising in influencing customer attitudes toward the product or service (Day, 1971). Identified WOM antecedent includes consumer degree of satisfaction or dissatisfaction with consumption experience (Reichheld and Sasser, 1990;Anderson, 1998).
Involvement or commitment is also regarded as an antecedent of product-related word-ofmouth (Richins, 1984in Anderson, 1998. Customer who has a positive feelings about the benefits generated from the relationship with the organization and who has emotional attachment to the organization produces positive word-of-mouth behaviors (Bendapudi and Berry, 1997;Hennig-Thorau et al., 2002). Affective commitment has a positive impact in developing customer' advocacy intention amongst friends and family (Fullerton, 2003;Bowen and Shoemaker, 2003). Normative commitment had also found for having positive relationship to WOM (Gruen et al., 2000), however other study resulted the other side (Phillips, 2007). It is measured by a proxy of the perceived relating pleasant, vivid, or novel experiences, recommendations to others, and even noticeable display (Anderson, 1998).
Immunity is described as a tendency to remain loyal to the organization and resist to attractive offer from competitor or negative information about the organization (Bolton et al., 2000). The behavior is also found affecting consumers in having lower expectations of service recovery (Hess, Ganesan & Klein, 2003). Hence, immunity is believed protecting organizations against losing customers regardless of rising prices or lower levels of service (Bolton et al., 2000). Customer who has emotional feelings with the organization will produce immunity behavior against the negative information from competitors (Phillips, 2007). In the meanwhile, customer who feels pressure or controlled in the relationship is likely to the pursuit in alternatives (Bendapudi and Berry, 1997). Affective commitment had positive effects on retaining customer from other alternative offerings (Fullerton, 2003;Bendapudi and Berry, 1997;Allen and Meyer, 1990;Wetzel et al., 1998). Recent study also supports this relationship (Phillips, 2007). The impact of normative commitment to immunity had varied ranging from positively related (Gruen et al., 2000) and also weakening immunity (Bansal, Irving & Taylor, 2004;Phillips, 2007).
In the meanwhile, customer who feel pressure or controlled in the relationship is likely to the pursuit in alternatives (Bendapudi and Berry, 1997). Immunity is measured by a proxy of the resistance to competitor offerings (Bolton et al., 2000).
Openness is defined as a willingness to provide information to the organization without pressure by means of feedback, complaints, suggestion for improvement, provide requested information, responding to customer surveys, focus group (Hirschman, 1970;Melancon et al., 2011). In the contextual of relational behavior, openness is often related to an alternative to losing a customer with no indication of what went wrong in a relationship (Hirschman, 1970). Customers respond to disappointment in two different actions: by leaving the organization ("exit") or staying with the organization and complaint "voice". The exit-voice model was introduced by Hirschman (1970). Customers, who are willing to give honest information to the organization, help the organization in carrying out relationship marketing goals (Morgan & Hunt, 1994). Voice behavior has also related to overall satisfaction.
Individuals who feel comfortable and satisfied with their relationship with the organization are likely to share constructive criticism to the organization (Ping, 1993). In a contractual business-to-business exchange, parties are expected to proactively provide useful information to the partner as anticipation against unforeseeable events that may impact the partner. This behavior is defined as information exchange and recognized as one of important behaviors in marketing relationship (Heide & John, 1992). Openness is also referred in voluntary-relationship as willingness to become an advocate for organization including provide information to the organization (Bowen & Shoemaker, 2003).
Having accurate information from the customers, organization can adjust its service to the customer preference. Hence, customers having affective commitment are more likely to d e m o n s t r a t e o p e n n e s s b e h av i o r t o organization because of their belief that the information and complaint provided to the organization would help organization to rectify the problem and guide the business. Sense of being obligated to the organization would likely impact customer with normative commitment to openness behavior (Gruen et al., 2000). However, Phillips (2007) found that normative commitment negatively related to openness. Openness is measured by a proxy of the providing feedback, complaints, information (Bendabudi & Berry, 1997;Hirschman, 1970).
Acquiescence is defined as a willingness to adapt to necessary changes in arrangements, comply with request or policies of the organization as they arise (Melancon et al., 2011). Acquiescence or flexibility in dealing with the organization refers to customer tendency to be cooperative in all dealings with the organization in order to maintain the relationship with the organization. The dimension of acquiescence is adopted from extant literatures, including: flexibility (Ivens, 2 0 0 4 ) , a d a p t a t i o n ( W i l s o n , 1 9 9 5 ) , agreeableness (Morgan & Hunt, 1994) and adherence to request (Bendapudi & Berry, 1997).
Behavior flexibility is defined as individual readiness to adapt an existing implicit or explicit agreement to new environmental condition (Ivens, 2004). Individuals with emotionally and dedicated based relationship with organization would likely demonstrate acquiescence behavior (Bendapudi and Berry, 1997;Morgan and Hunt, 1994;Ivens, 2004;Heidi and John, 1992). Individual' commitment with sense of obligation would reveal acquiescence behavior (Gruen et al., 2000). Acquiescence is measured by a proxy of behavior flexibility as individual readiness to adapt an existing implicit or explicit agreement to new environmental condition (Wilson, 1995;Ivens, 2004).
The following table describes research roadmap of relationship between customer relationship quality and relational worth drawn from past academic works. It is believed that commitment reduces the tendency to leave the relationship with the organization, reduces uncertainty, increases acquiescence and increases the belief that conflict will be functional (Morgan and Hunt, 1994). The behavioral consequences of commitment include voluntary partnership activities by customer. These partnership actions vary from spreading WOM, making business referrals, provide references & publicity, and provide information and feedback to the organization (Bowen and Shoemaker, 2003). Commitment is also found to be a key driver for customer's resistance to change (Iwasaki & Havitz, 1998). An attitude of resistance to change could be viewed as resistance to leave the relationship with the organization or rejection for alternative offer from other organization. Therefore, it is believed that relationship commitment has a relationship with the dimensions of relational worth, such as WOM, immunity, openness, and acquiescence.
Furthermore, it is believed that customer satisfaction is a key driver for customer retention through customer loyalty and positive WOM communication (Anderson, 1998;Anderson and Sullivan, 1993;Fornell, 1992;Hallowell, 1996). The impact of customers' overall level of satisfaction is generally measured by the constructs of w i l l i n g n e s s t o r e c o m m e n d a n d recommendations to others (Anderson, 1998). Ekinci, Dawes & Massey (2008) identified the consequences of customer satisfaction include customers' overall attitude to the firm at the post purchasing point and intention to return were amongst the consequences. Satisfaction also forms intention on the customer side including intention to use (to return), intention to switch and also intention to recommend (Caruana, 2002).
The literature proposed the outcome construct of customer satisfaction are loyalty, retention and WOM of the extent to which customer informs friends and family on their experience with the firm. Customer retention involves emotional-cognitive behavior of customer in the form of willingness to recommend and intention to repurchase. Hence, the following hypothesis is proposed to test the relationship between Perceived Relationship Quality and Relational Worth.
H2: Perceived relationship quality through FFP will have positive relationship with relational worth

Influence of Perceived Rewards on Relational Worth
Accordingly, referring to the literature review and previous study discussed in earlier section there is also believed that perceived reward has a direct relationship with the dimensions of relational worth, such as WOM, immunity, openness and acquiescence. Hence the following hypothesis of the research is suggested.

Conceptual Model
The study analyzed how the rewards offered through FFP influence non-financial behaviors of the members to the airline. Derived from the above approached, a conceptual model for the current study is developed as presented in Figure 1.
In the context of analyzing loyalty reward program, this research expanded previous study of Melancon et al. (2011) on the influence of effort-type program to customer outcome behavior. The fact that the reward is obtained through exerted effort might bind a customer to the organization when he must abide a certain policy in earning/redeeming the rewards. This behavior aligns with CET literature in which customers feel of having more freedom in their autonomy or competence. The policy may increase customer commitment through influencing perceived autonomy or competence. When the informational dimensions of a reward are more salient, the reward will enhance intrinsic motivation (Tietje, 2002).
Understanding how customer exertions in getting the reward (ie. non-paid and accumulated reward program) enhances or undermines customer behavior is a critical area of study. The impact of customer efforts in obtaining rewards has not been considered yet on the past studies. The study also examined the impact of direct impact of perceived rewards offered through Frequent Flyer Program to relational worth generated by its members. Furthermore, this study is the first to consider constr uct satisfaction as a relationship quality contributing significant relationship to specific non-financial valuation. Satisfaction, in addition to commitment, is a critical construct during maintenance phase of relationship quality in continuing the relationship (Dwyer et al., 1987;Parvatiyar & Sheth, 2000;Wetzels et al., 1998). Hence, satisfaction is also critical construct to be included on the model of this study.

Research Design
Online survey was used to test the influence of frequent flyer program' reward benefit on the relationship quality and non-financial behaviors of the members. The online survey was conducted by Garuda Indonesia through the Garuda Frequent Flyer, GFF's system. The survey was run in August 2013.
Respondents were invited to participate in the study through Garuda' e-blast system which included an embedded URL link to the online survey hosted by googledrive. The survey was divided into two sections.
The first part enclosed questions on customer profiles and preferences of the reward benefits, while the second part covered main questions regarding customer non-financial valuation. Respondent who participated on the study was awarded an incentive of 300 GFF free miles credited to his/her GFF account.

Sampling Method
A stratified random sampling was generated by the system. Stratified sampling is determined by the tier level, which is the membership level of FFP based on the accumulated miles. is required for representing 32 itemquestionnaires. In term of representation of the population variance and desired precision, Krejcie & Morgan (1970) recommended an optimum sample size (n) of 384 for the population of GFF members at 0.05 of standard error. Hence, the sample size of 475 in the study is denoted as adequate representing the model.

Variable and Measurement
The variables and scales used in this study are adapted from previous studies with adjustment to the context of FFP. The perceived social reward benefit was measured by the modification scales of Hennig-Thurau et al. (2002). The construct was gauged through four indicator variables: feel special, special attention, special privileges, priority reservation given to customers. The perceived economic reward benefit was measured by the modification scales of Hennig-Thurau et al. (2002), such as: upgrade, free ticket, and nonflight package. The controlling reward policy was measured by the modification scales of Melancon et al. (2011). The construct was gauged through three indicator variables: flexibility, few limitations and easy redemption. The affective commitment was measured by the scales of Meyer & Allen (1991) through three indicator variables: emotional-bond, family and belonging.
The normative commitment was measured by the scales of Meyer & Allen (1991) through four indicator variables: guilty feeling, owe feeling, ethics feeling, and obligation. The satisfaction was measured by the modification scales of Oliver (1980) through three indicator variables: wise choice, good service, and overall satisfaction. The WOM was measured by the modification scales of Anderson (1998) through three indicator variables: promote, spreading positive news, and defense to negative news. The acquiescence was measured by the scales of Ivens (2004) and Wilson (1995)) through three indicator variables: adapt, comply, and adjust.
The openness was measured by the scales of Bendapudi & Berry (1997) and Hirschman (1970) through three indicator variables: feedback, complain, and provide information. The immunity was measured by the scales of Bolton (2000) through three indicator variables: uninterested in better offer, remain stay in the company, and reject competitor offer. All indicator is measured on a 6-point Likert scale where 1=strongly disagree, 2=disagree, 3=mildly disagree, 4=mildly agree, 5=agree, 6=strongly agree. A 6-point of Likert scale format was employed to avoid neutral answer from the respondents. All items and scale sources of the study are presented in Table 2.  Bendapudi & Berry (1997), Hirschman (1970) 3 Acquiescence Ivens (2004), Wilson (1995) 3 Immunity Bolton (2000) 3

Analysis Method
The analysis of non-financial value of the members to the company was conducted using two-step approach to SEM/Structural Equation Modelling (Anderson & Gerbing, 1988). The model developed for this study involved latent constructs that could only be approximated through observed or measured variables. Therefore SEM was used for its ability to incorporate latent variables into the analysis. The other analysis method, Path Analysis, only works for observable variables which can be directly measured. The first step consisted of assessment of measurement model fit and the second step consisted of analyzing the str uctural (theoretical) relationship hypothesized within the conceptual model. LISREL version 8.30 was used in this study.
The measurement model specifies how the latent variables are measured in terms of the obser ved variables. It describes the measurement properties (validities and reliabilities) of the observed variables. This includes the determination of construct validity of each of the scales presented in the model, including information relative to the testing of scale reliability, dimensionality and validity. LISREL version 8.30 was employed to test the measurement model. A Confirmatory Factor Analysis was conducted on the FFP member samples to check the reliability and validity of sub-model of the research design. Construct Reliability (CR) more than 0.7 and Variance Extracted (VE) more than 0.5 were deemed to have acceptable fit (Hair et al., 2010). For measuring the model, a ratio of factor loadings to standard error (lambda or loadings) more than 0.5 was classified as reliable (Joreskog & Sorbon, 1996).

Pre-Test
A pretest was conducted on FFP member samples (n=30) to provide validation that the questions were read and understood the same way by the respondents. The validity test was done by Pearson correlation, while the reliability test was conducted by Cronbach alpha test. All indicators passed the validity test with results between 0.704-0.975, and reliability tests showed more than 0.7 of Cronbach alpha. The results indicated that the validity and reliability of the questionnaire were acceptable. All of the questionnaires were positioned for full release in online survey to the respondent pool.

Descriptive Statistics
Four hundred seventy-five respondents were participated on the online survey. Sample distribution of GFF members for this study is dominated by male members (79%), Indonesian nationality (82%), age between 31-40 years (37%), having the lowest level of FFP's membership in the respected airline, namely blue tier (53%), graduated from university or equivalent (60%), working as government official or professional/manager each at 32%, and living in Indonesian country of region (82%). Most of the respondents have an exclusive FFP membership only with the airline (69%) and preferring flight-related redemptions, free ticket and upgrading (75%).

Measurement Model Assessment
The measurement model showed that the factor loadings for each construct were statistically significant (i.e. ranging from 0.65 to 1.0) and sufficiently high to demonstrate that the indicators and their underlying constructs were acceptable.
Reliabilities ranged from 0.79 to 0.94, and the variances extracted were above the threshold value from 0.63 to 0.83. These results indicate that the measurement model is highly reliable and reasonably valid. The research ideas and hypotheses were well operationally represented by the questionnaire items. The results of the CFA for each constructs are presented in Table 3.

Structural Model Assessment
The results of standardized loading error and t-value of the model were presented in Figure 2 and Figure 3.   Conclusively, perceived reward had a significant impact on producing non-financial value of customer through relationship quality. The perceived rewards offered through frequent flyer program play important roles as antecedents of relationship quality by means of commitment and satisfied customer. The rewards that create a relationship quality will enhance relational worth through WOM, immunity, openness and acquiescence.

Discussion
This paper investigates whether perceived rewards offered through FFP influence the non-financial value of the customers to the airline. Using a sample of 475 FFP members who participated in the online survey, the influence of frequent flyer program' reward benefit on the relationship quality and nonfinancial behaviors of the members were examined. This research is a continuing study on how perceived reward influence customer value of customer to organization previously conducted by Melancon et al. (2011).
While Melancon et al. (2011) examined the reward effect specifically on paid and nonaccumulated' reward program, this research examined non-paid and accumulated reward program in which involving customer exertions in getting the reward. The results expanded the study of relation worth concluded by Melancon et al. (2011). This study covered the impact of direct versus indirect relationship on perceived reward to relational worth that has not been examined yet by Melancon et al. (2011). The result contributes to the knowledge of relational program in general and loyalty reward program in specific.
The result indicates that perceived reward benefits and relationship quality were significantly related to perceived relational worth, while perceived reward benefits was also significantly related to relationship quality. It is found that the impact of perceived reward benefits on non-financial value is higher if moderated by relationship quality than directly connected to non-financial value.
The perceived reward benefits offered through frequent flyer program play important roles as antecedents of relationship quality, such as perceived commitment and satisfaction value. The rewards and relationship quality enhanced relational worth through WOM, immunity, openness and acquiescence. This suggested that perceived reward benefit obtained by FFP members would result on social behaviors, as follows: q spreading positive word-of-mouth, taking on some of the promotional responsibilities of the airline including to influence family members and friends to use the airline (WOM); q keep staying with the airline, being not interested in better offers from competitor, and being remain staying with the airline even though they heard some negative information about the airline (immunity); q providing feedback or suggestion for any experiences of imperfect services, offering suggestion for improvement and any information for helping the airline to conduct better services (openness); q adapting to necessary changes on the relationship with the airline, complying with request or policy of the airline, adjusting the flight to match to the airline schedule (acquiescence).
Perceived rewards would develop relationship quality through the following: q emotional-bond : through feeling as part of the family, having a strong sense of belonging and emotionally attached to the airline, q obligation-bond: through feeling owe a great deal to the airline, having an obligation remain with the airline, feeling guilty for leaving the airline and having a sense of obligation to remain as a customer to the airline, q satisfaction of FFP members: by means of member feeling of using the airline is a wise one, member always being delighted with the airline' services, and the feeling of satisfaction.
Hence, perceived rewards would produce higher level of relational worth through committed and satisfied consumers than if the perceived reward transform directly to relational worth.

Managerial Implication
Perceived rewards offered through loyalty reward program were perceived higher value of relational worth through moderating by relationship quality than a direct relationship. Perceived rewards indirectly generate nonfinancial value of the customers through relationship quality of commitment and satisfaction. Awarded rewards without developing any engagement feeling would be easily forgotten by customers. Therefore, direct reward has not been successfully increased social value of customer to organization.
As a managerial recommendation, rewards offered through FFP should encourage emotional and obligation bonds and also feeling of satisfaction to members. reward program, this research examined nonpaid and accumulated reward program in which involving customer exertions in getting the reward. The results expanded the study of relation worth concluded by Melancon et al. (2011). This study covered the impact of direct versus indirect relationship on perceived reward to relational worth that has not been examined yet by Melancon et al. (2011). The result contributes to the knowledge of relational program in general and loyalty reward program in specific.
To academics world, this research will be beneficial as additional source of research that can be cited or experimented in other context or demography. This research develops nonfinancial values of customers to the organization especially in airline industry. The result will also propose an improvement of member value proposition from the point of view of beyond financial value which is social benefits. How customers are voluntarily taking part to the organization duties in respect of the perceived rewards that were obtained by the customers through exerted efforts. It will enrich the loyalty approaches especially in airline industry. This research has taken steps to break down and understand customer behavior to reward program offered through frequent flyer program in airline industry by concluding the following result.
1. Perceived rewards had a significant impact on generating non-financial value of customer by means of social behaviors involving WOM, immunity, openness and acquisition. 2. Perceived rewards offered through loyalty reward program were perceived higher value of relational worth through moderating by relationship quality than a direct relationship. Perceived rewards generate higher value of relational worth by an indirect relationship through relationship quality as indicated by Beta â=0.88, compared to a direct relationship that generates lower Beta â=0.1. Committed and or satisfied customers encourage perceived rewards generating high level of non-financial value.
The indirect relationship would enhance customers to generate better social behaviors in terms of: spreading WOM, reducing the tendency to leave the relationship with an o r g a n i z a t i o n , p r o v i d i n g i m p o r t a n t i n f o r m a t i o n a n d f e e d b a ck f o r t h e improvement of the organization, and increasing flexibility in dealing with the organization.

Future Research
Future research might explore these hypotheses in various contexts of non-paid and accumulated programs as well as involving many FFP' communities from other geographic area. A study of the impact of direct versus indirect relationship on perceived reward to relational worth in geographic area of United States would not only complete the previous study of Melancon et al. (2011) but also offer an opportunity to compare with the result of current study as well as access the effect of culture distinctions on customer' social behavior. A continuing research on the same program with an effect of multi lateral alliance would be also interesting.
This study assumed that the relationship between relational benefit and relationship quality in regard to FFP is occurred in a stable and continuing relationship. Future research might explore the relationship between relational benefit and relationship quality in the other relationship lifecycle, for example in the early stage of relationship. Variable trust is believed acting as the active construct during the phase of initiation (Wilson, 2005). A continuing research involving additional latent variable, trust, in the framework model may provide different outcome.

Conclusion
This research is a continuing study on how perceived reward influence customer value of customer to organization previously conducted by Melancon et al. (2011). While Melancon et al. examined the reward effect specifically on paid and non-accumulated'