Retail Loyalty Programs in Malaysia: The Relationship of Equity, Value, Satisfaction, Trust, and Loyalty among Cardholders

This study examines loyalty programs in Malaysia in terms of perceived equity, value, program satisfaction, program trust, and store loyalty. A total of 400 questionnaires were collected from cardholders in Klang Valley, Malaysia. The hypotheses were based on social exchange and equity theories. The structural equation model that assessed the relationship between the proposed variables was tested using AMOS 6.0. The findings revealed that perceived equity and perceived value of the loyalty program significantly influence program satisfaction and trust. Program trust was the most important predictor of store loyalty. Limitations of the study and recommendations for future research are discussed.


Introduction
Loyalty program has become one of the most popular retention strategies used by retailers to increase customer loyalty in mature retail markets. Few researchers (Gomez et al. 2006) have studied loyalty programs as a means to intensify the relationships and to stimulate customer loyalty. In fact, the Journal of Retailing devoted a special issue to customer loyalty to stimulate research on pertinent and current retailers' related to-pics such as loyalty programs and store loyalty (Meyer-Waarden 2007). Over the past few years, loyalty programs, i.e., loyalty cards, store cards, or frequency programs, have become a key component of Customer Relationship Management (CRM), serving a critical role in developing relationships, stimulating product and service usage, and retaining customers. Based on the realization that it is much less expensive for a business to retain its existing customers than it is to acquire new ones, most companies are adopting loyalty programs to retain its existing customer base. Marketers have implemented loyalty programs in a wide variety of industries (Blattberg, Deighton 1996). In fact, the effect of loyalty programs have been studied within the context of various industry settings such as automotive industry (Stausse et al. 2001), packaged goods (Roehm et al. 2002), fi nancial services (Bolton et al. 2000), airlines (Long, Schiffman 2000), retail stores (Noordhoff et al. 2004), telecommunications (Roos et al. 2005), or casinos (Palmer 2003).
The growth of loyalty programs around the world is remarkable. McKinsey (2001) reported that about half of the ten largest U.S retailers have launched loyalty programs, and the rate is similar among U.K. and Dutch retailers (Cigliano et al. 2000). In its efforts to boost the burgeoning tourism industry, the Malaysian government offers consistent encouragement and environment to foreign retailers including department and superstores operators. Consequently, the rise and expansion of retailers in Malaysia has contributed to a high competition among retailers leading to the growth of loyalty programs, which have become a part of structured marketing efforts used to build store traffi c, increase basket size, and create deeper relationship ties with customers. Accordingly, research by consulting fi rm Frost & Sullivan revealed that the loyalty program business in Malaysia is currently worth MYR 2.4 billion and is projected to grow to MYR 3.5 billion by 2010 (Ganesan 2006). Due to its popularity, extensive fi ndings demonstrate the rationale for adopting a loyalty program (Kim et al. 2001) and provide empirical evidence for its effectiveness (Bridson et al. 2008;Omar et al. 2007).
Despite its popularity, many researchers have questioned whether loyalty programs actually lead to customer loyalty or whether customer loyalty is due to some other factor (Capizzi, Ferguson 2005). The effectiveness of a loyalty program is likely to depend on the fairness as well as the value of the program. Critics of loyalty programs argue that perceived fairness is important in determining the length and quality of a relationship (Bolton et al. 2000). Parker and Worthington (2000) posit that when cardholders begin to calculate the equitability of the loyalty program and feel that it was unfair and/or perceive that a competitive program is of a higher value, it may affect their loyalty to the retailer. Several empirical studies suggest that equity is one of the signifi cant factors infl uencing satisfaction in social interactions (i.e. buyer-seller) that subsequently infl uence the formation of long-term relationships and increase favorable behavioral consequences (Olsen, Johnson 2003). However, to the best of the authors' knowledge, prior studies did not examine the impact of perceived equity and perceived value on satisfaction within the context of a retail loyalty program. This is a critical gap in the literature, given that loyalty programs are popular among retailers in retaining their customers globally. Hence, it is important for retailers to pay attention to factors that would improve their loyalty program. This study aims at fi lling the gap by studying the relationship between programs' perceived value, perceived equity, and their impact on cardholders' program satisfaction, program trust, and store loyalty within the context of a retail loyalty program.
Although some studies attempted to assess the relationship between consumer fairness and fi rms' use of loyalty programs to collect proprietary information (for example Lacey, Sneath 2006), this study has not investigated the impact of program perceived equity on attitudinal or behavioral loyalty. Considering that perceived equity and perceived value signifi cantly relate to satisfaction and trust (Grace, O'Cass 2005;Oliver, Swan 1989;Singh, Sirdeshmukh 2000;Wat, Shaffer 2005), whether these relationships would hold within the context of a loyalty program is questionable. Furthermore, studies on loyalty programs used mostly samples drawn from the western regions. With a few exceptions, for example Omar et al. (2007Omar et al. ( , 2010, investigated the impact of loyalty program satisfaction on customer loyalty among cardholders in Asia. Given their popularity, it is critical to investigate the effectiveness of loyalty programs as a retention strategy among cardholders in Malaysia. This study could provide greater conceptual and practical insights with respect to the design and application of retail loyalty programs. The fi ndings from this study could offer potential implications for the retailers that offer loyalty programs to their customers such as assisting the industry managers as well as marketers in formulating effective strategies to foster store loyalty.

Literature review
Loyalty programs have been around since the latter part of the 20 th century. They started in early 1980s and expanded from redeeming Betty Crocker points and S&H Green Stamps to earning free fl ights and hotel rooms. In fact, studies report that loyalty programs have blossomed and span to various industries including retail, hotel, airlines, financial and entertainment industries (Lecey and Sneath 2006). Their success was partly due to the simplicity of the concept like giving out seats and rooms as a way to reward returning customers. Depending on the fi rm's philosophy and program objectives, the fi rm can automatically offer loyalty program membership to all customers or only to certain customers who qualify for membership based on cumulative purchase of a given product or a service. Loyalty programs are different from the normal sales promotions since they provide customers with added incentives to continue their marketing relationship with the fi rm through repeat purchase and loyalty.
In Malaysia, most loyalty programs use a card to collect points and offer gifts or vouchers to redeem these points. According to the Nyang Koon Seng, CEO of Advance Information Marketing Sdn. Bhd. (AIM), there are two types of loyalty programs (or loyalty cards / schemes) in Malaysia. The fi rst type is the multiple-merchant loyalty program or multiple branding schemes. This loyalty program gives members the leeway to collect points on different merchandise from diverse merchants. The second loyalty program is a single merchant program. This is an in-house program where cardholders can only redeem points from a particular merchant's outlet. Most department stores and superstores chains in Malaysia have their own single merchant loyalty programs. Under the single merchant program, the program management is independent of any of the partners. The program bestows many benefi ts and privileges on the card members such as, discounted supermarket items, free insurance coverage, gift redemptions, cash vouchers, free parking, special member promotions, free magazines etc. It is important to highlight that this study focuses on the single merchant loyalty program owned by major superstore chains in Malaysia. Concentrating on the single merchant program rather than on the multiple-merchant loyalty program allows the authors to examine if a single program could lead to customer loyalty to the store.

Program perceived value
Researchers often categorize perceived value into two parts. Benefi ts received by the customer (e.g., quality, benefi ts, worth and utilities) represent one part and sacrifi ces made (e.g., price, time and effort) represent the second part (Dodds et al. 1991;Zeithaml 1988). Considerable number of studies has suggested that the customer's value, or derived benefi t, plays a signifi cant role in determining his or her long-term relationship with, or loyalty to, the fi rm (Huber et al. 2001). Research reveals that customers are less likely to switch fi rms if they can understand the actual economics, time, and energysaving value of staying in a relationship (Gwinner et al. 1998). In fact, O'Malley (1999 highlighted that if the consumer does not perceive value in building relationships with a fi rm, then they might only engage in a relationship because a better option is not available elsewhere.

Program perceived equity
Philosophy, political science, organizational sciences, and economics have all studied the concept of fairness for a long time. Findings from these diverse areas suggest that fairness forms a fundamental basis for maintaining relationships during social exchange (Lind et al. 1993). Swan and Mercer (1982) propose that equity involves primarily the relationships between the marketer's and the consumer's net gains, which have to be in balance. Interestingly, research found that if a customer perceives that the seller treated him/her equitably, a strong bond between the seller and the customer might form because the buyer considers the relationship important (Bolton et al. 2000;Nobel, Philips 2004). Further, research has suggested that fair interpersonal treatment contributes to satisfaction (Bitner et al. 1990) and subsequently affects future purchase intentions (Fisk, Young 1985).

Program satisfaction
Consumer satisfaction is the most effi cient and least expensive source of market communication because consumers who are satisfi ed with a product or service will be more likely to disseminate their favorable experiences to others. On the other hand, if they are dissatisfi ed, they will spread unfavorable appraisal of the product or service (Dubroski 2001). Scientists often perceive satisfaction as the result of all activities carried out during the purchase and consumption process and not only of observation and or direct product or service consumption (Oliver 1996). Satisfaction depends on the experience of using the product or services. Hence, satisfaction is defi ned as "a cumulative evaluation of a customer's purchase and consumption experience to date" (Lervik, Johnson 2003).

Program trust
Scientists often view trust as an essential ingredient for building and maintaining successful relationships (Berry 1995). According to Wilson (1995), trust is a fundamental relationship model building block and is included in most relationship models. For that reason, earning customer trust is usually one of the earliest goals in building a long-term relationship with customers (Peppers, Rogers 2004). Trust has been defi ned by Morgan and Hunt (1994) as "one party's confi dence in an exchange partner's reliability and integrity". Central to this defi nition of trust is the notion of credibility or the beliefs that one can rely on another party to fulfi ll written and spoken promises.

Store loyalty
With current environmental changes, such as increase in competition, the managers face greater challenges to increase and retain store loyalty. Gomez et al. (2006) noted that loyalty programs are marketing strategies based on offering an incentive with the aim of seeking customer loyalty to a retailer. According to Sharp and Sharp (1997), the effectiveness of marketing efforts to promote the relationships and loyalty should be evaluated in terms of behavioral changes they bring about. Loyalty consists of two distinct components, attitudinal and behavioral (Evanschitzky et al. 2006). Attitudinal loyalty measures the emotional and psychological attachment inherent in loyalty (Bowen, Chen 2001). Behavioral loyalty refl ects unique combination of behavioral indicators such as purchase frequency and purchase amount (De Wulf et al. 2001). Rauyruen and Miller (2007) defi ned loyalty based on two underlying concepts, the willingness to continue the transaction with the seller, and the willingness to do business with the seller relatively independently of own interest. In addition, the concept of store loyalty is further defi ned by Knox and Denison (2000) as "the consumer's inclination to patronize a given store or chain of stores over time".

Theories and hypotheses
A general relationship marketing theory has proposed that both consumers and businesses engage in relationships for the same reasons: to optimize expected equity (a trade-off between exchange costs and benefi ts) and to increase the predictability of exchange outcomes (Christy et al. 1996). Hence, both equity theory and social exchange theory are useful frameworks for investigating the reasons for engaging in relationships by explaining the impact of perceived value and perceived equity of the programs on satisfaction, trust, and customer loyalty. Generally, in any loyalty program, both the cardholders and providers will have to give up something in exchange, with the expectation that the advantages of this exchange outweigh the cost (De Wulf et al. 2003). If a customer has joined a particular loyalty program, the participant will have some freedom to select a provider and may or may not pay some form of entry fee in exchange for the benefi ts available from a loyalty program. In return, the provider accepts the burden of offering the customer some benefi ts, a concession which helps fi rms retain customers. Equity theory is one of the guiding theories in this study because it provides a useful framework for understanding cardholders' behavior.
The under-investigated but related issue of perceived fairness is of primary importance since previous studies (Bolton et al. 2000) indicate that perceived fairness is important in determining the length and depth of a relationship. Firms that fail to establish an image of fairness often cannot gain potential customer confi dence needed to establish loyalty. Another popular relationships theory that is also crucial to this study is social exchange theory (Fisher, Bristor 1994). According to Bagozzi (1995), the principle of reciprocity states that people should return good for good, in proportion to what they receive. Most of the existing literature has noted that relationships, in which rewards outweigh the costs, will yield enduring mutual trust and attraction (Morgan, Hunt 1994).
Thus, social exchange theory and equity theory seem promising for explaining consumer-retailer relationships within a loyalty program context. The ideas and constructs in both theories help explain marketing relationships and are in line with retail situations (De Wulf 1999). Only a limited number of studies support perceived equity as a direct antecedent of perceived value. Empirically, only a few studies found that customer perceptions of overall fairness relate to customer perceptions of value directly and in a positive direction (Goodwin, Ross 1992). Several researchers (Bitner et al. 1990;Hutchinson et al. 2009) found that perceived equity had a signifi cant and direct infl uence on both value and satisfaction. In the social contract setting, the perceptions of fairness are likely to promote enhanced feelings of job satisfaction and employee trust in the supervisor for making a good faith effort to be fair (Wat, Shaffer 2005). A study on the treatment of sales persons by supervisor found that equity (interactional and distributive fairness) had a strong positive effect on supervisor trust (Ramaswami, Jagdip 2003).
Consistent with the evidence presented by previous researchers, it is possible to expect that program perceived equity would affect the customer's perceived value, satisfaction and trust in the retail loyalty program. We therefore postulate that: There is considerable merit to the view that creating and offering value to customers might lead to satisfaction (Gale 1994). Interestingly, many past studies (Ball et al. 2006;Grace, O'Cass 2005) found that customers' perceived value had a strong and signifi cant impact on satisfaction. The proposed relationship between perceived value and satisfaction is supported by value disconfi rmation experience when customer expects to receive benefi t greater than the cost on each purchased made. An unexpected decrease or increase in the cost or the benefi t after the purchase alters the perceived value, leading to either a decreased or an increased customer satisfaction. With respect to the link between the perceived value and trust, Morgan and Hunt (1994) suggested that the benefi ts of the relationship are precursors of trust. Shah Alam and Yasin (2010) believe that people trust an online brand based on their own experience of benefi ts (such as quality information, security, and reputation) they received. Thus, one can consider that the perceived value received from a loyalty program is part of the benefi ts of the relationship. The proposed perceived value-program trust is consistent with the early fi ndings that perceived value is an antecedent of trust (Singh, Sirdeshmukh 2000). In addition, Moliner et al. (2007) showed that perceived value of a tourism package is an antecedent of the tourist's trust in the travel agency. Based on the reported research, we proposed the following hypotheses: H 2a : Within the retail loyalty program card context, program perceived value (PPV) will have a signifi cant positive relationship with program satisfaction (PS). H 2b : Within the retail loyalty program card context, program perceived value (PPV) will have a signifi cant positive relationship with program trust (PT).
Research links satisfaction to the trust in a relationship (Bauer et al. 2002). Accordingly, Flavian et al. (2006) suggested that trust should be greater when the business or product provide a greater consumer satisfaction. Similarly, Singh and Sirdeshmukh (2000) proposed that satisfaction had a direct effect on post purchase evaluation of trust. Hence, based on previous literature, we predict following relationships: H 3a : Within the retail loyalty program card context, program satisfaction (PS) will have a signifi cant positive relationship with program trust (PT).
Indeed, prior studies have found satisfaction to be a leading factor in determining loyalty and profi tability (De Wulf et al. 2001;Zeithaml et al. 1996) as well as a key factor in generating positive customer word-of-mouth behavior (File et al. 1994). Moreover, Dick and Basu (1994) suggested satisfaction as an antecedent of relative attitude because without satisfaction, consumers will not hold a favorable attitude towards a brand as compared to other alternatives available. Based on this, we state the following hypothesis: H 3b : Within the retail loyalty program card context, program satisfaction (PS) will have a signifi cant positive relationship with store loyalty (SL).
Few scholars, such as Anderson and Weitz (1989) and Berry (1995), suggest that marketing relationship builds on the foundation of trust, which seems to show a strong correlation with loyalty (Jarvenpaa et al. 2000). Previous researchers pointed out that trust is a critical factor in stimulating purchase and loyalty (Ball et al. 2006;Flavián et al. 2006). Inspired by these ideas and fi ndings, we investigate these hypotheses: H 4 : Within the context of retail loyalty program, program trust (PT) will have a signifi cant positive relationship with store loyalty (SL).

Method
The target population was adult consumers (over 18 years of age) who have been members of a retail loyalty program for at least a year and live or work within the area of the Klang Valley, Malaysia. The authors selected the respondents with at least a one-year membership since it this period is suffi ciently long to facilitate quitting from the program in case of disappointment (Morais et al. 2004). We selected this particular location because the capital of Malaysia is situated in Klang Valley. On average, households living in this area spent 1.5 times more compared to households living in rural areas.
Most of the modern retail establishments in Malaysia are located in the Klang Valley (Euromonitor 2001). Because loyalty program is still in its early stage in Malaysia yet popular among those living in urban areas, it is more likely that individuals in more populated urban areas would use this program, and therefore, it is reasonable to confi ne the target population to those in the Klang Valley.
Among the samples collected, female respondents (75%) were the majority. Regarding participants' age, 41% were 30 to 39 years old, 39% were 20 to 29 years old, 13% were 40 to 49 years old, and 7% were older than 50. Married respondents accounted for 63% of the sample and single respondents for 35.5%. The rest of the respondents were divorced (1.5%). About 41.2% of the respondents were members of a loyalty program for 1 to 2 years, 28.8% were members for 3 to 4 years, and 30% were members for 5 years or more.

Procedure
The data for this study was collected using self-administered questionnaires and distributed using the quota sampling 'drop off and collect' technique. This study used gender as the key mechanism to control the composition of the sample. The ratio of men to women in the sample was 1:3 Studies conducted in Malaysia and many other countries suggest that women make up a signifi cantly larger percentage of purchasers (ACNielsen  2002). Indeed, research shows that women are more likely to use a retail loyalty program rather than any other price reduction methods (Harmon, Hill 2003). In order to ensure that all potential respondents fall within the sampling quota, we selected sample carefully to ensure that we would not select the participants based on perceived friendliness or other characteristics that might infl uence the researcher's selection. To avoid such bias, the authors personally contacted the key personnel in several organizations to negotiate access to their staff. The key personnel explicitly explained the criteria to the respondents. Upon agreement from the key personnel and before an appointment was set for delivery of the research instrument, the authors requested a list of names of those who are interested in participating. Each respondent would select a particular loyalty program that he/she is a member of and would like to focus on in the questionnaire. Overall, 400 out of 460 participants completed questionnaires, representing a response rate of 87%.

Measures
All of the selected items in the questionnaire were based on previous literature and measured on a fi ve-point Likert scale from (1) "strongly disagree" to (5) "strongly agree" except program satisfaction. Likert scale approach is favored because it maintains the scale interval-level properties (Allen, Rao 2000). We conceptualize program perceived value as the customer's overall assessment of the loyalty programs in terms of all the relevant benefi ts and rewards incurred by the program's members. In order to measure this construct, we utilized a fi ve-item measure to measure the respondents' perceived value towards the retail loyalty programs (Duman, Mattila 2005). We defi ned program perceived equity as what participants give up or sacrifi ce in order to subscribe and acquire benefi ts in a loyalty card program. The scale was adapted from previous researchers (Oliver, Swan 1989). We utilized four items to assess the degree to which respondents perceived the retail loyalty program as fair, particularly in terms of the program's outcomes and treatment. In terms of the program satisfaction scale, respondents were required to rate their overall feeling towards the retail loyalty program that they chose to focus on in this study. Five semantic differential scales, commonly used in satisfaction studies (Oliver and Swan 1989), measured overall satisfaction. We conceptualized trust as the level of members' confi dence that their expected behavior will lead to valued outcome by participating in the particular loyalty program. Program trust was measured on a fi ve scale commonly used in previous studies (De Wulf et al. 2003;Kim, Cha 2002). Finally, we conceptualized store loyalty as high positive attitudes towards a particular store and repeat purchase behavior. Five item scale based on existing measures (Yi, Jeon 2003;Zeithaml et al. 1996) was used to measure store loyalty.

Results
A two-step approach was employed; examination of the measurement model was followed by an examination of the structural model used to test the hypothesized relationships ). The Structural Equation Modeling (SEM) procedure enabled us to evaluate how well a proposed conceptual model that contain observed variables and unobservable constructs fi ts the collected data (Bollen 1989). A confi rmatory factor analysis (CFA) was conducted to test the robustness, reliability of the scales (via AMOS and the maximum likelihood estimation technique), to confi rm the factor loading of the fi ve constructs (i.e. program perceived equity, perceived value, satisfaction, trust and store loyalty) and to assess the model fi t. The model adequacy was assessed by the fi t indices suggested by Hair et al. (2006). The analysis showed an excellent overall fi t of the model as indicated by the comparative fi t index (CFI), normed fi t index (NFI), relative fi t index (RFI), and root mean square error of approximation (RMSEA) with values of 0.98, 0.96, 0.95 and 0.03, respectively. However, chi-square statistic was signifi cant (χ2 = 186.70, df = 127, P = 0.00), which is common given the large sample size (Bagozzi, Yi 1988). A better measure of fi t is chi-square over degrees of freedom. This ratio for our model is 1.47, which is within the suggested three to one bracket (Chin, Todd 1995). Thus, the measurement model fi ts well enough to suggest adequate validity and warrant a closer look.
We assessed convergent and discriminant validity with several test suggested by Anderson and Gerbing (1988). Table 1 lists the standardized loadings, cronbach alpha, composite reliabilities, and variance extracted estimates (AVE). Cronbach's alpha of all dimensions ranging from 0.86 to 0.89, this clearly indicates that the scales used in this study were highly reliable (Nunnally 1978). Moreover, all standardized loadings were greater than 0.60, thus the criterion for convergent validity was satisfi ed ). We calculated the composite reliability and variance extracted using Fornell and Larcker's (1981) formula. The composite reliability ranged from (0.90 to 0.93), all exceeding the minimum reliability standard of 0.70. Variance-extracted estimates ranged from (0.75 to 0.78), all exceeding the recommended lower standard of 0.50 (Fornell, Larcker 1981). All tests supported convergent validity of the scales.
We assessed discriminant validity also with tests recommended by Anderson and Gerbing (1988). The correlation index among factors was low and moderate and did not exceed the cut-off point of .85 (Kline 2005). This supports discriminant validity of the scales (Churchill 1995). An additional discriminant validity criterion was also established when AVE for each construct was greater than the squared inter-construct correlations between the relevant constructs (Fornell, Larcker 1981). All constructs passed this test (Table 2). Common method variance (CMV) was also assessed (Zhao et al. 2006). A one-factor model with all items loading on a single construct did not provide a good fi t to the data suggesting that CMV does not pose a serious threat. In this study, the one-factor test yields a (χ2 = 1216.1, df = 135, P = 0.00). The one-factor fi t provided a worse fi t to the data compared to the proposed measurement model (difference in χ2 = 1029.4, df = 8, P = 0.00). Overall, the results indicate that the study measures are reliable and valid, and that the proposed model provides an adequate fi t to the data.
After establishing the adequacy of the model, it is appropriate to examine individual path coeffi cients. Table 3 summarizes the results of this analysis. It is worthwhile to highlight that in this study, the model incorporated two marketing relationship constructs (perceived equity and perceived value). This has allowed the authors to examine, test, and ultimately generate knowledge of the effects of these constructs on the formation of program member's loyalty. The effect of program perceived equity on program perceived value was signifi cant ( = 0.82, p < 0.001), thus, supporting the H1a hypothesis. As expected, program perceived equity had a strong positive and highly signifi cant impact on both program satisfaction ( = 0.32, p < 0.01) and program trust ( = 0.48, p < 0.001), supporting the H1b and H1c hypotheses.   In the near future, I will surely purchase from this retail store again 0.823 Program perceived value was found to be a signifi cant factor in determining program satisfaction ( = 0.29, P < 0.01) and program trust ( = 0.34, P < 0.001), supporting the hypotheses H2a and H2b. The hypotheses H3a and H3b were only partially supported. The hypothesis H3a posited that program satisfaction would be positively associated with program trust. As expected the path from program satisfaction to program trust was positive and signifi cant ( = 0.12, P < 0.05). However, the path from program satisfaction to store loyalty was not signifi cant, disproving the hypothesis H3b. The results indicated that program satisfaction and store loyalty are not related. These results suggest the mediating effect of program trust on the relationship between program satisfaction and store loyalty.

Discussion and implications
This study contributes to the current literature in several ways. First, an integrative model simultaneously analyzed the relationship between perceived equity, perceived value, satisfaction, and trust as well as investigated satisfaction and trust as the prediction of store loyalty. The fi ndings support the hypothesized path of perceived equity and perceived value. In essence, this fi ndings offer empirical evidence that when cardholders are treated equitably, they are likely to reciprocate by engaging in behaviors that enhance the fi rm-cardholder relationship. This is similar to the fi ndings in a study by Hutchinson et al. (2009), which suggested that customer perceptions of fairness were positively associated with customer perceptions of the value of the service received. This provides further support for the importance of rewards, discounts, and free products offered by the loyalty program. This also demonstrates the importance of ensuring fair and honest treatment of the program participants by the service providers through creating a win-win situation between company and customers.
The second important implication in this study is the strong positive relationship found between program equity and program satisfaction as well as program trust. Clearly, these fi ndings support the notion that loyalty program characterized by equity and fairness can consistently produce satisfaction and trust toward the loyalty program. This is consistent with the fi ndings of other empirical studies by Oliver and Swan (1989), and Ramaswami and Jagdip (2003). Therefore, cardholders who perceive that their inputs are greater than the outputs may feel that it is unfair and may be reluctant to form relationships with the programs or retailers. It is crucial to the provider who manages the loyalty program to maintain a constant utility level, to satisfy the members, and to increase their trust in a loyalty program. The outputs and the inputs need to be balance in order to increase consumer trust and ultimately program participation.
Third, the fi ndings unveiled that program perceived value signifi cantly infl uences program satisfaction and program trust. This confi rms the importance of using the measure of perceived value when predicting program satisfaction, as suggested by prior researchers (Grace, O'Cass 2005;Cronin et al. 2000), as well as trust (Moliner 2007;Singh, Sirdeshmukh 2000). Based on the results, it is important to note that program perceived equity exerted a much higher infl uence on program satisfaction and program trust compared to program perceived value. This implies that feeling equitable in their loyalty program increases cardholders' satisfaction and trust in the program. This calls for the use of perceived equity and fairness in the loyalty program, as recommended by prior researchers, since there is no published work examining the issue of fairness in a loyalty program (Lacey, Sneath 2006).
Fourth, program satisfaction had a positive infl uence on program trust. This fi nding is consistent with that of Singh and Sirdeshmukh (2000) who similarly identifi ed a direct effect of satisfaction on post-purchase evaluation of trust. The fi ndings also suggest that when loyalty program's performances exceed cardholders' initial expectations, customers' positive confi dence (trust) in the competence of the loyalty program is likely to increase.
Fifth, our study does not support the proposed relationship between program satisfaction and store loyalty. It is important to note that the effect of program satisfaction on store loyalty does not follow a direct path; instead, the program trust mediates this effect. This fi nding is unexpected because most previous studies (for example, Bloemer, Ruyter 1998;Macintosh, Lockshin 1997) demonstrate that customer satisfaction signifi cantly infl uences store loyalty. On the contrary, this result is parallel to that by Mo Koo (2003) and Omar et al. (2011) who stipulated that store satisfaction does not relate directly to store loyalty. This may be because, within the context of this study, we did not defi ne retail loyalty program as a product. Hence, we could summarize that in the retail loyalty program setting; "program satisfaction" is not an infl uential factor in determining the cardholder's loyalty to the store examined in this study. Therefore, it appears that program trust may mediate the effects of program satisfaction on store loyalty (program satisfaction -program trust -store loyalty).
Sixth, program trust seems to be an important predictor of store loyalty. This supports prior research in that customer trust in the product or service increases store loyalty (Flavián et al. 2006;Ball et al. 2006;Rauyruen, Miller 2007). Undoubtedly, this fi nding supports the notion that program trust is one of the important factors in promoting member loyalty to the store. Clearly, these results indicate that in order to achieve store loyalty, it is not enough for cardholders to feel satisfi ed with the loyalty program but to trust the program. Briefl y, these fi ndings imply that program providers should explore ways to increase members' trust towards the program. In fact, this paper suggests few approaches such as providing fair treatment and delivering value to program members.

Limitations and directions for future research
When discussing the results of this study, one must keep in mind the particular limitations of the study. First, we did not select cardholders from the Klang Valley area completely randomly; therefore, they may not accurately represent the population of loyalty program members. Thus, the ability to generalize beyond this sample is limited. A second pertinent weakness concerns the cross-sectional research design employed. Certainly, longitudinal research is required to capture fully the dynamic nature of customer post-consumption evaluation. Obviously, any efforts to utilize longitudinal design to test the present model would require an enormous amount of sustained consumer cooperation over time.
It may be fruitful for future research to replicate and validate all parts of the current model in order to determine the robustness of the fi ndings. Cross-national as well as cross-cultural studies are essential in order to examine the generalizability of the model. This research direction appears to be potentially fertile because many consider loyalty program an important strategy and mechanism for increasing revenue growth (Young, Stepanek 2003). In addition, further studies need to investigate insignifi cant relationship between program satisfaction and store loyalty. Is this result peculiar to this sample only or is it a widespread phenomenon? As suggested in the article, future research could perhaps study program trust as a mediating variable. The current study has contributed to the loyalty program literature, specifi cally to the understanding of the effects of equity, value, satisfaction, and trust on increasing cardholder loyalty, particularly in the retail sector in Malaysia. Future research could focus on comparing different segments (such as new and matured cardholders, single and multi loyalty program ownership) or different cultures or industries (airlines, hotel and fi nancial) on the relational outcomes of loyalty programs would also be an interesting area of study.

Conclusions
The spread of loyalty programs in business circles has prompted numerous studies in various aspects. This study that focused on analyzing the relationship among program perceived equity, program perceived value, program satisfaction, program trust and store loyalty is important for retailers. Through loyalty programs, fi rms can potentially gain more repeat business and customer loyalty. Previous literature on the impact of loyalty programs towards loyalty have focused on members and non-loyalty program members (Passingham 1998), and consumers' behavior before and after enrolling in loyalty program (Sharp, Sharp 1997). However, there are relatively few existing studies that empirically study fairness and value of loyalty programs (e.g., Lacey, Sneath 2006;Sunny Hu et al. 2010). Thus, the present study that examined the relationship between program perceived equity, perceived value, satisfaction, trust and store loyalty adds up to the empirical evidence on this matter.
This study has brought to light knowledge pertaining to the critical role of program perceived equity as one of the main predictor of program perceived value, program satisfaction and program trust. The fi ndings of this study support the notion that cardholders tend to be satisfi ed and valued when the loyalty program delivered with fair and equitable treatment to the cardholders. In addition, the cardholders tend to trust the program, which in turn creates loyalty among the members. This is in line with De Wulf et al. (2003), that customers behave rather opportunistically and in self-interest when deciding to participate in a loyalty program, thus cardholders will always want to minimize their inputs, and maximize the outcomes expected.
This study also provides important implications, particularly for retailers offering loyalty program to their customers. The fi ndings reveal that the retailers needs to be highly aware of the cardholders perception of fairness toward their loyalty program. Although one may view the relationships as long-term, it may be a sense of reciprocity that develops from the enactment of fairness may elicit a behavioral response on the part of a cardholder. By taking full advantage of these results, managers can use loyalty program to strengthen their marketing positions without compromising on their customers' perception of fairness towards their loyalty program. Moreover, this study also offers evidence to substantiate the importance of program perceived value in program satisfaction and program trust.
Another important observation that is critical for the retailers is that program satisfaction has no direct effect on store loyalty, but has an indirect infl uence through program trust. Clearly, these results indicate that in order to achieve store loyalty among program's cardholders, it is inadequate to have cardholders to feel satisfi ed with the program but, they need to fully trust the program. This is in line with other researchers like Macintosh and Lockshin (1997) and Sirdeshmukh et al. (2002) who also recognize the role of trust in creating loyalty. Therefore, from program-related factors it is vital for fi rms to ensure that customers build bonding with the loyalty program. As most loyalty programs face competition from rival programs that offer similar benefi ts and rewards, fi rms need to fi nd ways to build bonding with their cardholders in order to make them loyal to the store.