Examining customers’ experience with the Nigerian Bank Verification Number (BVN) policy from the perspective of a dual-lens theory

Purpose – Although customer experience has been widely researched, its effects on behavior towards a government policy is still unclear. This paper draws on two theories with some similar and opposing perspectives to investigate the effect of three components of customer experience (i.e., utilitarian experience, hedonic experience, and relational experience) and customers’ intention on behavior within the context of the BVN policy implemented by the Nigerian apex bank. Design/methodology/approach – Data emerged from one of the most populous districts in south-eastern Nigeria. Participants were recruited by mall-intercept. Out of the 283 participants approached, 246 participated but only 82.9% were valid for analysis. After subjecting data to statistical screening to confirm its suitability for parametric statistical analysis and examining data for the potential effects of common method variance as well as sample representativeness, a partial least squares structural equation modelling (PLS-SEM) technique and the Preacher and Hayes bootstrapping procedures were utilized to test the hypothesized relationships. Findings – Based on data obtained from Nigerian bank customers, the paper demonstrated that the customer loyalty arising from the implementation of a government policy is determined more by hedonic experience, followed by relational experience and very much less by utilitarian experience. Findings also indicate that the relationship between the components of customer experience and customer loyalty is complementarily mediated by intention to open new account(s). Research limitations/implications – Though the theoretical grounding of our paper strongly support the study design, we strongly recommend that future research should examine customer experience-customer behavior models in situations of policy implementation with longitudinal design. Additionally, since intention to open new account(s) is a complementary mediator of the links between the components of customer experience and customer loyalty, there is need for future researchers to integrate other mediators into the conceptual framework that we examined in this paper. Practical implications – This paper cautions that whilst the research findings play out effectively in situations where the benefits of the introduced policy and the consumers’ belief in the good intent of the policy are congruent and customers are susceptible to the manipulations of the social class leading to absence of volitional control; firms should not be deceived into relying too heavily on this kind of loyalty because it is situational and consequently promiscuous. Nevertheless, deploying more resources to seamlessly meet the needs of customers in such situations is counter-productive for service organizations. Social implications – Based on our findings, it has come to the fore that consumers will be at the receiving end of a government policy poorly implemented by service organizations. When such policies are rolled out therefore, governments should enforce operational modalities that will forestall potential negative experiences that consumers could possibly encounter. Originality/value – By examining the effects of three components of customer experience and intention to open new account(s) on customer loyalty within the context of BVN implementation in an emerging banking sector, we contribute to the broad stream of literature that focuses on the effect of customer experience on company bottom-line. The strength of this contribution is based on the premise that this paper draw on the similarities and opposing orientations of two theories to uncover these effects. We show that the effects of the three components of customer experience on customer loyalty is different from the results of previous research because of the unique perspective adopted in this paper.

Purpose -Although customer experience has been widely researched, its effects on behavior towards a government policy is still unclear.This paper draws on two theories with some similar and opposing perspectives to investigate the effect of three components of customer experience (i.e., utilitarian experience, hedonic experience, and relational experience) and customers' intention on behavior within the context of the BVN policy implemented by the Nigerian apex bank.
Design/methodology/approach -Data emerged from one of the most populous districts in south-eastern Nigeria.Participants were recruited by mall-intercept.Out of the 283 participants approached, 246 participated but only 82.9% were valid for analysis.After subjecting data to statistical screening to confirm its suitability for parametric statistical analysis and examining data for the potential effects of common method variance as well as sample representativeness, a partial least squares structural equation modelling (PLS-SEM) technique and the Preacher and Hayes bootstrapping procedures were utilized to test the hypothesized relationships.
Findings -Based on data obtained from Nigerian bank customers, the paper demonstrated that the customer loyalty arising from the implementation of a government policy is determined more by hedonic experience, followed by relational experience and very much less by utilitarian experience.Findings also indicate that the relationship between the components of customer experience and customer loyalty is complementarily mediated by intention to open new account(s).
Research limitations/implications -Though the theoretical grounding of our paper strongly support the study design, we strongly recommend that future research should examine customer experience-customer behavior models in situations of policy implementation with longitudinal design.Additionally, since intention to open new account(s) is a complementary mediator of the links between the components of customer experience and customer loyalty, there is need for future researchers to integrate other mediators into the conceptual framework that we examined in this paper.
Practical implications -This paper cautions that whilst the research findings play out effectively in situations where the benefits of the introduced policy and the consumers' belief in the good intent of the policy are congruent and customers are susceptible to the manipulations of the social class leading to absence of volitional control; firms should not be deceived into relying too heavily on this kind of loyalty because it is situational and consequently promiscuous.Nevertheless, deploying more resources to seamlessly meet the needs of customers in such situations is counter-productive for service organizations.
Social implications -Based on our findings, it has come to the fore that consumers will be at the receiving end of a government policy poorly implemented by service organizations.When such policies are rolled out therefore, governments should enforce operational modalities that will forestall potential negative experiences that consumers could possibly encounter.
Originality/value -By examining the effects of three components of customer experience and intention to open new account(s) on customer loyalty within the context of BVN implementation in an emerging banking sector, we contribute to the broad stream of literature that focuses on the effect of customer experience on company bottom-line.The strength of this contribution is based on the premise that this paper draw on the similarities and opposing orientations of two theories to uncover these effects.We show that the effects of the three components of customer experience on customer loyalty is different from the results of previous research because of the unique perspective adopted in this paper.Accenture in 2013 indicates that over 50% of consumers leave firms due to poor experiences (Wollan, 2014).Thus, negative experiences can activate consumer exodus and attenuate customer loyalty to a service organization.Yet, theory suggests that the rules that govern behavior may be externally imposed by "agents who have our interests in mind" (Prelec and Herrnstein, 1991, p. 321 in Soman and Cheema, 2011).Pettinicchio (2017) quoted Hillary Clinton who is said to hold the following view: "I don't believe you change hearts.I believe you change laws, you change allocation of resources, you change the way systems operate" (p.156).Since governments and its agencies have the power to enact laws whilst the latitude of citizens to contest the laws diminishes if majority of the masses perceive the policy as largely in line with theirs' and the entire society's interest, the views that consumers hold about services organizations and their services as well as their loyalty propensity may likely remain unchanged if the service organization is simply implementing a directive from the government or its agency.Thus, intention and behavior towards business organizations is to some extent non-volitional, implying that consumers affected by externally imposed rules must comply.The implicit proposal of the foregoing is that consumer intention and behavior is contingent on whether the service experiences provided by an organization is linked to a government directive or not.Thus, there is no single way of demonstrating why consumers remain loyal to a service organization.
Recently, the Central Bank of Nigeria (CBN) introduced the Bank Verification Number (BVN) policy with bank customers having to go through onerous experiences to comply.Building on the contrasting theoretical perspectives presented above, we pose the following queries: i) why will a customer find it desirable to go through these experiences just to obtain this so-called unique BVN? ii) Why will customers still remain loyal to their banks irrespective of the onerous experiences they passed through in obtaining a BVN? and iii) In the face of these upturns, what is the role of intention to open new accounts?
Addressing these queries is important in two key directions.First, although the field of customer experience has attracted significant academic discussions (see for instance Adhikari, 2015;Rose et al., 2012;Gentile et al., 2007), literature has till date, failed to contextualize the roles of customer experience within the context of unsuspecting government policy.This is surprising especially as Verhoef et al. (2009) argued that experience subsist at prior, during, and post-purchase phases of the consumer buying cycle.This article focuses on the 'during' stage of customer experience as portrayed by Verhoef et al. (2009).The 'during' stage of customer experience reflects the implementation stage of the BVN policy when commercial banks were directed by the apex bank to deliver this service to their customers.We focused on the implementation stage of the BVN because as noted by Klaus (2013), utilitarian, hedonic, and relational attributes of customer experience play out effectively in the 'during' stage of the consumer buying process.This paper draws on the BVN policy implemented by the Nigerian apex bank to demonstrate how customer experience will affect consumer behavior.Second, by addressing these questions, this paper advances the customer experience literature through drawing on two theories to investigate the effect of three components of customer experience (i.e.utilitarian experience, hedonic experience, and relational experience) and customers' intention on behavior within the context of the BVN policy.Through examining the effects of these three components of customer experience on customer loyalty as well as the mediating role of intention to open new account(s) in this link within the context of BVN implementation, we contribute to the broad stream of literature that focuses on the effect of customer experience on company bottom-line by demonstrating why consumers may remain loyal despite poor experiences.The strength of this contribution is based on the premise that this paper draw on the similarities and opposing orientations of two theories to uncover these effects.The rest of the paper is organized as follows: First, we explore the theoretical background and develop the research hypotheses.
The methodology is thereafter discussed.We subsequently analyzed data and discuss the findings as well as the research contributions.Finally, the paper draws logical conclusions and points out some limitations and future research directions.

Theoretical background and conceptual framework
The Bank Verification Number (BVN) policy BVN was initially introduced by the Central Bank of Nigeria (CBN) on the 14th of February 2014 (Idowu, 2015;Aroloye, 2015) with registration due for completion on or before 30th of June 2015 (Premium Times, 2014).As this deadline was not met, CBN had to extend registration deadline to 31st October 2015 (Idowu, 2015).BVN reflects the CBN's initiative to give every Nigerian bank customer a unique identification number that is verifiable across all banks in the country (Aroloye, 2015).The registration process involves customers walking into their banks and having their biometrics (which include facial image and 10-finger prints) captured (Idowu, 2015).Unique identity numbers are thereafter created.Irrespective of the number of accounts and banks operated by a customer, one BVN registration can serve for all, meaning that customers will have to register only once.BVN is part of the CBN's cashless policy aimed at capturing customers' data and checking fraudulent practices in the Nigerian banking system (Premium Times, 2014).Aroloye (2015) similarly argued that the central intent of the BVN policy is to safeguard bank customers, mitigate fraud and strengthen the Nigerian banking system.
According to Idowu (2015), the customers' data captured through the BVN serve as online and mobile banking gateways while at the same time helping in the Know Your Customer device.With BVN, the credit history of customers can be easily ascertained, thereby making acquiring of loans by customers easier while also enabling banks to easily track fraudulent customers (Aroloye, 2015).The policy is also meant to reduce queues in the banking halls (Leadership Editors, 2014).All these mean that the intent of the BVN were all aimed at improving the service experiences of banking customers in Nigeria but this was never going to come easy.
At the first roll out of the policy, the apex bank insisted on 40 per cent compliance in the first year that the policy was introduced (Leadership Editors, 2014).The CBN's initial position was that bank customers who fail to undergo the BVN registration exercise will be unable to perform transactions on their accounts.After the rush by customers created several frustrating queues at various banking halls across the country, the CBN had no option than to extend the deadline for registration from 30th June to 31st Evidences suggest that the frustrating experiences engendered by the BVN policy amongst others include, the difficulty in linking BVN to other accounts (Udo, 2016), the length of time taken to provide enrolled customers with BVN as well as fraudsters seizing the opportunity to send fake messages to bank customers (Komolafe, 2015).For instance, since we started developing this article till date, I have received more than ten of such fraudulent text messages.Additionally, the approval granted by the apex bank to commercial banks to allow customers who are finding it difficult to link their BVNs to different accounts to effect needful corrections (Udo, 2016) foretells that customer experiences were not seamless but rather torturous.While such torturous experiences may on the surface, suggest possible negative consequences for Nigerian banks and imply the deployment of resources by banks to mitigate it for the sake of survival, a contrasting theoretical viewpoint indicates that this may not be the case especially as the bank is simply implementing the directive of a government agency.
Given the depth of customer experience research, definitions of the concept abound.Customer experience is variously defined as summation of all the clues that customers detect in the buying process (Meyer and Schwager, 2007;Berry et al., 2002), activities spanning prior, during and post-purchases (Verhoef et al., 2009), and engaging customers in a personal way (Pine and Gilmore, 1999).Gentile et al. (2007, p. 397) also noted that customer experience "originates from a set of interactions between a customer and a product, a company, or part of its organization, which provoke a reaction.This experience is strictly personal and implies the customer's involvement at different levels (rational, emotional, sensorial, physical, and spiritual)".Gentile et al.'s definition is remarkable in the sense that it not only emphasizes how the concept originates but also points out its multifaceted nature.Drawing on the above, Verhoef et al. (2009, p. 32) more elaborately submitted that "customer experience construct is holistic in nature and involves the customer's cognitive, affective, emotional, social and physical responses to the retailer.This experience is created not only by those elements which the retailer can control (e.g., service interface, retail atmosphere, assortment, price), but also by elements that are outside of the retailer's control (e.g., influence of others, purpose of shopping)".Thus, in addition to pointing out the multifaceted nature of customer experience, Verhoef et al.'s (2009) definition reflect its holistic character and the influence of factors beyond the marketers' control.This definition is therefore suitable for our purpose in this paper particularly because its emphasis on external forces is strongly in line with policy introductions by government or its agencies since consumers and firms adjust to what they cannot control.
Just as it is nearly impossible to find a dominant definition of customer experience, its components are also diverse.But whilst a wide range of perspectives abound, there tends to be an agreed structure of the components of customer experience.Verhoef et al's (2009) conceptual portrayal of the components of customer experience as a construct comprising the cognitive, affective, emotional, social, and physical elements.Broadly conceptualized, customer experience is also noted as comprising cognitive and affective elements (Rose et al., 2012) or utilitarian and hedonic components (Babin et al., 1994).Although other components of customer experience abound in the literature (see for instance Klaus, 2013;Pine and Gilmore, 1999), Schmitt's (1999) five-dimensional framework comprising sensorial, emotional, cognitive, behavioral, and relational experience is by far the most expanded framework of customer experience components.What can be concluded from the foregoing however is that as noted before, customer experience is a multifaceted construct comprising several components.These components are holistically perceived by customers.It has been argued that not all of these components are important in every context or situation (Gentile et al., 2007) because some components of customer experience are more influential than others in some contexts (Åkesson et al., 2014;Schmitt and Zarantonello, 2013).These context-specific nature of customer experience imply that scholars should focus on specific components of customer experience that suits the study context.To this end, previous operationalizations of customer experience within the banking sector (see Garg et al., 2014;Klaus et al., 2013) broadly reflect the cognitive/utilitarian, emotional/hedonic, and relational components.Thus, this paper focuses on utilitarian, hedonic, and relational components of customer experience.While hedonic experience as was noted before aligns with shopping entertainment, shopping enjoyment or what is lately referred to as emotional experience, utilitarian experience is cognitive in nature and aligns with task instrumentality (Holbrook and Hirschman, 1982) which Babin et al. (1994) and Overby and Lee (2006) described as utilitarian value.
The focus of this study on hedonic, utilitarian and relational components of customer experience is also strictly in consonance with the tenets of the theory of reasoned action because as noted by Schiffman et al. (2012), it builds on the tricomponent attitude model which is compatible with the above components of the construct.

Power politics theory vs the theory of reasoned action (TRA)
The power politics theory, also referred to as political elites or power elites theory is a sociological theory that posit that the power to influence policy is concentrated in the hands of a few (Stachowiak, 2013).
Elites refer to a selected and small group of people/citizens and/or organizations that holds an enormous amount of power (Vergara, 2013).These selected few powerful individuals are at the top of every society and are usually politicians and business tycoons.Accordingly, Higley (2006) stressed that elites are the major determinants of decision making in modern societies because they command key business enterprises, important pressure groups, mass movements and political parties and so on and they are therefore key regular and substantial influencers of political outcomes.A key lesson to take away from Higley's reasoning is that as members of larger groups within a broader social system, elites influence and can also be influenced by the views and interests of group members.Thus, a significant contribution of the power elites theory is that the social class in which one belongs shapes his or her personal identity (Vergara, 2013).Pareto's (1991) description of elites indicate that the rest of the society are susceptible to the manipulations of the ruling class and by implications their policies.Thus, there is little or nothing that the rest of the society can do about the policy initiated by political elites rather than shape their intention and behavior towards those policies.The latitude to contest the policy further diminishes or even amounts to zero if majority of the masses perceive the policy as largely in line with theirs' and the entire society's interest.Accordingly, John and Street (2013) noted that individual choice interacts with institutions; and ideas become important when they interplay with individual interests.
The theory of reasoned action (TRA) (Fishbein and Azjen, 1975) on the other hand posits that customers' attitude towards an episode which in turn leads to behavior is a function of customers' beliefs or rational cognitive assessment of the episode.The TRA not only predicts intention and behavior very well, "it also provides a relatively simple basis for identifying where and how to target consumers' behavioral change attempts" (Sheppard et al., 1988, p. 325).Although postulated in the 1980's, the continual relevance of the TRA is reflected in its recent applications.For instance, Hassandoust et al. (2011) applied the TRA to uncover the behavioural factors that influence virtual knowledge sharing in Malaysia.Lujja et al. (2016) also utilized the TRA to investigate the feasibility and adoption of Islamic banking in Uganda.Within the health services sector of Uganda also, Mafabi et al. (2017) applied this theory.The parsimonious character of the TRA is also an influencing factor of its wide application (Lujja et al., 2016) whilst TRA has also been touted as "the best-known social-psychological attitude-behaviour model which incorporates external factors (normative beliefs) on intention to adopt an overt behaviour (Prager, 2012)" (Lujja et al., 2016, p. 586).
The intersecting point of the power politics theory and the TRA as applied in this paper is that irrespective of the perspective adopted, intention and behavior are influenced.We illustrate this with the concept of earmarking investigated by Soman and Cheema (2011).Soman and Cheema (2011) theorize imposing spending rules to counteract impulsive behavior as one of the ways in which earmarking -labelling money for a particular purpose -can be guided to fruition.The point of distinction is that earmarking as described by Soman and Cheema (2011) is volitional especially when individually-induced and its ideation towards self-control.But earmarking practice and a governmental policy may have similar effects because Soman and Cheema (2011) argue that breaking rules set by one leads to negative emotions such as feelings of failure or lack of faith in oneself.Prelec andHerrnstein (1991, p. 321 in Soman andCheema, 2011)  them".Thus, although compliance to rules set by individuals for themselves is to a great extent within the confines of volitional behavior, consequences of violation may propel it to assume the status of a rule enforced by a government or one of its agencies.In attempt to avoid these negative emotions, consumers are constrained to obey rules they set for themselves even though compliance is to a very less extent, a product of compulsion.
In contrast, consumers comply with government policies because in most instances, it is a matter of compulsion.The competing point of the power politics theory and the TRA is the nature of influence because behavioral outcomes in each perspective are driven by divergent motivations.Given that opportunities that lead to breaking of rules evolved by individuals for themselves abound and can be simple in some instances (Soman and Cheema, 2011), we argue that its effect might not be as acute as a policy initiated by a government probably because each behavioral disposition stem from different motivations.While Fishbein and Azjen's (1975) model was initially proposed to predict and explain volitional behavior, the model has also been frequently applied to situations in which the target behavior is not completely under the subjects' volitional control (Sheppard et al., 1988, p. 325).As such, it can form a good theoretical point of departure for understanding and explaining behavioral responses during the implementation phase of a policy.But given that the studies that applied the TRA to non-volitional situations have been criticized on the grounds that the model is suitable mainly for predicting and explaining volitional behavior (Sheppard et al., 1988), we perceive the theory and its assumptions as the most suited for supporting the power politics theory.Since the TRA assesses how belief factors affect intentions which in turn affect behavior, its adaptation to the BVN policy is quiet straightforward.Schiffman et al. (2012) is of the view that the TRA aligns closely with the tricomponent attitude model because it incorporates cognitive, emotional, and conative components.The TRA submits that a full understanding of intention can be gained if subjective norm is explored in addition to the cognitive and emotional drivers of intention (Schiffman et al., 2012).The three components of experience are treated as the belief factors whereas intention to open new bank account(s) and customer loyalty are intention and behavior variables respectively.Subjective norm measures the influence which relevant "others" have on an individual customer.Thus, it corresponds to relational experience because subjective norm is similar to what Grove and Fisk (1997) termed 'other customer influences' whilst numerous studies likened relational experience to social value, customer-to-customer interactions, and social presence (see Mohd-Any et al., 2015;Blasco-Arcas et al., 2014;Cyr et al., 2007).These terms reflect the influence of other customers on an individual customer's service experience.Utilitarian experience corresponds to cognitive component of belief while hedonic experience corresponds to the emotional component of belief.
When the power politics theory is applied to the BVN policy in Nigeria, the policy has till data remained uncontested as a result of two key reasons: a) it was initiated by a powerful government agency, and b) it is for the benefit of the entire society.Since social classes constantly interact, whilst attitudes and behaviors are defined by the ruling class (Vergara, 2013), a case can therefore be made that as citizens experience a policy initiated by the ruling class, their intentions and behaviors are shaped accordingly with no need for protest especially if the policy is for the socio-economic good of the people.This is absolutely in line with the reasoning that services experiences are created by elements that firms can control and elements that lies outside firms' control (Verhoef et al., 2009).However, our interest in this paper is not to dig deep into what makes citizens to comply in the strict sense of it.Rather, we are much more interested in investigating how citizens' (in our case, bank customers) experiences with the implementation of a policy (BVN in our case) shape their subsequent intention and behavior.Thus, by isolating and testing the effects of the three customer experience components (utilitarian experience, hedonic experience, and relational experience) on behavior (i.e.customer loyalty) and through intention (i.e.intention to open new account(s)) in the context of BVN policy implementation, we tell the contrasting story of why loyalty does not ensue only from positive experiences but can also remain unchanged if customers encounter negative experiences as a result of government policy implementation by service organizations.

Hypothesis development
Since the TRA is mostly suited for predicting and explaining volitional behavior, the nature of relationships between belief factors and intention and behavior is bound to be different if people are performing a behavior because they have to do so as implied in the power politics theory.According to Fishbein (2008), the relative importance of the three belief factors that determine intention will vary as a function of behavior and the population being considered.Fishbein (2008) further noted that some of the three broad belief factors may fail to carry any significant weight on intention.Accordingly, Åkesson et al. (2014) noted that certain drivers of experience are dominant and are more important than others in some situations.Although previous research (e.g.Adhikari, 2015, Shang et al., 2005) found positive association between the three components of experience and intention, it has been widely noted that hedonic experience is a better predictor of intention and behavior than utilitarian experience (Adhikari, 2015;Ding and Tseng, 2015;Shang et al., 2005;Zaltman, 2003) even though a contrasting evidence abound (see Mohd-Any et al., 2015).In line with the prospect theory also, decision making is much more likely to be a product of heuristics than rational judgement when consumers are faced with uncertain situations (Tversky and Kahnman, 1974) whilst consumers are more likely to react to possible losses than possible gains (Chen et al., 2009).Thus, the negative consequences of failing to comply with the BVN exercise and fear of the unknown will make emotional experience a better predictor of customer loyalty than utilitarian experience.The following hypotheses is therefore proposed: H1. Utilitarian experience will be a significant positive predictor of customer loyalty H2.Hedonic experience will be a better significant positive predictor of customer loyalty than utilitarian experience Additionally, the importance of interactions between institutions including powerful ones and individual interests has been stressed (John and Street, 2013;Vergara, 2013).This viewpoint is also reinforced in the service dominant logic of marketing that posit that consumers co-create value when they interact with firms' services (Vargo and Lusch, 2004).The experiential consumption literature has also recognized the effect of customer-to-customer interactions on intentions (see Verhoef et al., 2009) and behavior (see Blasco-Arcas et al., 2014;Shang et al., 2005).Although a contrasting evidence on the importance of relational experience abound in the literature (see Mohd-Any et al., 2015;Gentile et al., 2007), given that greater coherence is fostered especially when the people affected by a policy similarly perceive the policy as they interact with each other, we expect that the influence of relational experience on customer loyalty will even be higher than the effect of utilitarian or hedonic experience because the consequences of behaving otherwise and fear of the unknown will be escalated through customer-to-customer interactions about the policy.This progression is highly logical because consumers were first solely thought of as rational decision makers and later as emotional beings capable of processing feelings, fun and fantasies (Holbrook and Hirschman, 1982); and more recently as active co-creators of value (Vargo and Lusch, 2004).We draw on the foregoing to propose the following hypothesis: H3. Relational experience is very likely to be a better predictor of customer loyalty than hedonic experience and utilitarian experience.
The theory of reason action (TRA) posit that intention is a product of the favorability or unfavorability of the intended behavior and subjective norm (i.e. the perceived opinion of 'important others' regarding the behavior in question) (Fishbein and Ajzen, 1975).Shimp and Kavas (1984) also alluded to the fact that consumer intention is a product of perception of whether 'important others' will view the effort expended as worthwhile.According to Fishbein and Ajzen (1975), intention will predict behavior of any voluntary act.Thus, when a behavior is compulsive, the role of intention might change or remain unchanged.The role/effect of behavioral intention as discussed above is therefore limited to the context of volitional behavior.Extending its role to the context of compulsive behavior might seem interesting for two reasons.
First, in Sheppard et al.'s (1998) critique of the TRA, they argued that the influence of intention should not be limited to volitional behaviors.Second, since the power politics theory implicitly suggest that intentions are to some extent non-volitional, implying that citizens affected by a policy must comply, the role of intention in this circumstance may differ or remain the same when compared to situations of volitional control.We therefore submit that the role of intention to open new account(s) ought to be examined.Since theories (see Davis et al., 1989;Ajzen, 1991;Fishbein and Ajzen, 1975)

Operationalization of the research constructs
Except intention to open new account(s) which had no existing measures, the measures of all constructs were adapted from previous research.However, the measures of intention as used in this paper is very similar to that adapted by Rose et al. (2012).Additionally, it was previously argued that both utilitarian and hedonic dimensions of value are experiential because the former reflect cognitive aspects of experience while the later reflect emotional aspects of experience such as entertainment for instance (Overby and Lee, 2006).Previous research (see for instance Lemke et al., 2011) also studied customer experience from the viewpoint of value-in-use.Thus, value and experience are strictly interwoven whilst value scales can serve as surrogates for experience scales.For instance, Nambisan and Watt (2011) measured customer experience with measures adapted from value scales.In line with the foregoing therefore, the five and six scale items that respectively measured utilitarian experience and hedonic experience were adapted from Overby and Lee (2006).Measures of relational experience (four scale items) were adapted from Hsu and Tsou (2011) while the five scale items used to measure customer loyalty was adapted from Zeithaml et al.'s (1996)  in the appendix.They were measured in 5-point Likert format with anchors of "strongly agree" representing 5, "strongly disagree" representing 1, and an option to respond neutrally representing 3.

Sample and data collection
The data emerged from the general area of Aba, Umuahia urban, one of the most populous districts in south-eastern Nigeria.Respondents' marital status was relatively even as 48% were single while 52% were married.Majority of the respondents are sufficiently educated since over 88% had a minimum of HND/BSc qualification, while only 11.8% had lower qualifications.Regarding income, 91.7% are low and middle income earners while 8.3% are high income earners.This is clearly in line with the demographic composition of the Nigerian population arising from the uneven nature of wealth distribution.

Data screening, sample representativeness and common method variance
Prior to the main analysis, data was assessed for the appropriateness of parametric statistical tests by examining the parametric assumptions of multicollinearity and normality.The multicollinearity diagnostics were employed to ensure that the predictors are not redundant while the normality tests were aimed at detecting and resolving extremely skewed data.variables respectively ranged from -0.133 to -2.041 and 0.011 to 5.173.Given the above findings, we conclude that the data is normally distributed because the range of skewness and kurtosis values were respectively below 2 and 7 recommended in the literature (see Curran et al., 1996).With the above results, we conclude that data is suitable for parametric statistical tests.
To assess the representativeness of our sample, nonresponse bias was evaluated using the time trends extrapolation method (Armstrong and Overton 1977).Since respondents were conveniently and purposively drawn, a distinction was made between two kinds of respondents to our survey (i.e. respondents who completed the survey immediately and respondents who completed and returned the survey at a later time or date).This yielded 119 immediate responders and 86 later responders producing two groups.An independent-samples t-test on the key constructs' aggregate scores revealed that nonresponse bias was not a major source of concern because apart from hedonic experience that indicated statistically significant difference (t = 2.67, df= 202, p< 0.05), no such difference was found in respect of the four key remaining constructs.Thus, the sample employed is relatively representative of the population.
Since responses on the predictor variables and the endogenous constructs were simultaneously retrieved by a means of self-report, the data may be prone to a common method variance (Podsakoff et al., 2003).
Thus, three main ways that broadly fit into the design of the study and statistical controls as recommended by Podsakoff et al. (2003) were employed to examine common method bias.First, by ensuring that scale items are not complexly or ambiguously worded, common method bias can be reduced (Podsakoff et al., 2003).As such, in addition to a review of the scale questions by a professor of marketing, a handful of respondents reviewed the survey instrument for clarity or vagueness of the scale items.The observations that emerged from these reviews were built into the final survey questions to ensure clarity.Second, in terms of questionnaire design, scale items were mixed, reverse-coded items were used, and respondents were reminded and assured that there are no right or wrong answers.Due to the inherent disadvantages of these procedural remedies (see Podsakoff et al., 2003), statistical remedies were also applied.To further confirm whether common method bias posed any serious threat to the analysis, the Harman's one-factor test recommended by Podsakoff et al. (2003)  that common method variance is not a serious concern since the variance of the single factor solution is less than 50%.For the sake of presentation parsimony, the entire outputs from these tests were not reported.However, the data that produced these summarized results can be made available on request.
Generally, we conclude that common method bias does not hinder the interpretation of our results.

Validity and reliability assessment
To examine the psychometric properties of the measurement scales, the validity and reliability status of the dataset were ascertained.While reliability is assessed through Cronbach's alpha, which is believed to be a conservative measure; or composite reliability, which is believed to be less conservative (Hair et al., 2014;Bagozzi and Yi, 1988), construct validity is established when a measurement scale demonstrates both convergent and discriminant validity.Since Cronbach alpha of 0.60 or over and composite reliability of ≥ 0.70 represent minimum thresholds of acceptability (Hair et al., 2014;Bagozzi and Yi, 1988), we conclude that the reflective indicates of the measurement constructs are internally consistent because outputs indicate that Cronbach alpha ranged from 0.60 to 0.85 while composite reliability is in the range of 0.79-0.89.Additionally, the average variance extracted (AVE) for all constructs ranged from 0.47 to 0.63, a clear demonstration of convergent validity of all measurements (see Table 2).The measurements also display discriminant validity, since the square root of the AVE of each construct is higher than the highest correlation pair of the constructs (Fornell and Larcker 1981) (see Table 2).Additionally, although not reported in bid to ensure that findings are parsimoniously presented, the cross loadings indicate that the loadings of all the items except one item measuring hedonic experience on their primary factors were 0.20 away from their loadings on the other factors.The problematic item was deleted and excluded from subsequent analysis for constituting discriminant validity threat.Thus, the outputs reported in Table 2 were obtained after deleting the problematic item.

Test of hypotheses
The proposed research hypotheses were examined in two phases.First, H1, H2, H3 and H4 were examined through a partial least squares structural equation modelling (PLS-SEM) technique through the SmartPLS 2.0 software (Ringle et al., 2005).The PLS-SEM technique was preferred over the covariancebased SEM approach because of its robustness and efficiency with small and medium samples (Segarra-Moliner et al., 2013) as well as the technique's ability to make predictions (Blunch, 2008).Second, to test H5a-c, the model was examined for possible mediation effects by adopting the procedures laid out by  -----Insert Table 3 about here----- To test H1 to H4 as noted above, the structural model of the PLS-SEM outputs was examined.Table 3 reports the PLS-SEM outputs.As shown in Table 3, utilitarian experience has a positive but insignificant linear effect on customer loyalty (β = 0.12; t = 1.46; p > 0.05).Additionally, hedonic experience (β = 0.36; t = 5.04; p < 0.05), relational experience (β = 0.29; t = 4.15; p < 0.05), and intention to open new account(s) (β = 0.15; t = 2.39; p < 0.05) all have significant positive linear effect on customer loyalty.
Thus, only H2 and H4 were fully supported whereas H3 was partially supported because relational experience was a better predictor of customer loyalty than utilitarian experience but hedonic experience predicted customer loyalty better than relational experience.H1 was not supported.An R 2 of 0.44 (see To ensure that Type I and Type II errors are minimized, the statistical power of the above tests were ascertained.The output (see Table 3) also indicates that the conducted tests were sufficiently powered.
----Insert Table 4 about   positive, it is a complementary mediation.In sum, the above results show that we have sufficient evidence to confirm that H5a-c are all supported.

Discussion and contributions
This paper initially set out to investigate the predictive capacity of customer experience on customer behavior towards commercial enterprises during a policy implementation initiated by government or its agency.The paper draws on two theories with some similar and opposing orientations as well as an important progression in the marketing literature to propose hypotheses that advance the outcomes perspective of customer experience and subsequently examined these hypotheses with a partial least squares structural equation modelling (PLS-SEM) technique and Preacher and Hayes bootstrapping scripting procedure.We found that hedonic experience and relational experience are significant predictors of customer loyalty while utilitarian experience is not.However, hedonic experience was a stronger direct predictor of customer loyalty than both relational experience and utilitarian experience.Results also show that the effects of utilitarian experience, hedonic experience, and relational experience on customer loyalty are significantly mediated by intention to open new account(s).We also found that intention to open new account(s) is a poor predictor of customer loyalty.The above findings are consistent with the reasoning that certain components of customer experience exhibit context-specific peculiarities such that they are more important than others in some situations (Åkesson et al., 2014) as well as previous researches that established the superiority of hedonic experience (see Adhikari, 2015;Ding and Tseng, 2015;Zaltman, 2003).Our findings also significantly add to the body of literatures (see for instance Mafabi et al., 2017) that postulated and confirmed the mediating role of intention.
The explanation of the above results essentially lies in the power politics theory which holds that attitudes and behavior are defined by the ruling class (Vergara, 2013) whilst majority of the society are susceptible to the manipulations of the societal elites (Pareto, 1991).This assumption contravenes the volitional nature of behavior as implied in the TRA.Since behavior is more compulsive than volitional during governmental policy implementation, consumers are more likely to rely on the emotional appeal associated with engaging in the behavior and interactions with co-consumers than engaging in rational evaluation of the policy in question.Thus, behavior is reduced to a case of, "let me comply because every other person is complying; there is nothing I can do about it because it is a government directive; let me just ascertain from others what they are doing and simply follow suite; and moreover, the entire thing is fun", rather than behaving based mainly on well-thought out expected rational benefits.These evidences also account for why intention to open new account(s) poorly predicted customer loyalty in our model.
The explanation of our results is also tangentially linked to some classic thinking in the mainstream loyalty literature.For instance, Oliver's (1999)  what is more recently referred to as 'consumption community' demonstrated how a knitting between a constituency representative and the constituency member's congruence shape loyalty even though this postulation aligns more closely with the volitional behavioral perspective.Oliver (1999) further stressed that the effect of cognitive constructs (e.g.satisfaction) is situational as loyalty can set in through other mechanisms (e.g.social bonding at the institutional and personal level).Additionally, the explanation of the findings can also be found in culture-bound evidences.For instance, Chung and Shin (2010) found that societies that are highly sociable, and interdependent are more susceptible to the influence of reference groups on individual behavior.With the high collectivistic ranking of Nigeria (Izogo, 2016), this point further highlights the credibility of our findings.Drawing also on the TPB (Ajzen, 1991), individuals are more likely to engage in a behavior if they believe that significant others want them to engage in such behavior (Downs and Hausenblas, 2005).
Based on our findings, we make three important contributions.First, by demonstrating that customer experience positively influences customer loyalty if a service organization is merely implementing the policy of a government agency irrespective of whether the implementation of the policy induces negative experiences, we challenge the reasoning that negative experiences induce customer switching because even though the operational modalities for the implementation of the BVN policy was not seamless, it never threatened customers' loyalty to their banks.Specifically, hedonic experience emerged as the most significant predictor of customer loyalty.This paper therefore contributes to the services science literature by demonstrating how the components of customer experience affect the nature of customers' intention and behavior towards service organizations during the implementation of a policy initiated by a 'powerful' regulatory agency of the government.Previous studies contextualized within a normal competitive business setup (e.g.Mohd-Any et al., 2015) found that e-value is mainly driven by utilitarian experience, less by emotional experience whilst relational experience do not significantly contribute to evalue formation.Similarly, Gentile et al. (2007) found that customers perceived emotional experience almost as much as the utilitarian experience but the values associated with the relational component of customer experience were low and did not reveal any sensible variation across the 12 product classes tested.In contrast, our paper demonstrates that under times of government policy implementation, hedonic experience is a better determinant of customer loyalty than relational experience and utilitarian experience.We also show that relational experience was a better predictor of customer loyalty than utilitarian experience.These findings therefore highlight the importance of engaging in an activity for its own sake as well as 'influential others' in an individual customer's experience.
Secondly, we also advance the services science literature by showing that intention to open new account(s) is a complementary mediator of the relationship between the three components of customer experience (i.e.utilitarian experience, hedonic experience, and relational experience) and customer loyalty.These effects are unique to the extent that they were demonstrated in a situation of compulsive behavior but previous studies (e.g.Mafabi et al., 2017;Lujja et al., 2016;Hassandoust et al., 2011) explored volitional behavioral situations or contexts.Thus, by drawing on the tenets of the power politics theory, we have deepened the current understanding of how customer loyalty set in beyond the traditional views that negative experiences induce customer switching.The perspective that we propose in this article accommodate the reasoning that the loyalty that consumers exhibit toward a service organization cannot be threatened by negative experience arising from the implementation of a policy initiated by a government or one of its regulatory agencies.
Finally, by demonstrating that intention become a poor predictor of behavior when the motivator of behavior is compulsive or nonvolitional, we also advance and apply the TRA to situations characterized by the absence of volitional control.Although the theory of planned behavior (TPB) (Ajzen, 1991), the successor of TRA extended TRA to include nonvolitional behavior by integrating perceived behavioral control, the TPB fails to account for how behavior is shaped under situations of compulsion.We show that contrary to the reasoning that intention will predict behavior of any voluntary act (Fishbein and Ajzen (1975), intention also predicts behavior of compulsive acts.Insights can also be drawn from the competition theory.According to Pine and Gilmore (1999) it is competition (i.e.presence of alternative options) that necessitates differentiation.Thus, absence of alternative courses of action can constrain customers to be loyal to the available option.Summarily therefore, we argue that a combination of hedonic experience and relational experience suppress the influence of rational evaluation and intentions on behavior under situations of absence of volitional control.Corroborating our findings with the tenets of the power politics theory and the theory of reasoned action, we advance customer experience literature by showing that this postulation explains why customers find it desirable to go through favorable or unfavorable experiences to comply with government policy and remain loyal to their service firms as well as the role which intention plays in situations of nonvolitional behavior.

Conclusion, limitations and future research directions
By identifying the nature of the effect of the components of customer experience on customer loyalty arising from the implementation of a government policy, this paper provides a deeper understanding of what happens to the loyalty that consumers exhibit towards a service organization when negative service experiences ensue because the organization is simply implementing a policy initiated by a government or its regulatory agency.It is clear from our findings that a policy initiated by the government or its agency will not threaten customers' loyalty to the service organization implementing that policy provided that the organization's role in the implementation of the policy possesses some elements of emotional appeal while the customers rely more on their interactions with co-consumers.These evidences play out effectively in situations where the benefits of the introduced policy and the consumers' belief in the good intent of the policy are congruent and customers are susceptible to the manipulations of the social class leading to absence of volitional control.However, firms should not be deceived into relying too heavily on this type of loyalty because it is situational/contingent and according to Mascarenhas et al. (2006), situational loyalty is promiscuous.
As with every research that is survey-based, our study suffers some evident limitations.One of the inherent shortfalls of this paper is that it utilized a cross-sectional dataset to test a causal model.
According to Hong and Goo (2004) Hedonic experience (Overby and Lee, 2006): • The entire BVN registration exercise engaged me positively • The entire BVN registration exercise was fun for me   papers were published in error?Or were the authors of those articles right after all?We were still thinking.We came to an agreement that advancing the position that previous studies utilized same approach as our reason for utilizing parametric statistical techniques was an inadequate reason to give.As a remedial option, we went back to research methodology textbooks.
We discovered that before applying parametric tests, researchers ought to examine data to see if such data meet parametric statistical assumptions.
In our revised manuscript therefore, we tested our data for multicollinearity and normality assumptions (see the "Data screening, sample representativeness and common method variance" section for details of the revisions made).Outputs recommended in the literature (see Curran et al., 1996).With the above results, we conclude that data is suitable for parametric statistical tests".
We kindly request the reviewer to check out the revisions made to the revised manuscript.
Additionally, the reviewer may have noticed that most of the methodology and the analysis sections have been completely reorganized and different Thank you very much for this comment.Thank you very much for this positive overall assessment.We wish that the revisions that we made thereafter will meet your assessment standard this time around.To some extent Thank you very much for this comment.Thank you very much for recommending Mayer and Sparrowe (2013) to us.We had already read this article long before now.The integration approach that we adopted is the 'one phenomenon, seemingly disparate theoretical perspectives'.This enabled the dialogue we initiated in our paper.Statements such as the following enabled us to initiate the dialogue: "The intersecting point of the power politics theory and the TRA as applied in this paper is that irrespective of the perspective adopted, intention and behavior are influenced".To make our ideas more coherent however, we have massively reorganized the theoretical background section.In revised the discussion and contribution section accordingly.Please do kindly refer to the revised manuscript for details of all the revisions that we made.

Results and conclusion
You need to relook into the discussion of the findings and implications given the above comments.
Thank you very much.As noted above, changes in the results emerged.These changes affected our discussion, implications and conclusion as well.
We have therefore revised these sections accordingly.Because we made enormous revisions in this respect as well as in the analysis section, we didn't capture those revisions here.
We therefore kindly implore the reviewer to have a relook into the revised manuscript to confirm that we have actually revised our manuscript accordingly.

Best of luck with your research
Thank you very much for this encouraging comment.We really appreciate it.

Reviewer 3
Reviewer's comments Authors' responses The literature review has improved compared to the previous draft.Please see my comments.
Thank you very much for this comment.Thank you very much for this very comment.We have had a second look at our original manuscript.
First, we would like to inform the reviewer that we have made some revisions is this respect.Thank you very much for this comment.We have  Thank you very much for this comment.We have had a second look at the statement.We agree with you that the statement is not clear.We have therefore re-casted it as follows: "Additionally, the approval granted by the apex bank to commercial banks to allow customers who are finding difficult to link their BVNs to different accounts to effect needful corrections (Udo, 2016) foretells that customer experiences were not seamless but rather torturous".Additionally, if you look at the revised manuscript, we have moved the entire discussion on BVN policy to the  Thank you very much for this comment.Like we noted earlier, the entire introduction was rewritten.We believe that these issues are better clarified now than before.Please do kindly refer to the last paragraph of the introduction to see how we addressed this issue.
Like we said earlier, comments like this show that you actually read our article word for word.We are grateful for such sincere effort.We have revised our manuscript accordingly.Please kindly review the revised manuscript for confirmation.
6.The author/s should provide the definition of utilitarian, hedonic and relational experience in the literature review section pertaining to the customer experience rather than in power politics theory.
Thank you very much for this comment.The theoretical background section has been massively reorganized.We have therefore revised the manuscript accordingly.Please do kindly refer to the revised manuscript for details.section has been reorganized.The 'power politics theory' section, the 'theory of reasoned action' section as well as the 'similar and competing aspects of the power politics theory with TRA'  Blasco-Arcas et al., 2014;Cyr et al., 2007).All of these terms reflect the influence of other customers on an individual customer's service experience".We hope this meets your standard.
8. I suggest that the section highlighting 'similar as well as competing aspects of the Power Politics theory with TRA' should be separated from the hypothesis development.
Thank you very much for this comment.We have revised our original manuscript accordingly.A separate section titled 'hypotheses development' has been created.Please see the revised manuscript for confirmation.
9. I agree with the concern raised by reviewer 1 regarding the face validity of customer loyalty.
Especially if we observe the last two items of customer loyalty scale adopted in this study i.  this paper for confirmation.Additionally, we developed measures of customer loyalty based on the work of Zeithaml et al. (1996).The reviewer may wish to also have a look.The above articles were published in top journals.Finally, the reviewer 1 that initially pointed out this concern tends to be happy with the response we gave in the first round of revisions.We are not trying to be mean about this.We just wish you could see this concern in the above lights as well as the response we provided in our initial revision.
Below was the response we provided in our first round of reviews: "Thank you very much on the comments relating to the validity of the measures of customer loyalty.While we quiet agree with the definition of customer loyalty put forward by Bell and Eisingerich (2007), we still believe that our measures of the construct passed face validity.This is because Bell and Eisingerich (2007) drew on the work of Zeithaml et al. (1996) to define and adapt measures of customer loyalty.In our conceptualization and measurement of customer loyalty in this paper, we did the same thing.The measures of customer loyalty that we adapted was strictly based on the behavioral intention scale of Zeithaml et al. (1996).Specifically, the customer loyalty measures that we adapted for the purpose of this research not only reflect consumers' intention to stay, commitment to increase depth and breadth, it also includes customers' willingness to recommend as operationalized by Zeithaml et al. (1996).Since the work of Zeithaml et al. (1996) passed face validity and construct Thank you for this kind observation.Again, this is a point blank evidence that you read through our work word for word.We have revised the manuscript accordingly.
11. Overby and Lee (2006) has used a scale on customer value (utilitarian and hedonic shopping values) and not utilitarian and hedonic experience.revised manuscript: "…because some components of customer experience are more influential than others in some contexts (Åkesson et al., 2014;Schmitt and Zarantonello, 2013).These contextspecific nature of customer experience imply that scholars should focus on specific components of customer experience that suits the study context.
To this end, previous operationalizations of customer experience within the banking sector (see Garg et al., 2014;Klaus et al., 2013) broadly reflect the cognitive/utilitarian, emotional/hedonic, and relational components.
Thus, this paper focuses on utilitarian, hedonic, and relational components of customer experience.While hedonic experience as was noted before aligns with shopping entertainment, shopping enjoyment or what is lately referred to as emotional experience, utilitarian experience is cognitive in nature and aligns with task instrumentality (Holbrook and Hirschman, 1982) which Babin et al. (1994) and Overby and Lee (2006) described as utilitarian value".Second, we adapted measures of relational experience from Hsu and Tsou (2011) because Overby and Lee (2006) did not measure relational experience in their research but Hsu and Tsou (2011) clearly spelt out those measures in their paper.We did not adapt the measures of utilitarian and hedonic experience from Hsu and Tsou (2011) even though they respectively correspond to think and feel because those measures were not as elaborate as the value scale we adapted from Overby and Lee (2006).We hope that these help clarify the above concerns.Thank you very much for this comment.Reviewer 2 suggested a different analytical approach which we subsequently executed in this round of revision.If you look at the revised manuscript, the entire analysis was re-analyzed.This led to new results.We have therefore revised our discussion and conclusion sections accordingly.We have also reinforced our contributions based on these changes.We kindly implore you to have another look at the revised manuscript for details of the corrections made.Please also kindly refer to our responses to the concern you raised in the '5.
confirmed that intention leads to behavior whilstMafabi et al. (2017) established that behavioral intentions mediate knowledge sharing behavior, we expect that this relationship will hold in the link between intention to open new account(s) and customer loyalty on the one hand and the mediating effect of intention to open new account(s) on the other hand in situations of compulsion.We therefore propose the following hypotheses: H4.Intention to open new account(s) will positively predict customer loyalty H5.Intention to open new account(s) will positively mediate the effects of a) utilitarian experience; b) hedonic experience; and c) relational experience on customer loyalty -----Insert Figure 1 about here----- Accordingly,Zhao et al. (2010, p. 204)  outlined the following conditions for establishing mediation effects: "i.If a × b is significant but c is not, you have indirect-only mediation.ii.If a × b is not significant but c is, you have direct-only nonmediation.iii.If neither a × b nor c is significant, you have no effect nonmediation.iv.If both a × b and c are significant, determine the sign of a × b × c by multiplying the three coefficients, or by multiplying c by the mean value of a × b from the bootstrap output.If a × b× c is positive, it is complementary mediation; if a × b × c is negative, it is competitive mediation".Zhao et al. (2010) further argued that the relevant outputs to capture when reporting the results of a mediation test are the coefficient of a × b or mean value of a × b, the 95% confidence interval from the bootstrap analysis, and the unstandardized regression coefficients of a, b, and c.
here----To examine H5a-c, the Preacher-Hayes bootstrapping script based on 5000 resampling and 95% confidence interval recommended byZhao et al. (2010) was employed.The outputs are shown in Table4.With regard to the mediation effect of intention to open new account(s) on the relationship between utilitarian experience and customer loyalty, we found that the mean indirect effect from the bootstrap analysis is positive and significant (a × b = 0.215), with a 95% confidence interval excluding zero (0.125 to 0.335).In the indirect path, a unit increase in utilitarian experience increases intention to open new account(s) by a = 0.616 units; b = 0.348 (see Table4), so holding constant utilitarian experience, a unit increase in intention to open new account(s) increases customer loyalty by 0.348 units on a scale of 0 to 1.The direct effect c (0.449) is also significant (p < 0.001); holding intention to open new account(s) constant, a unit increase in utilitarian experience leads to 0.449 increase in customer loyalty, perhaps reflecting rationality or cognitive reasoning effects.Since a × b × c (0.097) is positive, it is a complementary mediation.With regard to the mediation effect of intention to open new account(s) on the relationship between hedonic experience and customer loyalty, we found that the mean indirect effect from the bootstrap analysis is positive and significant (a × b = 0.116), with a 95% confidence interval excluding zero (0.066 to 0.182).In the indirect path, a unit increase in hedonic experience increases intention to open new account(s) by a = 0.413 units; b = 0.280 (see Table 4), so holding hedonic experience constant, a unit increase in intention to open new account(s) increases customer loyalty by 0.280 units on a scale of 0 to 1.The direct effect c (0.366) is also significant (p < 0.001); holding intention to open new account(s) constant, a unit increase in hedonic experience leads to 0.366 increase in customer loyalty, perhaps reflecting enjoyment or entertainment effects.Since a × b × c (0.042) is positive, it is a complementary mediation.Finally, with regard to the mediation effect of intention to open new account(s) on the relationship between relational experience and customer loyalty, we found that the mean indirect effect from the bootstrap analysis is positive and significant (a × b = 0.167), with a 95% confidence interval excluding zero (0.085 to 0.277).In the indirect path, a unit increase in relational experience increases intention to open new account(s) by a = 0.604 units; b = 0.276 (see Completing the BVN registration exercise diverted my attention away from most of my worries • Having a BVN makes me feel like I'm in another world • Overall, linking up my accounts with the BVN was fun for me • I feel good after linking up all my accounts with BVN Relational experience (Hsu and Tsou, 2011): • I think the BVN registration exercise offered me the opportunity to be seen as a genuine member of the community of Nigerian bank customers • With a BVN, I am sure that my relationships with other bank customers (e.g.transferring money) is guaranteed • The BVN registration exercise makes me think about social rules and activities • Overall, the entire BVN registration and account linking offered me the opportunity to share my experiences with other bank customers Intention to open new account(s) (developed in accordance with Rose et al., 2012): • I intend to open new accounts with my bank in the future • I intend to open new accounts with other banks in the future • I think the account(s) I have currently will serve me in the future Customer loyalty (Zeithaml et al., 1996): • I say positive things to other people (e.g.friends, colleagues etc) about how my bank handled the entire BVN exercise • I recommend my bank to someone who seeks my advice because of the way they handled the entire BVN exercise • I encourage friends and relatives to use my bank because of how I was treated during the BVN exercise • I consider my bank as the first choice because of the way they handled the entire BVN exercise show that our data was indeed suitable to warrant the application of parametric statistical techniques.To this end, we made the following revisions to the original manuscript: "Prior to the main analysis, data was assessed for the appropriateness of parametric statistical tests by examining the parametric assumptions of multicollinearity and normality.The multicollinearity diagnostics were employed to ensure that the predictors are not redundant while the normality tests were aimed at detecting and resolving extremely skewed data.The tolerance value and variance inflation factor (VIF) were used to test for potential multicollinearity issues.Results indicate that the tolerance value ranged from 0.544 to 0.622 while the VIF ranged from 1.607 to 1.839.Based on the criterion set by Hair et al. (2010), the estimation of our model is not because the tolerance values were all substantially above 0.10 while the variance inflation factor (VIF) were all well below 5. Data normality can be assessed through histogram (which is a subjective measure) and skewness and kurtosis (which are objective measures).To further examine whether our data is suitable for parametric statistical tests, both subjective and objective measures were utilized to assess the distribution of our data.First, the histogram (although not reported here in the interest of presentation parsimony, the data that produced the results can be made available on request) for the aggregate measures of the constructs within the model indicate that data is normally distributed.Additionally, results indicate that negative and positive values of skewness for all the measurement variables were respectively in a range of -0.228 to -0.947 and 0.000 to 1.665.Results also indicate that negative and positive values of kurtosis for all the measurement variables respectively ranged from -0.133 to -2.041 and 0.011 to 5.173.Given the above findings, we conclude that the data is normally distributed because the range of skewness and kurtosis values were respectively below 2 and 7

5.
Implications for research, practice and/or society: Does the paper identify clearly any implications for research, practice and/or society?Does the paper bridge the gap between theory and practice?How can the research be used in practice (economic and commercial impact), in teaching, to influence public policy, in research (contributing to the body of knowledge)?What is the impact upon society (influencing public attitudes, affecting quality of life)?Are these implications consistent with the findings and conclusions of the paper?You should not ignore social implications because it is a social science.Thank you very much for this comment.We have included a social implication of our research.The following has been included in the abstract of our paper: "Social implications -Based on our findings, it has come to the fore that consumers will be at the receiving end of a government policy poorly implemented by service organizations.When such policies are rolled out therefore, governments should enforce operational modalities that will forestall potential negative experiences that consumers would have encountered".Please do kindly refer to the revised manuscript for details 6. Quality of Communication: Does the paper clearly express its case, measured against the technical language of the field and the expected knowledge of the journal's readership?Has attention been paid to the clarity of expression and readability, such as sentence structure, jargon use, Thank you very much for this comment.It is indeed very encouraging.DETAILED COMMENTS: Paper is better than last version, thank you.I understand your point about statistics choice but parametric tests are not good enought for social science.

2.
Relationship to Literature: Does the paper demonstrate an adequate understanding of the relevant literature in the field and cite an appropriate range of literature sources?Is any significant work ignored?No Thank you very much for this comment.

4
. Results: Are results presented clearly and analysed appropriately?Do the conclusions adequately tie together the other elements of the paper?To some extent Thank you very much for this comment.

5.
Implications for research, practice and/or society: Does the paper identify clearly any implications for research, practice and/or society?Does the paper bridge the gap between theory and practice?How can the research be used in practice (economic and commercial impact), in teaching, to influence public policy, in research (contributing to the body of knowledge)?What is the impact upon society (influencing public attitudes, affecting quality of life)?Are these implications consistent with the findings and conclusions of the paper?

6.
Quality of Communication: Does the paper clearly express its case, measured against the technical language of the field and the expected knowledge of the journal's readership?Has attention been paid to the clarity of expression and readability, such as sentence structure, jargon use, Thank you very much for this comment.too long and does not get to the objective straight forwardly.The discussion on customer experience in the first two paragraphs can be included in your discussion of customer experience.The discussion about the policy can be included in a separate section in theoretical background.The author(s) need to provide a brief background and present why it is important to measure customer experience in their context.Please refer to Grant, A. M., & Pollock, T. G. (2011).Publishing in AMJ-Part 3: Setting the hook.Academy of Management Journal, 54(5), 873-879 for how to write the introduction.Thank you very much for this comment.First, we had like to specially thank you for referring us to Grant, A. M., & Pollock, T. G. (2011).Publishing in AMJ-Part 3: Setting the hook.Academy of Management Journal, 54(5), 873-879.That referral alone has shaped the way we see the introductory part of a research paper and what its contents should be.As such, we have done exactly what the reviewer asked us to do.The first two paragraphs have already been moved to the customer experience section.We have also housed the discussion on policy in the theoretical background section.More importantly, the entire introduction was rewritten to reflect the approach suggested by Grant and Pollock (2011).We kindly request the reviewer to refer to the revised manuscript for these revisions.Literature review I have certain issues in this part.While i like the idea of using Power Politics theory in the background of the policy change, it is definitely not the theory that is being used to examine their research objective.While it is always good to bring in novel perspective to understand a phenomenon by integrating or bringing new theories, it should be done properly.Again refer to Mayer, K. J., & Sparrowe, R. T. (2013).Integrating theories in AMJ articles.Academy of Management Journal, 56(4), 917-922 on how to integrate theories.In this study, the theoretical framework on which the model or relationships are proposed and tested in TRA and power politics theory is more to do with balance of power which is not the case in your study.
4. Results: Are results presented clearly and analysed appropriately?Do the conclusions adequately tie together the other elements of the paper?Yes Thank you very much for this comment.

5.
Implications for research, practice and/or society: Does the paper identify clearly any implications for research, practice and/or society?Does the paper bridge the gap between theory and practice?How can the research be used in practice (economic and commercial impact), in teaching, to influence public policy, in research (contributing to the body of knowledge)?What is the impact upon society (influencing public attitudes, affecting quality of life)?Are these implications consistent with the findings and conclusions of the paper?Most of the results are obvious and in line with prior studies.This limits the contribution of this study as we still don't know does an exercise such as BVN forces the banks to change its ways it provides experience to customers.The author/s need to clearly highlight this aspect.
included the following clarifications to the revised manuscript to justify why we focused on the during stage and to also further clarify that 'during' as used here implies the implementation stage of the BVN policy when commercial banks were directed by the apex bank to deliver the service to their customers: "The 'during' stage of customer experience reflects the implementation stage of the BVN policy when commercial banks were directed by the apex bank to deliver this service to their customers.We focused on the implementation stage of the BVN because as noted byKlaus (2013), utilitarian, hedonic, and relational attributes of customer experience play out effectively in the 'during' stage of the consumer buying process.This paper draws on the BVN policy implemented by the Nigerian apex bank to demonstrate how customer experience will affect consumer behavior".
3. "Additionally, for the apex bank to have granted approval to commercial banks to allow customers finding it difficult to link their BVNs to their different accounts to effect needful corrections(Udo, 2016) implies that customers must have been having torturous experiences with their banks as a result of this policy" (pg. 3. Ln 50).This statement is not clear.

4.
Following are your research questions: "why will a customer find it desirable to go through these experiences just to obtain this so-called unique BVN? ii) Why will customers still remain loyal to their banks irrespective of the onerous experiences they passed through in obtaining a BVN? and iii) In the face of these upturns, what is the role of intention to open new accounts?"I believe that these research questions needs better positioning.It is mandatory for each customer to go through such an exercise irrespective of whether he/she finds it desirable or not.The BVN policy is implemented by all the banks based on the instructions given by the CBN.So, I believe that the source of undesirable customer experience is not in the control of the banks.Therefore, why should the customers' intention to open bank accounts and their loyalty towards a particular bank get affected?It needs to be clearly highlighted in the introduction.
section have been merged to form 'Power politics theory vs the theory of reasoned action (TRA).In the fourth paragraph of this section, we further discussed how the concept of subjective norm was integrated into this study.The integration now reads as follows: "Subjective norm measures the influence which relevant "others" have on an individual customer.Thus, it corresponds to relational experience because subjective norm is similar to what Grove and Fisk (1997) termed 'other customer influences' whilst as numerous studies likened relational experience to social value, customer-to-customer interactions, and social presence (seeMohd-Any et al., 2015; e. "I consider my bank as the first choice because of the way they handled the entire BVN exercise" and "I intend to use my bank more in the coming years because of the way they handled the BVN Thank you very much for this comment.We had like to stress here that we didn't develop these measures off hand.Like we pointed out in the 'operationalization of the research constructs' section, we developed measures of 'intention to open new account(s)' based on the work of Rose et al. (2012).The reviewer may want to refer to Please provide justification for equating customer value with customer experience.Besides,Hsu and Tsou (2011) have provided 5 dimensions of customer experience viz.sense, feel, think, act and relate.Then why have you only employed only "relational experience" from the scale.Thank you very much for this comment.We revised our manuscript accordingly.Regarding your concern on equating customer value with customer experience, the following has been included in the 'operationalization of research constructs' section of the revised manuscript: "….Additionally, it was previously argued that both utilitarian and hedonic dimensions of value are experiential because the former reflect cognitive aspects of experience while the later reflect emotional aspects of experience such as entertainment for instance (Overby and Lee, 2006).Previous research (see for instance Lemke et al., 2011) also studied customer experience from the viewpoint of value-in-use.Thus, value and experience are strictly interwoven whilst value scales can serve as surrogates for experience scales.For instance, Nambisan and Watt (2011) measured customer experience with measures adapted from value scales.In line with the foregoing therefore…….".Regarding your concern on why we employed only relational experience from Hsu and Tsou's (2011) scale, we have the following responses.First, the following has been included in the 'Customer experience: a brief state-of-the-art, meaning and its components' section of the

Examining customers' experience with the Nigerian Bank Verification Number (BVN) policy from the perspective of a dual-lens theory Abstract
IntroductionIt has been established that over 50% of B2B and B2C customers stopped buying after a negative experience(Dimensional Research, 2013)whilst a global Consumer Pulse Research conducted by

http://mc.manuscriptcentral.com/ijoem International Journal of Emerging Markets
also noted that rules that govern behavior may be externally imposed by "agents who have our interests in mind" or internally "'constructed' by ourselves as we see the need for

ijoem International Journal of Emerging Markets
behavioral intention battery.All the measures used are http://mc.manuscriptcentral.com/ This location was selected because it has the highest concentration of bank customers who participated in the BVN exercise in Abia State of Nigeria.Participants were conveniently and purposively recruited by mall-intercept.283participantswereapproached.246subjectsparticipated of which 42 responses were incomplete and were discarded.The balance was considerably complete and yielded a valid response rate of 82.9%.Table1summarizes the demographic profile of the respondents.Respondents' gender distribution indicates that 54.9% were male while 45.1% were female.In terms of age, 91.7% were 55 years or younger while the 8.3% were 56 years or older.This clearly indicate that the respondents were within the very active population which is the main target for this research.
The tolerance value and variance inflation factor (VIF) were used to test for potential multicollinearity issues.Results indicate that the tolerance value ranged from 0.544 to 0.622 while the VIF ranged from 1.607 to 1.839.Based on the criterion set byHair   et al. (2010), the estimation of our model is not affected by multicollinearity because the tolerance values were all substantially above 0.10 while the variance inflation factor (VIF) were all well below 5. Data histogram (although not reported here in the interest of presentation parsimony, the data that produced the results can be made available on request) for the aggregate measures of the constructs within the model indicate that data is normally distributed.Additionally, results indicate that negative and positive values of skewness for all the measurement variables were respectively in a range of -0.228 to -0.947 and 0.000 to 1.665.Results also indicate that negative and positive values of kurtosis for all the measurement normality can be assessed through histogram (which is a subjective measure) and skewness and kurtosis (which are objective measures).To further examine whether our data is suitable for parametric statistical tests, both subjective and objective measures were utilized to assess the distribution of our data.First, the http://mc.manuscriptcentral.com/ijoem

ijoem International Journal of Emerging Markets
Zhao et al. (2010)e equations procedure,Zhao et al. (2010)indicated that the one and only condition to establish a mediation effect is that the indirect effect is significant when either a Sobel test or a superior bootstrap is conducted.Zhao et al.'s (2010)recommended bootstrapping procedure was employed in this paper because of its advantages over the Sobel's test which is insufficient to establish a mediation effect.
Zhao et al. (2010)through the Preacher-Hayes bootstrapping script in SPSS.As against the Baron and Page 14 of 48 http://mc.manuscriptcentral.com/

Table 3 )
indicate that 44% of the total variance in customer loyalty is explained by a combination of utilitarian experience, hedonic experience, relational experience, and intention to open new account(s).Thus, according to the rule of thumb specified byHair et al. (2011), the predictive accuracy of the model is moderate.The effect size which was derived with the formula below indicate that the sizes of the detected effects were small based on theCohen's (1988 in Field and Hole, 2003)convention.

Table 4
customer loyalty by 0.276 units on a scale of 0 to 1.The direct effect c (0.598) is also significant (p < 0.001); holding intention to open new account(s) constant, a unit increase in hedonic experience leads to 0.598 increase in customer loyalty, perhaps reflecting 'other customers' effects.Since a × b × c (0.100) is http://mc.manuscriptcentral.com/

•
, in any model where causality is proposed especially for consumer behavior, longitudinal design provide stronger inferences.Additionally, one might argue that since the paper deals in a policy issue, longitudinal design would have provided stronger inferences also.Thus,Ding, C. G. and Tseng, T. H. (2015), "On the relationships among brand experience, hedonic emotions, and brand equity", European Journal ofMarketing, Vol.49No.7/8,pp.994 -1015.Downs, D. S. and Hausenblas, H. A. (2005), "The theories of reasoned action and planned behaviorLemke, F., Clark, M. and Wilson, H. (2011), "Customer experience quality: an exploration in business and consumer contexts using repertory grid technique", Journal of Academy of Marketing Science, The time I spent in completing the BVN registration was not a waste for me • I think completing the BVN registration was the best decision I took to ensure that my account is Zhao et al. (2010)eoretical grounding of our paper strongly support the study design, we strongly recommend that future research should examine customer experience-customer behavior models in similar situations with longitudinal design.Additionally, given that the explanatory power of our model is only modest.Future research can integrate other predictors of customer loyalty and test the model across several services fields.Of particular interest are non-financial services because the risky nature of financial services makes it unique and different from other services types.It may also be insightful to integrate variables like switching barriers and perceived relational investments into such models because both constructs align very suitably to the power politics perspective and could be applied to further explain situational loyalty arising from a policy implementation.Additionally, it should be noted that intention to open new account(s) is a complementary mediator in the relationship between the components of customer experience and customer loyalty.But according toZhao et al. (2010), when a complementary mediation effect is detected, it shows that even though the identified mediator is consistent with the hypothesized theoretical framework, the model is theoretically incomplete.This implies the likelihood of an omitted mediator in the 'direct' path.Thus, further research can integrate other possible mediators into the conceptual model tested in this paper.To this extent, it might be interesting to examine whether regrets and switching intentions are possible mediators.http://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoemhttp://mc.manuscriptcentral.com/ijoem

Page 27 of 48 http://mc.manuscriptcentral.com/ijoem International Journal of Emerging Markets
I intend to use my bank more in the coming years because of the way they handled the BVN exercise • http://mc.manuscriptcentral.com/ijoem

Table 2 Correlation matrix and descriptives for measurement scale
Note: α = SD = standard deviation; Cronbach alpha; CR = composite reliability; AVE = average variance extracted; Values in bold in the diagonal are the square root of the AVE.

com/ijoem International Journal of Emerging Markets
7. pg.7 ln.39 "Subjective norm measures the influence which relevant "others" have on an individual customer."How have you incorporated Thank you very much for this comment.Like we noted earlier, the entire theoretical background http://mc.manuscriptcentral.

Page 44 of 48 http://mc.manuscriptcentral.com/ijoem International Journal of Emerging Markets
exercise".I wonder how these items do not overlap with the items of customer's intention to open new accounts.The current explanation in this regard is inadequate.

Page 45 of 48 http://mc.manuscriptcentral.com/ijoem International Journal of Emerging Markets
Overby and Lee (2006)t perceive the face validity issue you raised as acute as it seems at first glance".We hope this helps.10.You have statedOverby and Lee (2007)in the main text whereasOverby and Lee (2006)in the references.Please ensure that all the references are correctly stated.

Page 47 of 48 http://mc.manuscriptcentral.com/ijoem International Journal of Emerging Markets
12. The results of the study are obvious and are in line with the prior studies.This limits the contribution of the study as it seems that BVN exercise has no impact on the hypothesized relationships.