Rewarding and Punitive Switching Barriers in a Continuous Purchasing Setting : The Impact of Relationship Age

Continued growth and expansion of online banking services has further enhanced the importance of customer retention for banks. However, understanding of factors that impact the efficacy of switching barriers in the banking industry continues to be inadequate, particularly in the case of online banking. The reasons for switching service providers have been studied to some extent but what has not been examined adequately is why some customers decide not to switch after having thought of switching. This paper examines the reasons why some customers of banks decide not to switch after seriously considering a switch. The role that the age of the customer’s relationship with the existing bank in the decision not to switch is also studied. Effectiveness of switching barriers in this context is measured by four factors, namely, service recovery, trust, switching costs and lack of alternative attractiveness. Consistency and reliability of these four factors is confirmed by confirmatory factor analysis and reliability tests. Results show that the relationship age between the bank and the customer affect the effectiveness of switching barriers. In particular, those with longer relationships with their main banks claimed that it was more difficult to change to another main bank than those with shorter relationships.


Introduction
Banking has become an intensely competitive industry, more so because of the frequent disruptions caused by technological advancements, making it incumbent upon banks to make all out efforts to retain existing customers (Clayton-Smith, 1996; Thomas, 2001; Wong & Wong, 2012).The general belief is that it is always less expensive to retain an existing customer than acquiring new customers (Stone, et al., 1996), though new customers too need to be acquired for sustaining growth.Efficient service and long-term relationships are recognized as the key essentials for continued growth.Not surprisingly, banks are placing considerable emphasis on finding ways of satisfying existing customers to a degree that ensures low switching by them to competing banks (Reichheld, 1996;Wong & Wong, 2012).Empirical evidence presented by Reichheld (1996) suggests that under certain conditions, the ability to retain customers can have a hugely positive impact on a bank's profit.
In reality retaining existing customers is becoming increasingly difficult because digitalization of the banking industry has made it considerably easier for consumers to hunt for alternative suppliers of banking services.In fact digitalization has resulted in Non-Banking Financial Companies (NBFCs) to compete with banks in several areas, such as educational loans.Margins in the banking industry have been under pressure and entry of NBFCs in areas hitherto served by banks has made things even more difficult.The increasingly aggressive consumers are demanding more at lower costs, which means banks have to think more creatively and be more efficient to retain customers (Shankar, et al., 2003).
While enhanced customer satisfaction is certainly the best way of retaining customers, the need for erecting suitable and effective switching barriers cannot be ignored (Lee, et al., 2001;Ranaweera & Prabhu, 2003;Wong & Wong, 2012).The cost of switching does impact decisions of customers who are apparently dissatisfied with the service.Considering banking services are essentially a continuous purchasing setting, customer retention is not feasible without some switching barriers (Ranaweera & Prabhu, 2003;Wong & Wong, 2012).

Research gaps
Technological advancements are resulting in increasing proliferation of online banking and are forcing banks to rethink almost all components of their operations.While potential opportunities for enhancing customer satisfaction are growing, implying it should become easier to retain customers, it is also true that customers are finding it increasingly easier to switch to alternate service providers.In the increasingly complex scenario, new research gaps are emerging rapidly.Apparently there are three critical gaps in extant research that need to be addressed, particularly in the context of online banking: i.
First, different dimensions and configurations of switching barriers used traditionally in retail banking (Tesfom & Birch, 2010; Valenzuela, 2010) need to be examined afresh in the context of the growing usage of online retail banking services. ii.
Secondly, the reasons for switching have been examined but why customers decide to stay after having considered switching has not been investigated adequately.
iii.The structure of this paper is as follows.
The second section provides a review of existing literature on switching barriers and the third section explains the methodology, while the fourth section reports the findings.Conclusions are described in section five.

Service recovery
When customers are satisfied with the service they receive, particularly recovery after some incident that triggers dissatisfaction, their commitment to the service provider strengthens further (Colgate & Danaher, 2000;Gwinner, et al., 1998).While incidents that trigger dissatisfaction are often the reason why a customer thinks of switching, effective steps taken by a bank to redress the situation, commonly described as service recovery, can often result in the customer-  1985).When the consumer has trust in the service provider, the relationship is strengthened since the risk perception and confidence lead to lower transaction costs for both.Naturally, trust encourages the buyer to stay with the same service provider (Ratnasingham, 1998).

Punitive switching barriers
Punitive switching barriers refer to the native reasons that remain in a relationship (Hirschman, 1970).Generally speaking, there are two main punitive barriers to switching in the retail banking industry: high switching costs and lack of alternative attractiveness.

Switching costs
There is a cost attached to not only the finding of an attractive alternative but also to termination of an existing relationship and it is these two

Relationship between switching barriers and relationship age
In this context, one factor that has not been examined adequately is the relationship age, i.e. the duration for which the customer has had a relationship with the current service provider.Like all other relationships, customers' switching barriers also evolve over time.In fact a _____________________________________________________________ ______________ Chi Bo WONG, Ka Li Kelly WONG, Hing Cheong NG and Chiu Ping CHOU (2018), Journal of Electronic Banking Systems, DOI: 10.5171/2018.812508 positive correlation between the market value of a firm and the average age of its relationships with its customers has been reported in prior research (Galbreath, 2002) since the relationship age has a positive effect on corporate reputation (Bartikowski, et al., 2011) and profitability also (Reinartz & Kumar, 2003).As the age of a relationship increases, the effectiveness of switching barriers is enhanced since the trust the customers have in the relationship grows with passage of time and accumulation of experience (Palmatier, et al., 2006).Besides, customers are often able to gain more benefits from the same relationship as the relationship matures, which implies lower risk perceptions and a higher sense of security (Dagger & O'Brien, 2010).

Selection of industry
The context of this empirical research was online retail banking industry in Hong Kong.Banking, per se, constitutes a continuous purchasing setting where customer satisfaction and switching barriers have high impact on customer retention (Ranaweera & Prabhu, 2003).Continuous purchasing setting is qualitatively different from normal purchasing where a customer has the discretion of sourcing from different suppliers at its own discretion.The relationship between a bank and its customers is generally of a long-term nature.Switching a bank is not an easy decision, more so in case of online banking, because switching requires considerable time, money and effort (Ranaweera & Prabhu, 2003; Wong & Wong, 2012).Therefore, switching decision is often considered and reconsidered even after a decision has been made in principle.

Target population and criteria of sample
Online banking services are used by retail as well as business customers.The target population of this research only included retail customers.The sample for this empirical research was selected on the basis of four specific criteria.The four criteria are that a user must (1) be above 18 years of age, (2) have carried out at least one online banking transaction with his or her main online bank in the preceding month, (3) have registered at least one complaint with his or her main online bank in the preceding 12 months, and ( 4) have decided to continue with his or her existing main online bank after having thought of switching.

Measurements
Dimensions of switching barriers used in online banking have not been examined exhaustively.Therefore, dimensions of barriers used in the traditional retail banking were used.These barriers can apparently be applied to online banking services too since in the context of switching barriers, the traditional banking services and online banking services are quite similar.Four switching barrier constructs, namely, service recovery, trust, switching costs and lack of alternative attractiveness, which have been validated in the past, are used in this research.Each construct comprises of three items, measured on seven-point Likert-type scale with anchors "1 = strongly disagree" and "7 = strongly agree".Definitions of these four constructs are as listed in Table 1

Constructs Definitions
Service Recovery Service recovery means a bank improves its service in response to a customer's dissatisfaction.

Trust
Trust is considered as a factor affecting relationship commitment.

Switching Costs
Switching costs relate to the level of difficulty involved in switching to another main online bank.

Lack of Alternative Attractiveness
Lack of attractiveness refers to absence of other attractive online banks in the market.

Questionnaire design
The questionnaire prepared on the basis of the constructs and the items was tested by obtaining opinions of others, such as some online banking professionals in Hong Kong (Converse & Presser, 1986), in order to identify any bias, besides any additional item that may help answer the research objectives.As is routine in research, appropriateness and suitability of the structure, as well as comprehensibility of the questionnaire was examined by a pilot study.Five lecturers teaching marketing, five managers of three local banks, and twenty online retail banking users were consulted.This resulted in rephrasing of some of the questions.

Data collection
Data collection was by way of a questionnaire survey.A few students of marketing were recruited for holding personal interviews with respondents selected from among shoppers coming out of three large shopping malls (Times Square, Langham Place and New Town Plaza) in Hong Kong.The screening process was used for selecting suitable candidates who met the three specific criteria described earlier.Once the potential respondents and interviewees were selected, their consent was sought and obtained for voluntary participation.

Respondent characteristics
The respondent characteristics are shown in Table 2. Of the 800 responses received for the survey, 52.6% of respondents were male and 47.4% of respondents were female.For the age distribution, each of the age groups of 18-24 (24.6%), 25-34 (26.9%) and 35-44 (24.1%) accounted for about a fourth of the sample size.For the respondents' education levels, about 39.2% of the respondents held undergraduate or postgraduate degrees and 41.9% had less than a certificate or diploma.For relationship age, an overwhelming majority (77.4%) was of those who said they had relationship with their main online banks more than four years and the rest (22.6%)only had relationship with their main online banks up to four years.

Measurement model analysis
AMOS21 was used for confirmatory factor analysis (CFA) and the results confirm suitability of the four-factor structure comprising service recovery, trust, switching costs and lack of alternative attractiveness for measuring effectiveness of switching barriers.Model fit indices are as shown in Table 3.All model fit indices have acceptable values (Hair, et al., 1988).

Reliability, validity, descriptive statistics and correlation coefficients
Reliability and validity of the constructs is shown in Table 4, which included standard item loading, average variance extracted (AVE), construct reliability (CR) and Cronbach's Alpha (Anderson & Gerbing, 1988).All standard item loadings were found to be larger than 0.5, which confirms acceptable construct validity (Cheung, et al., 2000).As AVE values are all >0.5 and all  (Nunnally, 1978).

Discussion of Results
In the banking industry, customers often consider switching to alternative banks but eventually decide to continue with the existing bank.While reasons for switching banks have been researched quite extensively, why some customers give up the thought of switching after having seriously considered switching do not appear to have been examined adequately.
Validity of the four-factor structure used for measuring perceptions of switching barriers, internal consistency and reliability are confirmed by the confirmatory factor analysis.
The sample of this research included customers having short relationships (up to four years) as well as customers having long relationships (more than four years) with their respective main online banks.The results of independent-samples t-tests show that these two segments of customers clearly have different perceptions of switching barriers.
Specifically, those with longer relationships generally find it more difficult to switch to another bank than those with shorter relationships.Consequently, it is suggested that banks should develop various strategies to improve service recovery, to enhance customer trust, to increase customers' switching costs, and to improve overall attractiveness in order to retain existing customer.The rationale of these suggestions is that the increase of relationship age will also increase the customers' switching barriers.

Limitations and Future Research
Whatever methodology a research work uses cannot be true or false; it can only be more or less useful (Silverman, 1994).However, all research works face limitations of one kind or another and that applies to this research work also.

Table 5 : Switching Barriers by Relationship Age
There are two limitations and suggestions for future research in this research.First, generalizability of the conclusions of this research to a broader range of businesses should be attempted.Instead of focusing only on online banking, it may be desirable that future research may examine and test the four identified factors in other settings such as e-commerce space and other online services like travel related services.Secondly, grouping of the respondents into those with shorter and longer relationship seems too broad and generic in nature.Future research may compare if there are significant differences in the perceptions of switching barriers in terms of customers' .812508 sophistication level of banking service usage in both segments of short and longterm relationships as sophistication level of banking service usage may also impact perceptions and decisions related to switching.