Given the importance of customer loyalty in retail banking, there is a need for further investigation into various facets, internally within the firm and externally, that may influence it or contribute to its enhancement. This study therefore examines the influence of corporate governance practices on customer loyalty in the retail banking sector in Nigeria. This investigation is hinged on the observation of the general loss of confidence in banks as a result of the instability in the banking sector occasioned by recurrent bank failures (Adeyemi 2011). The banking failure scenario and subsequent erosion of customer confidence suggest that there is a need to focus on factors that will engender customer loyalty issues beyond the usual marketing strategies adopted by banks, such as advertising and personal selling. This position is substantiated by Cronin, Brady and Hult (2000) who aver that retailers generally have little knowledge of the antecedents of customer loyalty. Also, Brady et al. (2005) argued that despite various studies focusing on drivers of loyalty, scholars as well as practitioners are still lacking in the understanding of loyalty determinants vis-à-vis corporate governance practices and their relative importance. There is no consensus amongst researchers on whether there are specific factors, especially those that are related to corporate governance practices in the banks, that influence customers’ loyalty disposition. Several studies have investigated how relationship marketing, a discipline with a central focus on customers, enhances organisational performance (Bobalca 2014) and the effect of corporate governance on firm financial performance (Peters & Bagshaw 2014). Despite these efforts by academics, little or no attention has been directed at the interaction between relationship marketing and corporate governance-related research. The selected corporate governance practices utilised in this study are indeed those that are in the public domain, observed by the customers and therefore not the traditionally highly sensitive issues that are basically known by the bank role players alone. Empirical evidence is therefore needed to better understand how these corporate governance practices individually and collectively influence customer loyalty in Nigeria retail banks.

To investigate the relative interactive influence of corporate governance practices¹ on customer loyalty dispositions to banks.

Corporate governance is a system or an arrangement that comprises a wide range of practices (accounting standards, rules concerning financial disclosure, executive compensation, size and composition of corporate boards) and institutions (legal, economic and social) that protect the interest of the corporation’s owners (Obiyo & Torbira 2011). Literature suggests that, over the years, globalisation has substantially facilitated the evolution of best practices amongst corporations and countries. The intense competition amongst firms in international markets has necessitated the drive to update corporate governance practices on a regular basis, with the aim of attaining high-quality corporate governance standards.

Existing literature on banks’ corporate governance has focused on various practices and mechanisms to evaluate a corporation’s performance. These include how board composition affects performance, as well as to which extent board characteristics (diversity, external ties, and so on) affect a firm’s competitiveness and performance (Rashid 2018). Furthermore, Shen and Chih (2007) investigated corporate governance scores for Asian firms using characteristics that include management discipline, transparency, independence, accountability, responsibility, fairness and social awareness.

Other studies investigated to which extent board size, the director’s experience and other activities affect the performance (Mehran 1995; Muchemwa, Padia & Callaghan 2016). The performance measures varied from profit margins, expected stock returns, firm valuation to market valuation (Black, Jang & Kim 2006; Choudhary 2015). The variables for this study are obtained from an analysis of the Nigerian Code of Best Practices (2003), the Code on Corporate Governance for Banks (2006) and from previous empirical studies on corporate governance.

As the focus of this study is on customer loyalty, five main corporate governance indicators were identified to be relevant to this study, namely integrity of leadership, competence of leadership, transparency and disclosure of information, a complaints management system and corporate reputation or image. These five corporate governance practices identified were chosen based on the weight of research suggesting their importance to organisational or institutional performance and their relevance to customers. (Esmaeilpour & Barjoei 2016; Ghazizadeh et al. 2010; Gladson 2007; Kim & Brymer 2011).

Empirical studies highlight the notion that corporate governance has a leadership dimension, as it provides directional leadership to organisations by creating an enabling environment that integrates and systematises various collaborative efforts for setting objectives and achieving corporate goals (Fama & Jensen 1983; Ogbechie & Koufopolous 2010). Boards of directors and senior management represent leadership of any organisation and consequently have the ultimate internal authority within a company, and they play a key role in the overall overseeing of the company, especially monitoring top management (Chatterjee et al. 2003).

Competence of leadership is recognised as one of the key characteristics of a good leader. A competent leader has the ability to solve complex issues and to make good decisions that will increase efficiency and effectiveness in an organisation (Jones & George 2003). Acting in their capacity as leaders, the board of directors is expected to bring specialised expertise and a wealth of knowledge to the company (Weir & Lang 2001). Given the opacity and complexity that characterise financial institutions, it has become pertinent for a board of directors to understand the inner workings of the bank in order to discharge their functions effectively.

Transparency of financial information is very crucial in a firm’s operations, relationship with other stakeholders and sustainability. Accordingly, transparency and disclosure of information are widely considered key elements of good corporate governance in banks (Hawley 1996; Ngerebo & Sweneme 2012). They are described as important pillars of a corporate governance framework that enable adequate information flow to various stakeholders. As such, stakeholders are provided with the necessary information to judge whether their interests are being taken care of, or not, which helps them decide on their investments or other interests (Sanusi 2002).

Another aspect of corporate governance practices is evidenced in the manner in which the complaint management system, which is part of the redress mechanism expected to be in place in banks, is structured and managed. Generally, customers lodge complaints from time to time when they perceive that the bank has acted unethically or inappropriately in handling their affairs. This may arise because of excess charges, unauthorised deductions, dishonoured cheques, non-disclosure of terms and conditions, foreign remittance, automated teller machine (ATM) fraud and related matters (Obioha 2017).

Corporate reputation has been identified as one of the drivers of corporate governance. It is also acknowledged as having the potential to impact on customer loyalty towards the firm (Nguyen & Leblanc 2001). Corporate reputation is defined as a ‘collective assessment of a company’s ability to provide valued outcomes to a representative group of stakeholders’ (Fombrun, Galdberg & Sever 2000).

According to Taleghani et al. (2011), banking represents a service provider category with high interaction opportunities and customer–service provider employee contact. The customer is described as central to all marketing activities of banks the world over. Thus, customer loyalty and retention play a key role in the success of retail banks. Bank loyalty is defined as the:

Loyalty can also be described as a strongly held commitment to a bank in a manner that the customer desires to patronise and buy products consistently without resorting to switching factors and marketing appeals (Oliver 1999). Studies have highlighted three reasons why loyalty is a particularly important issue for service providers like banks. Firstly, loyalty is greater or more prevalent amongst services consumers than goods consumers. Secondly, services provide more opportunities for person-to-person interactions. This, in turn, often provides opportunities for loyalty to develop and perceived risk is often greater when purchasing services than goods (Parasuraman, Zeithaml & Berry 1985; Suprenant & Solomon 1987; Zeithaml 1981).

Therefore, it becomes necessary for bank management to carefully consider the factors that might increase customer loyalty and retention rates. Ghazizadeh et al. (2010) assert that, to achieve this goal, banks have to adopt a market-orientation approach that identifies consumer needs, designs new products or services and redesigns the old ones. Gummesson (1998) notes that banks have undertaken several initiatives, such as value chain analysis, customer satisfaction and loyalty programmes. Also included in these factors is minimising switching behaviour, perceived service quality, satisfaction, trust and switching costs amongst others (Afsar et al. 2010; Petruzzellis, Romanazzi & Gurrieri 2009). In the context of Nigeria, the occurrence of banking failures and the loss of confidence expressed by customers necessitate a move away from the usual drivers of customer loyalty and a consideration of corporate governance dimensions as potential drivers of enhancing bank–customer relationships.

The target population for this study were customers of the existing 22 retail banks located in Ibadan, one of the capital cities in Nigeria, as at the time of this study (2014–2016). Eight banks were purposively chosen from strata of categories of banks identified. In order to bring a more robust analysis, the selected banks were categorised into two distinct groups, namely banks by merger experience and bank by generation category. Three banks that have merger² experience within the past six years and five from those that have no merger³ or acquisition history within the same period were chosen. The bank generation categorisation was based on the length of time in the business of banking and the extent that modern technology is used in banking operations. A quota sampling technique was applied in a distributive manner to select 62 respondents from each bank. As the universe of the target population (bank customers) was not known, the sample from each bank was determined through convenience process. In this case, the parameters set by the researcher upon which participants were deemed selectable were considered. Thus, a total sample of 496 respondents affiliated to the eight banks were used for this study. The selection criteria for the sample were individuals who had a bank account maintained for at least 1 month, with the ability to read and write. At the end of the data collection, 424 questionnaires were found complete and usable for the analysis, which represented an 85% response rate. A total of 424 bank customers were conveniently selected, in which process respondents’ social categories were taken into consideration.

A survey (questionnaire) was administered to a sample of 424 bank customers selected from eight purposively chosen retail banks. A four-point Likert scale was used where respondents were asked to ‘Strongly Agree’, ‘Agree’, ‘Disagree’ or ‘Strongly Disagree’ with statements (items) referring to customers’ disposition to being loyal to their respective banks. Loyalty as a construct contained nine items that measure it. Similarly, the five corporate governance practices domains contain sets of indicators that measured each of them. The four-point Likert scale employed in the questionnaire was to avoid the tendency to swing to ambivalence by the respondents.

Data collected were analysed by the use of Statistical Package for Social Sciences (SPSS) tools. The reliability test produced a Cronbach’s alpha coefficient of 0.95 (pretest) and 0.87 (final analysis) for the customer loyalty construct. The final data analysis began with a reliability test of the constructs and items to ascertain the suitability of the final research instrument. Based on the acceptable benchmark value of average intra-class correlation of > 0.30 and Cronbach’s alpha ≥ 0.70 for scale reliability, the result of the test shows that all the five constructs (domains) have acceptable and high reliability with values ranging from 0.87 to 0.89. The order of their strength of reliability based on Cronbach’s alpha coefficient from the highest to the lowest was as follows: presence of bank management with integrity (0.89), transparency and adequate disclosure (0.87), corporate reputation (0.88), presence of competent bank management (0.87) and presence of customer complaint and dispute resolution system (0.89). Further analysis of data to capture the objectives of this study was performed by use of multiplicity of statistical tools, including partial least square structural equation model (PLS-SEM) Smart PLS (version 3).⁴ This was utilised to ascertain the influence of corporate governance practices domains on customer loyalty, with Pearson’s correlation moment analysis as a complementary tool to establish the association between corporate governance practices and customer loyalty.

Equation estimation formula (Partial least square structural equation model): η(m×1)=B(m×m)×ξ(n×1)+Γ(m×1)×ξ(n×1)+ζ(m×1)y(p×1)=Λy(p×m)×η(m×1)+ξ(p×1)X(q×1)=Λx(q×n)×ξ(n×1)+δ(q×1) where η is the latent endogenous construct associated with observed y indicators, B is the regression of one endogenous construct on another endogenous construct, ξ is the latent exogenous construct associated with observed x indicators, ζ is the error term associated with formative construct, ε is the error term associated with y indicators and δ is the error term associated with x indicators.

Equation estimation formula (Pearson’s correlation moment, r): r=n(∑ xy)−(∑ x)−(∑ y)√[ n(∑ x2)−(∑ x2) ][ n(∑ y2)−(∑ y2) ] where n is the number of pairs of scores, ∑xy is the sum of the products of paired scores, ∑x is the sum of x scores, ∑y is the sum of y scores, ∑x² is the sum of squared x scores and ∑y² is the sum of squared y scores.

This study strictly adhered to all the ethical prescripts of the host institution, Tshwane University of Technology, and related human science ethical issues by subjecting this research to the entire process. The designated departmental, faculty and university central committees approved the research; having met the ethical requirements a Clearance Certificate (Ref #: FREC2016/FR/04/001-MS) was issued to the researcher by the Faculty of Management Sciences Research Ethics Committee of the university. The participants’ informed consent was received and their confidentiality and privacy observed by the researcher.

In this study, the researcher sought to establish the effects of various corporate governance practices on customer loyalty to banks. This is with the aim to determine the strength of the effects of each corporate governance practice (domain), as individual components and the relative strengths comparatively. The major questions in this regard were: What is the strength of the effect of each of the corporate governance domains on customer loyalty? What is the relative strength of each of the domains? Which corporate governance domain is likely to have the highest influence on customer loyalty in the banks?

Table 1 shows the estimates of model fitness and quality indices of the constructs and indicators included in the test of the interaction between customer loyalty and corporate governance practices, as generated by PLS-SEM version 3. This interactive model met the standard requirement by the PLS-SEM version 3 for evaluating the reflective measurement model for the latent construct, customer loyalty (Custloyl).

The average variance extracted (AVE) of 0.498 (T = 14.200, p = 0.00), approximately 0.50, indicated is acceptable. The AVE is a metric to assess the communality of a construct. An AVE estimate of 0.50 or higher indicates that, on average, the construct, customer loyalty (Custloyl), explains more than half of the variance of its indicators’ convergent validity (Hair et al. 2017), which is the acceptable standard. Values below 0.50 are unacceptable. Also, the R-squared value of the endogenous latent variable, customer loyalty (Custloyl) 0.531 (T = 15.618, p = 0.00), establishes convergent validity. This indicates that the five constructs, namely presence of management with integrity (Preintegr), presence of competent bank management (Precompt), transparency (Transpar), presence of a complaints management system (Precompl) and corporate reputation (Corprepu), explain 53.1% of the variance of the endogenous variable, namely customer loyalty (Custloyl). Lastly, the Cronbach’s alpha (0.865 for customer loyalty) was above the 0.70 upper higher acceptable threshold, establishing construct reliability and validity.

The internal consistency reliability of the indicator variables for the customer loyalty construct (Custloyl), measured by Cronbach’s alpha, ranges between 0.79 to 0.87, which is regarded as satisfactory in social science research (Hair et al. 2017). Also, the internal consistency of the indicator (0.895) is established, because this value was above the threshold of 0.70, considered the standard (Hair et al. 2017). The outer loadings for each of the indicators on the customer loyalty construct indicate values ranging from 0.40 to 0.85. This was considered sufficient, because they contribute to the convergent validity of the endogenous construct, customer loyalty (Custloyl) (Hair et al. 2017:114). In establishing convergent validity, indicators with very low outer loadings (below 0.40) are considered weak and are therefore recommended to be eliminated from the construct (Bagozzi & Philipps 1991; Hair et al. 2011). Considering the strength of various corporate governance domains’ effects on customer loyalty, Figure 1 and specifically Table 2 show the total effect the domains had on customer loyalty.

The result of this analysis shows that presence of bank management with integrity had a significant positive effect (β = 0.220) and T-value (3.557) at p = 0.00 significant level on customer loyalty.

Considering the association and covariance between customer loyalty and bank management with integrity attributes, a positive relationship at significant level of p = 0.00 was found (Table 3). This result further confirmed the importance of these attributes in increasing customer loyalty to their banks. Bank management and staff’s level of assistance to customers was the most important attribute in this regard, while the attribute of bank management saying what they mean (being straightforward) in their transactions with customers had the least association in comparison with other attributes. As management integrity attributes increased collectively or individually, so did customer loyalty in the same direction.

The presence of competent management was found to have a positive total effect of β = 0.228 at a significant level of p = 0.00 on customer loyalty.

Further testing (Table 4) of relationship between customer loyalty and all the attributes of competent bank management revealed significant positive relationship between them. Responding to customer banking needs was found to be highly associated with customer loyalty, followed by promotion of excellent banking services. On the other hand, experience or knowledge in banking by the management team and clear vision in banking were the lowest ranking attributes of competency in banking associated with customer loyalty.

In the present study, the transparency or disclosure domain of corporate governance practices was found to negatively affect bank customers’ loyalty. The total effect (β = -0.005, p = 0.93) means that the effect of transparency or disclosure was very little in terms of what it contributes to customer loyalty.

The cross-loading result (Table 5) of the covariance between customer loyalty and the attributes of transparency or disclosure domain confirmed that customers’ accessibility to bank officials had the highest strength of correlation or association (r = 0.432) with customer loyalty, compared to other attributes. Banks’ provision of information on their contribution to the community had the weakest link to customer loyalty (r = 0.101).

The extent to which a customer complaint management system influenced or affected customer loyalty was explained in this study. The study found that a customer complaints management system in banks had a positive and significant effect on customer loyalty (β = 0.191, p = 0.00).

Further analysis of correlation of covariance (Table 6) between attributes of the complaint management system and customer loyalty confirmed apology for complaints by bank management and officials as the highest attribute of complaint management, while fast resolution of complaints was found to be least associated with customer loyalty to banks. However, all the attributes of the customer complaint system were positively associated with customer loyalty at a significant level.

This study tested and found that corporate reputation garnered by banks over time positively affected customer loyalty to the bank (β = 0.186, p = 0.00). As such, customers’ loyalty to their bank increased as customers recognise or are aware of bank officials and management’s reputation.

The study further tested the correlation between corporate reputation attributes and customer loyalty. The results (Table 7) confirmed that while all the attributes of corporate reputation are positively and significantly associated with customer loyalty to banks, banks’ adaptability to positive changes and financial stability were found to have the strongest and least link with it, respectively. This means that banks’ adaptability to positive changes in the industry increases customers’ loyalty to their banks.

In estimating the correlation between customer loyalty and the corporate governance practices (domains), Table 8 revealed that while all corporate governance practices (domains) had a positive association with customer loyalty at a very significant level, presence of competent bank management has the highest association with customer loyalty (r = 0.612, p = 0.00). This is followed by corporate reputation and customer loyalty (r = 0.592, p = 0.00) and the presence of bank management with integrity with customer loyalty (r = 0.59, p = 0.00). Transparency and disclosure was found to be least associated with customer loyalty (r = 0.403, p = 0.00) when compared with other corporate governance practices (domains). This result implies that customer loyalty increases as each of the corporate governance practices (domains) increases, especially with the presence of competent bank management.