Relationship marketing can be viewed as a business approach for establishing and managing the relationships between an organisation and its customer base. It encompasses all the functions of marketing with the purpose of establishing, developing and maintaining relationships with a partner at a profit through mutual exchange and delivery on promises (Egan 2011:266; Harwood, Garry & Broderick 2008:17).

Higher levels of competitiveness in the market place, increased marketing costs, and shortened technology life cycles necessitate a greater focus on the building of long term relationships with customers (Gilaninia et al. 2011:787). Theron, Terblanche and Boshoff (2012:32) support this argument by proposing that a renewed focus on relationship marketing over the last three decades has provided service organisations with an opportunity to secure a competitive advantage through relationship building initiatives. They argue further that role players in the financial sector have been progressive in their implementation of relationship strategies to strengthen their financial positions.

Organisations use a relational approach to establish long-term relationships with customers and such an approach should be to the benefit of all parties involved (Rootman, Tait & Sharp 2011:187). Venetis and Ghauri (2004:1579) confirm that managing long-term relationships is beneficial to all role players involved. From an organisational perspective, the benefits that accrue from such a relationship include a reduction in pricing competition, increased income generation from a customer base in the long term, and the establishment of barriers against competitors (Omar & Moh 2010:25). It is important to understand that a relational approach towards customers entails more than simply recruiting new customers (Ebersohn & Theron 2014:3). It also encompasses activities that focus on the management of the customer to enhance customer loyalty. Customers will remain loyal to an organisation if a relationship is established and professionally managed between them and the organisation (Omar & Moh 2010:25–26). Customer loyalty as such can be viewed as an obligation, based on emotion, to continue purchasing a specific brand of product or service without considering circumstantial influence or promotional efforts to secure switching actions (Rai & Medha 2013:140).

Literature theorises that factors such as trust, commitment, customer satisfaction, and supplier image are important when assessing customer loyalty (Van Vuuren 2011:18).

The marketing literature provides encompassing support for the inclusion of trust in building relationships. Trust exists ‘… when one party has confidence in an exchange partner’s reliability and integrity’ (Morgan & Hunt 1994:23). Jumaev, Kumar and Hanaysha (2012:41) further argue that trust is a key virtue in relationship building, as it is the basic foundation of a relational approach to ensure that both parties are fully committed to the relationship. Parties to a relationship desire ‘… predictable and obligatory behaviour’ from each other to such an extent that a high level of certainty is linked to future benefits (Negi & Ketema 2010:115). Parties are also more willing to commit themselves to such a relationship if it is developed from a sound foundation of trust (Morgan & Hunt 1994:24). Ultimately, trust can be viewed as a precursor to the establishment of customer loyalty and a key element in both the establishment and the management of long-term relationships with customers, supported by actions to deliver on promises made (Rauyruen & Miller 2007:24). Theron et al. (2012:33) further propose that the ‘… importance of trust has been confirmed in a variety of sub-industries, such as retail banking (Ndubisi, Wah & Ndubisi 2007), merchant banking (Liang & Wang 2007), financial planning (Johnson & Grayson 2005), electronic banking (Soureli, Lewis & Karantinou 2008) and asset management (Nijssen & van Herk 2009)’.

The marketing literature secondly emphasises the role of commitment when a long-term relationship is established. For example, Hunt, Arnett and Madhavaram (2006:78) and Hennig-Thurau, Gwinner and Gremler (2002:232) state that both trust and commitment are required to ensure a partnership approach towards relationship building. Commitment is defined as:

Marketing theorists such as Amin et al. (2011:789) further refer to commitment as an interest by parties to a relationship to remain with a specific course of action out of their own free will. Commitment implies that partners would rather consider the benefits that accrue from establishing a long-term relationship than the short-term benefits available (Read 2009:28). Therefore, commitment relates to an implied or unequivocal promise of continuation between parties to a relationship (Dwyer, Schurr & Oh 1987:13; Nyadzayo 2010:189). Commitment is a central element in the relationship building process and is a strong foundation for securing a long-term relationship building approach (Rauyruen & Miller 2007:3; Van Vuuren 2011:80). Morgan and Hunt (1994:5–26) note that commitment must be established within a trustworthy environment. They state that ‘… relationships characterised by trust are so highly valued that parties will desire to commit themselves to such relationships. Indeed, because relationship commitment entails vulnerability, parties will seek only trustworthy partners’. This argument is supported by Theron, Terblanche and Boshoff (2010b:998–999), who state that commitment can be secured only if a trustworthy environment is created through which parties to a relationship can commit to such a relationship.

Customer satisfaction also seems to be a key factor in marketing relationships (Amoako 2012:136; Balas & Tareef 2011:67; Ha & Lee 2010:1163; Theron et al. 2012:34).

Halimi, Chavosh and Choshali (2011:51) argue that organisations continuously focus on the satisfaction of customer needs as a precursor for securing their retention. In the current business environment, customer satisfaction has been identified as a priority concern for organisational management in securing future business growth. Therefore, if an organisation wants to survive in an increasingly competitive business environment, the customer must be provided with products and services of superior quality to enhance customer satisfaction levels. This can result in positive word of mouth referrals by the customer (Hu, Kandampully & Juhaweer 2009:115). Swar (2012:26) further holds the opinion that the process of service quality management through the continuous assessment of customer satisfaction levels has become a critical element in securing the success of services marketing compared to the marketing of products.

The core focus of services marketing is on satisfying both the needs and the wants of customers. This implies that customer satisfaction has become the pivotal point of all marketing activities (Hu et al. 2009:115). Luo and Homburg (2007:133) support this view and identify customer satisfaction as a key antecedent of customer loyalty. Ndubisi, Malhotra and Chan (2009:8) note that through empirical research it has been scientifically proven that the satisfaction of a customer’s needs and wants has a positive influence on the consumer’s decision to remain loyal to the relationship. When customers are dissatisfied they are both willing to exit from the relationship and to criticise the business.

Van Vuuren, Roberts-Lombard and Van Tonder (2012:98) hold the view that the concept of ‘supplier image’ has not been academically explored in the same manner as concepts such as trust, commitment and customer satisfaction. It is argued that supplier image is a central element in following a long-term relational approach towards a customer. Wang (2010:255) describes supplier image as a total impression made in the mind of an individual by an organisation, whether visually, through communication or in print. Van Vuuren (2011:75) concurs, arguing that supplier image refers to the customer’s perception, developed over the long term, of the sum of all experiences from previous interactions with an organisation.

Hoq, Sulatana and Amin (2010:72–73) propose that customer loyalty is influenced by image and that the association between image and loyalty is of a positive nature, and a result of customers’ evaluations of the perceived quality of service and product received from the business. They argue that if the experience is of a positive nature, the resulting perception of the organisation will also be positive. Da Silva and Alwi (2008:125) agree and argue that the image communicated by a supplier is a central prerequisite for securing customer loyalty. According to them, the development of loyalty is influenced by the customer’s perception (positive or negative) of the image formed by the supplier. Favourable supplier image is important, not only to improve loyalty but also to promote positive word of mouth. Customers need to be aware of a supplier, its products and its services for the correct reasons. Therefore, the management of supplier image is of critical importance as the outcome can be positive or negative if not managed professionally (Friedman, Brown & Taran 2007:375–376). Supplier image is therefore a key element that influences customer loyalty, and a positive image can encourage customers to purchase more. The loyalty of current customers is also dependent on whether an image is maintained and enhanced (Hu et al. 2009:117).

Supplier image further has both direct and indirect effects on customer loyalty (Hawabhay, Abratt & Peters 2009:4). An individual’s decision-making process is characterised by weighing the costs and benefits of a specific decision. An implication is that as perceived switching costs increase, the costs will offset the perceived benefits of switching to an organisation with a better image. Therefore, despite an increase in switching costs, an unhappy customer may decide to remain loyal to an organisation even though its image has been influenced negatively (Kuo & Ye 2009:752–753). Hence, there will be an insignificant relationship between supplier image and customer loyalty in cases of high switching costs. In comparison, in the case of low switching costs, customers with a negative perception of an organisation’s image will have greater motivation to leave the organisation, which strengthens the relationship between supplier image and customer loyalty (Wang 2010:255).

Struwig, Roberts and Gordon (2013:viii) state that 25% of South Africans in 2012 indicated that they assumed sole personal responsibility for daily money management in the household in which they reside. It was furthermore determined in the same study that:

In terms of financial planning it was established that mixed race and black South Africans ranked the lowest in terms of financial knowledge, saving and planning for retirement. The white and Indian population in South Africa ranked the highest for this category. It is therefore argued that after 20 years of democracy, education on issues of financial planning is still determined by socio-economic standing, influencing access to education which imparts changing behaviour (Struwig et al. 2013:91, 98). Joffe (2010:4–5) states that the financial service delivery process should be guided by the principles of trust, created through product knowledge, empathy and the service scape itself. Addressing these aspects fully in the service delivery process will lead to the commitment and the eventual loyalty of the customer. Joffe further argues that this approach should be generic to all customer levels, irrespective of the customer portfolio.

South Africa has experienced rapid urbanisation since 2000, similar to most other emerging economies in the world (Zuma 2013:3). A total of 75% of South Africa’s population currently resides in the urban regions of the country. The number of South Africans living in urban or semi-urban parts of the country has increased from 52% in 1990 to 62% in 2011 (South Africa Info 2013). The Gauteng province of South Africa is the smallest province and represents only 1.4% of the total geographical land area of the country. The province is primarily urban with a total number of 12 million citizens within its borders (South Africa Info 2015a). The North-West province is South Africa’s 4th-smallest province, consisting of 8.7% of South Africa’s land area with a population of 3.5 million people. The province is characterised by both semi-urban and rural areas, with Potchefstroom, Klerksdorp and Mafikeng the major semi-urban areas (South Africa Info 2015b).

A Financial Services Board (FSB) survey conducted in 2012 illustrated that aspects such as race, level of education, income level and the geographical area of residence play a central role in South African’s financial literacy levels. The report specified that differences in financial literacy are evident amongst black South Africans relative to other population groups, with the former having lower levels of financial literacy. Disparity in terms of financial knowledge is also evident between rural, semi-urban and urban populations and South Africans residing in provinces such as the Eastern Cape, Limpopo and North-West have lower levels of financial control compared to individuals living in Gauteng or the Western Cape province (Roberts et al. 2012:7; Struwig et al. 2013:xvi, 41, 97). In conclusion, South Africans living in provinces with a higher level of economic employment, educational literacy and urbanisation record (e.g. Gauteng, Western Cape and Kwazulu-Natal) illustrate a higher level of financial literacy compared to provinces such as the Eastern Cape, North-West and Limpopo (Struwig et al. 2013:35).

Accordingly, following the literature investigation, it is hypothesised in this article that:

The demographic profile of the respondents who participated in the study is summarised in Table 1.

It is evident from Table 1 that approximately a third (34.0%) of the total number of respondents who participated in the study have been with their financial adviser’s practice between two and five years. From a closer perspective, it seems that the two to five years category also received the most responses in the North-West and Gauteng provinces. Table 1 further indicates that most of the respondents who participated in the survey in the North-West province were females (51.8%), whilst the majority of the respondents who completed the questionnaire in the Gauteng province were males (53.7%).

In both provinces, the age group 36–47 years was selected most. As the majority of the respondents who participated in the survey were older than 36 years, they were in a position to make a valuable contribution to the study, as individuals at this mature level generally tend to have clearly manifested perceptions and opinions about the industry and the products it offers.

Finally, most of the respondents in the North-West province are from the African population group (86.3%), whilst the majority of the respondents from the Gauteng province are from the white racial group (71.5%). Statistical analysis was conducted to determine if there would be a significant difference between the perceptions of the various racial groups. The results are discussed in the last part of the results section.

Table 2 provides an indication of the types of products that the respondents bought over the last 10 years. A 10-year period was selected as financial services products are not purchased on a frequent basis. The respondents were allowed to select more than one option, and generally it seems that the respondents who participated in the survey had been exposed to a variety of financial products. Overall though, a life insurance policy was the product bought most by the respondents (55.8%), followed by a retirement annuity policy (33.7%). The same pattern could be detected in each of the two provinces, but it seems that the life insurance policy category was selected more often in the Gauteng province (68.3%) than in the North-West province (44.4%).

Table 3 portrays the Cronbach alphas that were calculated to determine the reliability of the scales used. The values per province as well as overall are all above 0.7, which, according to Pallant (2010:97), can be deemed the minimum level of acceptable scale reliability. It can therefore be concluded that all five scales used in the survey were in fact reliable.

Standard multiple regression analysis was used to ascertain whether the selected variables can be viewed as predictors of customer loyalty towards independent financial advisers in the Gauteng and North-West provinces. The Pearson product-moment correlation coefficients were determined next, to explore the associations between the independent variables of trust, commitment, customer satisfaction, and supplier image and the dependent variable customer loyalty. It became evident from the results that overall all the constructs were significantly correlated at the p < 0.05 level. The same pattern could be detected at provincial level. Considering Pallant’s (2010:134) guidelines, it seems that the level of correlation of all the variables investigated was mostly medium to high, but not equal to or more than 0.9, which would have demonstrated multicollinearity.

In the third step of the analysis, the researchers explored which of the independent variables included in the model contributed to the prediction of the dependent variable customer loyalty. The results are displayed in Table 4.

The study relied on a 95% level of confidence in the interpretation of the results. Considering the findings from both provinces, the total variance explained by the model as a whole was 75.5%, F (5, 242) = 149.3, p < 0.05. The total variance explained by the model in Gauteng was somewhat higher, 80.8%, F (4,118) = 124.188, p < 0.05 and lower in the North-West province, 67.8%, F (4,122) = 64.291, p < 0.05.

It further appeared that in relation to the total sample, the variables commitment, customer satisfaction and supplier image have p-values of less than 0.05 and can consequently be regarded as predictors of customer loyalty. Within each of the two specific provinces a different scenario is, however, presented. In the Gauteng province only trust and commitment can be viewed as predictors of customer loyalty, as only their p-values were smaller than 0.05. It further seems that when considering beta, the variance in customer loyalty explained by commitment (beta = 0.473, p < 0.05) is higher than the variance in customer loyalty that is accounted for by trust (beta = 0.249, p < 0.05). Within the North-West province, only the variable commitment seems to have a significant positive relationship with customer loyalty (p = 0.0001). When considering the beta value, one unit increase in commitment would result in a 0.594 increase in customer loyalty.

Consequently, following the above discussion, it is possible to conclude that HA¹ should be rejected. Only trust and commitment can be viewed as predictors of customer loyalty in the Gauteng province and only commitment has a significant positive relationship with customer loyalty in the North-West province. Similarly, HA² should then also be rejected, as not all variables appeared to have an equally significant positive relationship with customer loyalty.

Table 5 lists the mean values that were obtained for the four independent variables and the dependent variable customer loyalty. Considering the fact that a 5-point Likert scale was used to measure the various constructs, it appears that the respondents in both provinces were reluctant to agree with the statements formulated about their financial service providers. However, it also seems that the respondents in Gauteng had a slightly more positive perception of their financial advisers, given the fact that their mean scores (for all variables investigated) were higher than the mean scores for the respondents in the North-West province. Independent sample t-tests were then performed to verify whether the perceptions of the respondents in Gauteng were truly different from those of the respondents in North-West. A preliminary investigation was conducted first and confirmed that the data set complied with all the underlying assumptions for conducting the independent sample t-test.

The results of the independent sample t-tests then revealed that there are statistically significant differences between the perceptions of the respondents in the two different provinces with regard to all the variables investigated. More particularly, statistically significant decreases were detected in the following scores: trust (from Gauteng: M = 3.90, SD = 0.78 to North-West: M = 3.53, SD = 0.72; t(254) = 3.93, p < 0.05), commitment (from Gauteng: M = 3.71, SD = 0.91 to North-West: M = 3.44, SD = 0.80; t(253) = 2.46, p < 0.05), customer satisfaction (from Gauteng: M = 3.82, SD = 0.82 to North-West: M = 3.57, SD = 0.77; t(255) = 2.50, p < 0.05), supplier image (from Gauteng: M = 4.19, SD = 0.67 to North-West: M = 3.79, SD = 0.65; t(253) = 4.92, p < 0.05) and customer loyalty (from Gauteng: M = 3.70, SD = 1.08 to North-West: M = 3.39, SD = 0.90; t(250) = 2.46, p < 0.05). The mean decrease in scores displayed in Table 5 further shows that the difference in perceptions between the Gauteng and North-West respondents is largest with regard to the variable customer satisfaction.

The effect sizes calculated, using eta-squared, for all the variables listed in Table 5 are as follows: trust (0.057), commitment (0.023), customer satisfaction (0.024), supplier image (0.087) and customer loyalty (0.024). Considering these results and the guidelines provided by Pallant (2010:243), the actual difference in the mean scores between the groups can be regarded as relatively small with regard to the variables commitment, customer satisfaction and customer loyalty and moderate with reference to the variables trust and supplier image.

Following the above mentioned findings, it is possible to accept HA³.

ANOVA was used to investigate significant differences between the perceptions of the various racial groups. No further preliminary investigations were required because, as confirmed in the previous section, the study adhered to all the underlying assumptions for comparing groups. The findings are displayed in Table 6.

From Table 6 it is evident that there are statistically significant differences at the p < 0.05 level between the perceptions of the different racial groups with regard to the variables trust, commitment and supplier image. Because some of the racial groups were relatively small, the researchers decided to also apply the Robust Test of Equality of Means in order to verify the differences in perceptions that were detected. The findings are displayed in the last two columns of Table 6 and confirm that at the p < 0.05 level there are significant differences between the perceptions of the various racial groups with regard to the variables trust, commitment and supplier image.

Hochberg’s test was performed next in order to determine the exact location of the differences in perceptions that were found. The test was deemed appropriate given the fact that the data obtained for these factors passed Levene’s test for homogeneity of variance and the number of respondents interviewed from each racial group were not equal. The test confirmed that there were significant differences in perceptions between only the white and African population groups with regard to the variables trust (p = 0.004), commitment (p = 0.039) and supplier image (p = 0.0001). The effect sizes calculated, using eta-squared for each of the variables, were respectively trust (0.029), commitment (0.024) and supplier image (0.052). Considering these results and the guidelines provided by Pallant (2010:254), the actual difference in the mean scores between the white and African racial groups can be regarded as relatively small.

Accordingly, following the above discussion, it can be concluded that HA⁴ can only be partially accepted. Significant differences were found between only the white and African racial groups, with respect to the variables trust, commitment and supplier image.

Firstly, our research results suggest that the Commitment-Trust theory of Morgan and Hunt (1994) can no longer be perceived as the base of departure for all supplier-customer relationships in a business-to-consumer (B2C) environment. These two authors argued that both trust and commitment are two core factors of relationship marketing because they encourage marketers to remain invested within a relationship. However, our research findings challenge the seminal work of Morgan and Hunt by establishing that in South Africa, the extent to which trust and commitment predicts relationship loyalty is specific to both industry and geographical location. In the Gauteng province, it was established that both trust and commitment can be perceived as predictors of customer loyalty. However, in the North-West province it was determined that loyalty could result from commitment alone. In a semi-urban environment, trust is therefore not perceived as a prerequisite to secure customer loyalty but is relevant in establishing a relational approach to customer management. This finding is supported by Theron et al. (2012:31), who state that ‘… it is possible to build commitment without the presence of trust’. Theron et al. (2010b:1005) further established that financial services decision makers do not necessarily perceive trust as a factor that influences commitment in the relationship-building process. They established that both parties to a relationship (namely the financial service provider and the client) should be motivated to make an investment of time and other resources to secure the continuation of the relationship. The finding is valuable as it provides researchers with an initial understanding of what can be viewed as predictors of customer loyalty from both an urban and a semi-urban perspective.

Secondly, this study provides further understanding of the relationship between trust, commitment and customer loyalty from both an urban and a semi-urban perspective. Theoretical arguments on the variables required to strengthen customer loyalty are provided from a financial services environment perspective. The theoretical approach followed in the article guides independent financial advisers on relational success. It stipulates that to succeed in an evolving market environment, it would be imperative for advisers to build good relationships with their clients and ensure that they remain loyal to them over the long term. Such a relational focus should, however, be adjusted for either an urban or a semi-urban context, depending on the geographical location in which the relationship is built. Therefore, by understanding the factors influencing customer loyalty (trust, commitment, customer satisfaction and supplier image), the relationship marketing approach of the independent financial adviser could be aimed at improving customer loyalty initiatives, ultimately increasing the use of financial planning services.

The survey conducted for this article provided valuable information regarding the factors of relationship marketing that do predict customer loyalty in the independent financial adviser environment of both the Gauteng and North-West provinces. The empirical findings should, however, be considered within the context of some limitations to the study. Firstly, the study applied convenience sampling and interviewed respondents from only the Gauteng and North-West provinces. The results can therefore not be generalised to South Africa as a whole. A future research study would need to be conducted on a broader scale and should include a representative proportion of all the provinces of South Africa that is inclusive of both urban and semi-urban areas. Secondly, a mixed-method approach that includes both quantitative and qualitative research techniques could be followed to provide more insight into the respondents’ perceptions of their financial advisers, and understanding of the true motivation behind the clients’ desire to remain loyal towards their financial advisers.