Summary of Doctoral dissertation: Some factors affecting brand loyalty of individual saving depositors in banking

24 1

CHAPTER 1: GENERAL INTRODUCTION OF THE RESEARCH 1.1. Background households and micro-enterprises (having less than 10 people) are also classified as individual customers due to many similarities in the decision- making process; therefore, this may be an extensive research direction for the topic.

Individual customers, in addition to using savings, also use many other products such as loan, money transfer, bank insurance, guarantee, Internet banking, etc.; therefore, it is possible to study their loyalty to the bank. In particular, studying users of online savings deposit, to compare against the loyalty of depositors over the counter is also an interesting research direction, especially in the current context where Internet banking services are developing and becoming so popular. Personal savings deposits play a significant role in the stability and development of the national economy. According to statistics of the State Bank of Vietnam (SBV) at the end of 2016, the total deposits of the economy reached 3,489,426 billion VND, accounting for 49% of total payment instruments. In order to control the increase in interest rates by commercial banks to attract deposits from individual customers, in August 2011, the SBV stipulated the interest rate ceiling on savings deposits. The interest rate ceiling on savings deposits has been maintained at 5.5%/year for a 6-month term since June 2013.

Changes in the financial and banking market in Vietnam as well as the fierce competition among commercial banks have created more and more opportunities for individuals savings depositors, and maintaining deposits at the same bank or moving to another bank is affected by various factors. In this context, brand loyalty of individuals savings depositors at commercial banks has attracted great attention of many scientists and administrators. 1.2. Rationale

Individual savings depositors are directly influenced by emotional choice (Kalneman Daniel, 1986), which refers to the profitability of savings deposits in comparison with other savings and investment channels. Brand loyalty of individual savings depositors is an important factor in their behaviors of depositing savings. Loyalty is a stable commitment to repurchase products or services from the same brand, resulting in repetition or repurchase of that brand even though impacts from enterprises’ situations and marketing activities can lead to change in behavioral loyalty (Oliver, 1999).

Brand loyalty has been studied by many authors in terms of behavioral chains of customers using products and services of banks (JW Prinsloo, 2000; Tanvi Singhal, 2015; Kasri and Kassim, 2009; Durkin et al, 2003; Cengiz Erol, 1989; Gerrard and Cunningham, 1999).

In the world there have been a lot of research relating to factors which afftect customer loyalty in the field of banking and finance. The factors tend to include those relating to customers, those relating to banks and those relating to society. Some other researchers tend to focus on factors which have direct effect on brand loyalty. According to this direction, they often use multiple regression models for analysis. It shows that brand loyalty is affected by various factors and the number of factors studied vary according to market characteristics.

2 23 Commercial banks should strengthen commitment from relationships for individual savings depositors as follows: - Retaining employees with salary and bonuses through benefits and insurance programs for bank staff and relatives. - Implementing internal communication activities to show

Some researches believe that brand loyalty of individual customers is affected by both direct and indirect factors through intermediary variants .Authors following the second direction study factors affecting brand loyalty through intermediary variables. This group of authors argue that customer loyalty is a complicated factor, and that, according to behavioral psychology, the way to loyalty must go through many complicated steps such as customer satisfaction. This research direction often involves analyzing data according to SEM (Structural Equation Modelling). the connection, love and pride for banks such as: team building, gratitude to employees, commending and rewarding employees making good achievements, etc. - Training soft skills for staff to improve emotional bonds with savings depositors.

- Constructing a system of data of savings depositors (KYC - know your customer) to understand customers' information and help bank staff take care of and appreciate customers on special occasions. 5.4. Proposals and recommendations to the State Bank

The group of individual savings depositors is a very important group for commercial banks since they are the ones who have money and can pro- actively leave the banks at any time, which is different from the group of borrowing customers who are bound by collateral or that of customers with cash/cards bound by transactional accounts. Under the pressure of competition to attract personal deposits to ensure capital balance, commercial banks have a lot of policies and programs to entice individual customers to deposit savings; this is a segment of strong competition among commercial banks.

The thesis provides recommendations for state agencies to manage individual deposit mobilization of commercial banks to avoid unfair competition and risks and to bring benefits to customers. 5.5. Shortcomings and further research direction 5.5.1. Shortcomings

First: The author's research samples cover Hanoi city. Although the author has provided arguments and evidences to prove the representation of the urban part of Hanoi compared to the Vietnamese market, research results may be true for large and densely populated cities as well as for the high density of banks' transaction points in Hanoi, but may not be true for many other localities, especially in remote areas. It can be seen from the above mentioned analysis that there needs to be a research study to assess factors affecting brand loyalty of individual savings depositors at commercial banks, which will be an inheritance from previous studies while irrelevant factors will be eliminated and factors in line with the Vietnamese market supplemented. Aware of this importance, after thorough research and finding, the author has selected the topic "Some factors affecting brand loyalty of individual savings depositors in banking” for her doctoral thesis. 1.3. Objectives and questions of the research 1.3.1. Objectives of the research

Second: Individual customers are of great dispersion so some ideas of the questionnaire can cause difficulties or misunderstanding for some respondents. This more or less affects the accuracy of the answers. - Overall objectives: The thesis focuses on the main objective of studying factors affecting brand loyalty of individual savings depositors in commercial banks and a case study scoping the urban part of Hanoi.

Third: In theory, there are many models and schools of study on loyalty, behavioral loyalty and attitudinal loyalty. However, the thesis cannot study in detail and fully all those models and schools.

Fourth: In fact, there may be other factors that may affect customer loyalty to brands such as regional characteristics, culture, habits, etc., which have not been reviewed in detail in this research. 5.5.2. Further research direction - Specific objectives: (i) Interpreting the theoretical background of brand loyalty of individual customers in financial services; (ii) Assessing impacts of the factors on brand loyalty of Vietnamese individual customers to savings deposit services at commercial banks; (iii) Proposing some solutions to maintain, strengthen and develop brand loyalty of individual customers using savings deposit services at commercial banks in Vietnam. 1.3.2. Questions of the research - How is the brand loyalty of individuals customers depositing savings This research is conducted for individual savings depositors. However, according to modern and practical points of view, at commercial banks, at commercial banks?

3 - What factors affect brand loyalty of individual savings depositors to 22 5.2.4. Impacts of switch cost and commitment from relationships on the conversion from satisfaction to brand loyalty commercial banks? - To what extent do those factors affect brand loyalty of individual savings depositors at commercial banks? - What solutions can help maintain and strengthen brand loyalty of

individual savings depositors at commercial banks? 1.4. Subjects and scope of the research 1.4.1. Subjects of the research

- Subject of study: Theory and practice on impacts of some factors on brand loyalty of individual savings depositors at commercial banks, specifically customers in the urban part of Hanoi. Factors of switch cost and commitment from relationships have impacts on the conversion from satisfaction to behavioral loyalty and from satisfaction to attitudinal loyalty (P-Value < 0,05). Switch cost and commitment from relationships have more impacts on behavioral loyalty than on attitudinal loyalty. It can be understood that, under the influence of moderating variables, attitudinal loyalty can be more easily changed than behavioral loyalty. 5.3. Proposed solutions to commercial banks 5.3.1. Solutions to improve customer satisfaction - Objects of the study: Individual customers having deposited or

depositing savings at commercial banks in Vietnam. 1.4.2. Scope of the research

- Scope in terms of contents: Brand loyalty of individual savings depositors at commercial banks, to be studied according to characteristics of the Vietnamese market. - Solutions to improve satisfaction through overall service quality: (i) Improve the assurance of staff directly dealing with customers through professional training, skills training and training at work; (ii) Improve communication with customers through processing of complaints, events and interactive activities; (iii) Improve facilities outside and inside transaction points. (iv) Complete and simplify the system of forms and templates for individual savings depositors.

- Scope in terms of space: Hanoi – the urban part (urban districts: Hoan Kiem, Ba Dinh, Thanh Xuan, Hai Ba Trung, Hoang Mai, Cau Giay, Tay Ho, Long Bien, Dong Da, Nam Tu Liem and Bac Tu Liem) as it is highly representative of commercial banks in Vietnam in terms of scale, proportions of personal deposits and competitiveness of products and services. - Solutions to improve satisfaction through reputation: (i) Invest in advertising and communication activities in the short and long term; (ii) Improve the financial potential, network system and customer experience; (iii) Invest in core banking and technology infrastructure to bring convenience and ensure security for customers. - Scope in terms of time: The study was conducted from 2014 to 2018. 1.5. Approach of the research

- Analysis of secondary data: Consolidation and analysis of statistics; analysis and consolidation of theories; classification and systematization of theories; modelling; inductive and deductive analysis. - Solutions to improve satisfaction through social responsibility: (i) Implement social responsibility with communities and localities at local transaction points of commercial banks; (ii) Complete and diversify product and service package for individual savings depositors, and renovate procedures to fulfill responsibilities to customers. 5.3.2. Solutions to increase switch cost Commercial banks should increase switch cost to individual savings depositors as follows:

- Analysis of primary data: (i) Qualitative research: interview of experts and focus group interview. (ii) Quantitative research: conducted in two stages of preliminary research with samples of 200 customers and official research with samples of 650 customers. The thesis applies analysis techniques of Cronbach’s Alpha, Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), SEM and Hayes’ regression analysis (2013). 1.6. New contributions of the thesis

- Upselling and crossselling services apart from savings deposit. - Conducting loyal customer programs and promotional programs to increase benefits of Individual savings depositors such as: preferential interest rate, promotional gifts, lucky draw, preferential fees for other products and services, etc. 5.3.3. Solutions to strengthen commitment from customer relationships • New academic and theoretical contributions (i) Studying factors affecting brand loyalty of customers in banking, given the typical case of services of personal savings deposit, the thesis has

4 21

- Accepting hypotheses of H6, H7: Determining that switch cost and commitment from relationships have impacts on the conversion from satisfaction to brand loyalty.

supplemented and demonstrated positive impacts of the factor ‘commitment from relationships’. The relationships between customers and bank staff create commitments to depositing and maintaining deposits to a bank. Research results show that this factor can help better explain brand loyalty of customers in retail banking

The results of multi-group structural analysis by gender show that there is a difference between men and women in factors affecting customer satisfaction during personal savings deposit and the conversion from satisfaction to brand loyalty. 5.2. Discussion on research results 5.2.1. The impact of 5 factors in SERQUAL scale on the overall service quality (ii) The thesis has completed and developed a number of scales for research variables in accordance with the context of retail banking business in Vietnam, including brand loyalty (behavioral), tangible instruments, brand reputation and social responsibility. In particular, the author of the thesis has developed a new scale for the factor "commitment from relationships" based on some research in relationship marketing.

All SERQUAL dimensions have positive impacts on the overall service quality. Accordingly, the strongest impact is made by assurance (Beta = 0,273), followed by empathy (Beta = 0,263), tangibles (Beta = 0,261), responsiveness (Beta = 0,159) and finally reliability (B=0,083). This is the basis for commercial banks to select priorities in investment in service quality improvement. 5.2.2. Impacts of overall service quality, reputation, social responsibility on customer satisfaction impacts on

- Overall service quality has the most positive and strongest impacts on custome satisfaction (Beta = 0,576). Therefore, in order to improve satisfaction of individual savings depositors, overall service quality should be taken into consideration and given priority by commercial banks.

• New findings and recommendations from research and survey results (i) The thesis has identified the extent of impacts of the following factors: overall service quality, brand reputation and social responsibility on customer satisfaction, among which overall service quality has the most significant impact while satisfaction has significant impacts directly proportional to brand loyalty when no impacts of moderating variables have been available yet. The two moderating variables of switch costs and the commitment from relationships have proportional conversion from satisfaction to brand loyalty, including behavioral loyalty and attitudinal loyalty. With regard to the conversion from satisfaction to behavioral loyalty and attitudinal loyalty, impacts of each moderating variable/ simultaneous impacts of both moderating variables on the attitudinal loyalty are stronger than those on behavioral loyalty. Analyses from the thesis also show the difference between men and women in brand loyalty when it comes to depositing personal savings. - Reputation has positive impacts on satisfaction of Individual savings depositors (Beta = 0,412). Accordingly, customers pay attention to reputation based on good reputation, financial capacity, experience and service comparison when deciding to maintain savings deposits at a bank.

(ii) The thesis has recommended some solutions for commercial banks to improve brand loyalty, including one based on a new research finding, which is, in order to take advantage of commitments from customer relationships, banks should take into consideration relationships between customers and bank staff and enhance connections between staff and their banks. 1.7. Organization of the thesis - Activities related to social responsibility of commercial banks help improve satisfaction of individual savings depositors (Beta = 0,254). Therefore, the participation of commercial banks in many community activities will contribute to creating positive impressions on customers, making them more satisfied. Commercial banks also conduct customer – oriented activities such as making information transparent, cutting down on procedures, etc. 5.2.3. Impacts of satisfaction on brand loyalty The thesis consists of 5 following chapters: Introduction; Theoretical Background and research models; Methodology; Research results; Discussion and some proposals.

It can be seen from testing results of P-value < 0,001, Beta = 0,620 that customer satisfaction has strong and positive impacts on brand loyalty without influence of moderating variables. This is relevant with many previous studies that customer satisfaction will lead to more loyalty.

20 5

Table 4.13: Comparison of impacts of moderating variables on conversion to attitudinal loyalty and behavioral loyalty CHAPTER 2: THEORETICAL BACKGROUND AND RESEARCH MODELS

R2 change

Attitudinal loyalty

Behavioral loyalty

2.1. Bank brands and individual savings depositors 2.1.1. Bank brands

4,3% 4,8%

3,1% 3,8%

cost

and

5,1%

4,59%

Only moderation by switch cost Only moderation by relationships Moderation by both switch relationship

A number of scholars have given different definitions of brands; however, most definitions have followed two main perspectives: the first focuses on functions and owners of brands; and the second is based on impacts of brands on consumers.

Table 4.13 shows that impacts of the only moderating variable of switch cost/ relationships or impacts of both variables on the conversion from satisfaction to attitudinal loyalty are stronger than on the conversion from satisfaction to behavioral loyalty. 4.2.3.4. Multi-group structural analysis Based on the first perspective, the American Marketing Association (AMA) defines a brand as a ‘name, term, sign, symbol or design, or a combination of them intended to identify the goods and services of one seller or group of sellers and to differentiate them from those of other sellers’. Based on the second perspective, Feldwick (1996) defined a brand as ‘a collection of perceptions in the mind of the consumer’.

The testing results regarding differences in compatibility indicators between the variable and partially invariant models show that differences between two models are statistically significant (p = 0.00 <0.05). Therefore, the variable model is selected and enables us to conclude that there are differences between male and female customers in assessing brand loyalty of customers.

Individual savings depositors As the banking industry possesses characteristics of services that do not allow for packaging, labelling and displaying, the brand value must be inferred by the consumer (Cobb and Ruble, 1991). Banking products are intangible and customer loyalty to a bank brand is the customer loyalty to products provided by that bank. In other words, for banks, product loyalty and brand loyalty are homogenous. 2.2. 2.2.1. Definition

Table 4.14: Differences in compatibility indicators between variable and partially invariant models by customer gender P 0,000 0,000 0,000

Compared model Partially invariant Variable Difference value

χ2 1187,568 1086,048 101,520

RMSEA 0,058 0,058 0,000

TLI 0,919 0,917 0,002

IFI 0,910 0,908 0,002

CFI 0,908 0,907 0,001

df 437 432 5

Individual savings depositors are individuals or family households having transactions of personal savings deposits at commercial banks or credit institutions and the thesis refers to individual savings depositors at commercial banks. Specifically: χ2: chi-square; df: number of degrees of freedom; p: level of significance CHAPTER 5: DISCUSSION AND SOME PROPOSALS 5.1. Summary of research results

Hypothesis testing results are as follows: - Accepting hypotheses of H1a, H1b, H1c, H1d, H1e: Determining that tangibles, assurance, responsiveness, empathy, reliability have positive impacts on overall service quality. Individual customers’ savings deposits are their cash amounts deposited to savings accounts and determined on savings books/ cards issued by the bank, entitled to interest rates as prescribed by organizations receiving the savings deposits and SBV. Individual customers’ savings deposits are basically classified according to depositing time or purposes, including: non-term deposit, fixed- term deposit and special deposit. 2.2.2. Behavioral characteristics of individual savings depositors

- Accepting hypotheses of H2, H3, H4: Determining that overall service quality, reputation, social responsibility have positive impacts on customer satisfaction. - Accepting hypothesis of H5: Determining that customer satisfaction Behaviors of individual savings depositors are classified into behaviors before deposit, during deposit, after deposit and in the future. Repeated depositing behavior or loyalty is one of the most important behaviors in the behavior chain. This is a complicated development process in customers’ behavioral psychology before they make a decision. Apart from factors of has positive impacts on brand loyalty.

6 19 loyalty

personal psychology, customer is also affected by many complicated factors that require further research in order to help commercial banks to improve the effectiveness of their capital mobilization. 2.2.3. Research studies on behaviors of individual savings depositors commitment from relationships accounts for 4,8%, conversion to attitudinal loyalty, F(2;593)=3,7937; P = 0,0019. Therefore, impacts of satisfaction on attitudinal loyalty are influenced by both switch cost and commitment from relationships. • Testing for behavioral loyalty o Satisfaction affecting behavioral loyalty under the influence of the variable of switch cost

Research studies on behaviors of savings depositors at commercial banks have been conducted and analyzed by many world scientists. Some examples are those by JW Prinsloo (2000), Tanvi Singhal (2015), Kasri and Kassim (2009), Le Thi Thu Hang (2012), Pham Thi Tam – Da Lat University – and Pham Ngoc Thuy – Ho Chi Minh City University of Technology (2013), Nguyen Thi Thai Ha (2010). 2.3. Brand loyalty in banking 2.3.1. Definition of customers’ brand loyalty Regression coefficients for Int_1 (XM), b3 = 0,5588 and t(595) = 4,3180 are statistically significant (p = 0,0405 < 0,05). Therefore, impacts of satisfaction on behavioral loyalty are influenced by switch cost. This result enables us to conclude: Switch cost strengthens impacts of satisfaction on behavioral loyalty of individual savings depositors to brands of commercial banks. o Satisfaction affecting behavioral loyalty under the influence of the variable of commitment from relationships Nowadays there are three main approaches to customers’ brand loyalty: (1) Behavioral loyalty; (2) Attitudinal loyalty; (3) Combined behavioral and attitudinal loyalty.

Regression coefficients for Int_1 (XM), b3 = 0,0318 and t(595) = 2,854 are statistically significant (p=0,0401 < 0,05). Therefore, impacts of satisfaction on behavioral loyalty are influenced by commitment from relationships. This result enables us to conclude: Commitment from relationships strengthens impacts of satisfaction on behavioral loyalty of individual savings depositors to commercial banks. With advantages of the approach as well as methods of measuring the combination of behavioral loyalty and attitudinal loyalty, in this study, the author chooses to study brand loyalty in the direction of a combination of behavioral loyalty and attitudinal loyalty. The thesis uses the definition by Oliver (1999) to orient the analysis of loyalty of individual savings depositors to commercial banks. 2.3.2. Classification of brand loyalty o Satisfaction affecting behavioral loyalty under the simultaneous influence of moderating variables of switch cost and commitment from relationships In his Brand Loyalty pyramid, David A. Aaker (1991) [108] identifies five levels of brand loyalty, ranging from not loyal to very loyal; the highest level is depicted at the top and the lowest level at the bottom of the pyramid. Dick and Basu (1994) [99], depicts loyalty as a two-dimensional construct involving relative attitude and repeat patronage. Jones and Sasser (1995) [88], classifies loyalty based on relations between Regression coefficients show the following results: b4 = 0,0680; t(593) = 2,2087; P = 0,0005 and b5 = 0,0627; t(593)= 2,1902; P = 0,0402, both of which are statistically significant (p < 0,05), which means both switch cost and commitment from relationships are moderating variables affecting the conversion from satisfaction to behavioral loyalty of individual savings depositors. satisfaction and loyalty. 2.3.3. Overview of the research on some factors affecting brand loyalty in

banking There are various approaches to the research on factors affecting customers’ brand loyalty in banking, and the author has analyzed and consolidated the research works in two main directions.

The first direction involves research on factors directly affecting brand loyalty. Typical authors include: Zeyad and Norailis (2013), Benjamin (2006) Rujrutana & Yaowalak (2011), KokSal and Dema (2014), Nguyen Thi An Binh (2016). This group has presented many factors and Given the significance of “R2 increasing due to impacts” of moderating variables, impacts of both moderating variables account for 4,59%, conversion from satisfaction to behavioral loyalty, F(593) = 0,0305; P = 0,0102. The only moderation by switch cost accounts for 3,1%, conversion to attitudinal loyalty, F(1;593)=3,4760; P = 0,0305. The only moderation by commitment from relationships accounts for 3,8%, conversion to attitudinal loyalty, F(2;593)= 3,9802; P = 0,0008. Therefore, impacts of satisfaction on behavioral loyalty are influenced by both switch cost and commitment from relationships.

18 7 - Customer satisfaction have positive impacts on loyalty of saving depositors (0,620) - Impacts of satisfaction on loyalty are stronger than impacts of social responsibility, reputation and overall service quality on satisfaction. analyzed their impacts on brand loyalty. However, the shortcoming of this group is that their analysis model is still simple, thus impossible to look into complicated aspects of relationships among variables, which refers to the interrelation among independent variables and simultaneous impacts on dependent variables i.e. customers’ brand loyalty. - Social responsibility has positive impacts on satisfaction of savings depositors (0,254)

The second direction involves research on factors affecting brand loyalty through intermediary variables. Typical authors are Beerli et al (2004), Younes Megdadi el al (2013), Belén Ruiz et al (2016), Butt and Aftab (2013), Kaura et al (2015), Tran Duc Thang (2015). Accordingly, the intermediary variable that is often selected to affect loyalty is customer satisfaction. Therefore, all hypotheses of H1, H2, H3 are accepted, which means overall service quality, reputation and social responsibility have positive impacts satisfaction of savings depositors 4.2.3.3. Impacts of moderating variables on loyalty (Hypotheses H5, H6) • Testing for attitudinal loyalty o Satisfaction affecting attitudinal loyalty under the influence of the moderating variable of switch cost

Based on results consolidated from previous research studies on customer loyalty to bank brands and factors affecting brand loyalty in banking, the thesis has consolidated and listed 12 factors studied, including: Reputation, Trust, Service Quality, Reliability, Convenience, Switch cost, Satisfaction, Habit, Expectation, Perceived value, Relationship and CSR. 2.4. Research model in the thesis 2.4.1. Analysis of factors in the model Regression coefficients for Int_1 (XM), b3 = 0,0468 and t(595) = 3,1268, are statistically significant (p = 0,0220 < 0,05). Therefore, impacts of satisfaction on behavioral loyalty are influenced by switch cost. This enables us to conclude: Switch cost strengthens impacts of satisfaction on attitudinal loyalty of individual savings depositors to brands of commercial banks. o Satisfaction affecting attitudinal loyalty under the influence of the moderating variable of commitment from relationships

The thesis selects the following factors affecting brand loyalty of individual savings depositors: (1) Satisfaction, (2) Overall Service Quality, (3) Reputation, (4) CSR, (5) Switch Cost, (6) Commitment from relationships. Below are detailed analyses leading to selection of such factors: 2.4.1.1. Customer satisfaction.

Regression coefficients for Int_1 (XM), b3 = 0,0203 and t(595) = 2,3724 are statistically significant (p=0,0360 < 0,05). Therefore, impacts of satisfaction on attitudinal loyalty are influenced by commitment from relationships. This enables us to conclude: Commitment from relationships strengthens impacts of satisfaction on attitudinal loyalty of individual savings depositors to commercial banks. o Satisfaction affecting attitudinal loyalty under the simultaneous influence of switch cost and commitment from relationships

Regression coefficients show the following results: b4 = 0,0524; t(593) = 1,9862; P = 0,0305 and b5 = 0,0583; t(593)=1,9477; P = 0,0319, and both are statistically significant (p < 0,05), which means both switch cost and commitment from relationships are moderating variables affecting the conversion from satisfaction to attitudinal loyalty of individual savings depositors. According to Kotler (1991), satisfaction is defined as a person’s feelings of pleasure or disappointment resulting from comparing a product’s perceived performance (or outcome) in relation to his or her expectations. There are two directions in analyzing the interrelation between satisfaction and brand loyalty: first, satisfaction acts as an independent variable and second, satisfaction acts as an intermediary variable. The thesis chooses to study satisfaction as a variable directly affecting brand loyalty, which is an intermediary variable among variables of overall service quality, social responsibility and reputation with regard to impacts on brand loyalty of individual savings depositors. 2.4.1.2. Overall service quality.

Given the significance of “R2 increasing due to impacts” moderating variables, impacts of both moderating variables account for 5,1%, conversion from satisfaction to attitudinal loyalty, F(593) = 2,4134; P = 0,0245. The only moderation due to switch cost accounts for 4,3%, conversion to attitudinal loyalty, F(1,593) = 3,4950; P = 0,0475. The only moderation due to Overall service quality reflects the effectiveness of an enterprise’ core activities and is often considered an important determinant of loyalty, either directly or indirectly, through customer satisfaction that affects brand loyalty (Cronin and Taylor, 1992; Sasser, 1995; Parasuraman et al, 1991).

8 17

The thesis selects to study impacts of overall service quality on satisfaction of individual savings depositors to be included in the research model.

service quality including 5 items:

Results show that the variance inflation factor (VIF) of the analyzed aspects are < 2 and Sig.<0,05, making all variables statistically significant. It can also be seen from table 4.5 that all aspects of tangibles, responsiveness, empathy, reliability, assurance and tangibles have impacts on the overall service quality. Accordingly, assurance has the strongest impact (Beta = 0,273), followed by empathy (Beta = 0,263), tangibles (Beta = 0,261), responsiveness (Beta = 0,159) and finally reliability (B = 0,083). 4.2.2. Testing main research model Parasuraman (1985, 1988) presented SERVQUAL model, a scale for measuring overall reliability, responsiveness, assurance, empathy, and tangibles. The thesis has analyzed these five items separately in terms of their impacts on overall service quality which will later affect customer satisfaction and loyalty. 2.4.1.3. Reputation. Some researchers have shown

Testing relevance of the research model with SEM analysis shows the following results: Chi-Square / df = 2,982; GFI = 0.902; CFI = 0,916; TLI = 0.908; RMSEA = 0.068. This result shows that the model, if not yet affected by moderating variables, is entirely consistent with market data. the close relationship between reputation and development of turnover and market shares, leading to greater customer loyalty (Andreasen and Lindestad, 1998; Robertson, 1993).

Average estimated coefficient

Standard error

Critical value

SEM analysis of the linear structure model shows the test results below: Table 4.6: Model testing results without impacts of moderating variables Level of significance P

According to Belén Ruiz et al (2016) and many previous research studies, reputation is an important factor that affects brand reputation and can be inherited by the Vietnamese market as the system of commercial banks here reflects separation between public and non-public commercial banks. The thesis chooses brand reputation as one of the factors to be included in the research model on impacts on satisfaction and loyalty in Vietnam. 2.4.1.4. Social responsibility.

*** *** *** ***

***

0,028 0,088 0,116 0,788 0,050

9,168 5,168 4,487 10,135 10,420

<--- TNXH SHL <--- UTTH SHL <--- CLTT SHL LTT <--- SHL TTTĐ <--- LTT TTHV <--- LTT

0,254 0,412 0,576 0,620 0,519 1,000

***

Corporate social responsibility (CSR) programs can be greatly beneficial to enterprises, including increased customer loyalty (Berens et al, 2005, 2007 ; Lichtenstein et al, 2004).

P: level of significance; *** = p <0,001 All impacts in the research theory model without impacts of moderating

variables have the P-value significance <0.05. After analysis of Active CSR initiatives can affect customer attitudes to an enterprise and its products and services, reflect its core capacity (Bhattacharya and Sen, 2003 ; Folks and Kamins, 1999 ) and affect both emotional values and satisfaction of customers, thus affecting brand loyalty.

The author has included social responsibility in the research model as a factor affecting customer satisfaction and social responsibility in the context of the thesis is viewed from two perspectives i.e. responsibility to social environment and responsibility to customers. 2.4.1.5. Switch cost. impacts of 3 factors on satisfaction: social responsibility, reputation and overall quality, the strongest impact is made by overall quality (Beta = 0.576), followed by reputation (Beta = 0.412) and finally social responsibility (Beta = 0,254). Satisfaction positively affects loyalty at a strong level (Beta = 0.620). 4.2.3. Testing research hypotheses 4.2.3.1. Impacts on customer satisfaction (Hypotheses H1, H2, H3)

Estimates show that *** or P<0,001 and weights are positive, which means variables of social responsibility, overall quality and reputation have a proportional impact on satisfaction of individual savings depositors. 4.2.3.2. Impacts of satisfaction on customer loyalty without impacts of moderating variables (Hypothesis H4) Model testing results show that: Savings depositors at commercial banks in Vietnam, if switching to other banks, will face up with many obstacles: financial obstacles i.e. loss of preferential interest rates; time obstacles upon new accounting opening and depositing, or more time and efforts to be spent on familiarizing themselves with new transaction officers. Switch costs, according to many authors, directly affect brand loyalty (Beerli et al, 2004), (Bahareh et al, 2013); (Nguyen Thi An Binh, 2016).

16 9

Switch cost

9. 10. Commitment from relationships 11. Reputation 12.

5 5 6 8

0,831 0,869 0,885 0,884

60,407% 65,740% 63,624% 56,143%

Following previous research studies, the thesis selects switch costs as a moderating variable affecting the conversion from satisfaction to brand loyalty. 2.4.1.6. Commitment from relationships.

Relationships is emphasized by a number of research studies as an important variable in close connection with customer trust, satisfaction and loyalty (Cronin and Taylor, 1992; Crosby et al, 1990).

Social responsibility Results show that all Cronbach’s Alpha coefficients of studied variables are >0,6. The total variances extracted are all > 50%, making all scales reach reliability. Results of the rotated component matrix show that 62 observed variables can be divided into 8 groups.

To sum up, after Cronbach’s Alpha analysis and EFA, all factors in the

model ensure fulfillment of requirements for the next steps. 4.1.3. CFA testing criteria Summary of CFA testing results

Scale

Relationships between commercial banks and individual customers through relationships between bank staff and customers, can also be those between partners, acquaintances, friends or other social connections. A relationship is the foundation for a beneficial and sentimental commitment when an individual customer chooses a commercial bank to deposit savings, thus affecting the maintenance of his savings at that bank or his switch to another bank. Bank staff often take advantage of their relationships to mobilize and maintain personal savings from individual customers.

Loyalty Satisfaction Overall service quality Reputation Social responsibility Switch cost Commitment from relationships

Testing criteria Chi-square/df 4,238 4,335 3,261 3,451 4,567 4,426 3,263

GFI 0,913 0,955 0,923 0,921 0,936 0,927 0,939

TLI 0,921 0,965 0,916 0,955 0,968 0,944 0,941

CFI 0,941 0,948 0,925 0,938 0,944 0,981 0,925

RMSEA 0,057 0,064 0,053 0,074 0,062 0,075 0,053

Commitment from relationships, as studied, is a moderating variable between satisfaction and brand loyalty, and this is considered a new point of the subject in comparison with previous studies based on characteristics of the Vietnamese market.

Based on the above mentioned overall analysis of the research on factors affecting brand loyalty, the author selects to approach the research model through intermediary variables with main reference to the research model by Younes et al (2013) due to similarities in the research area, i.e. retail banking, in subjects of the research, i.e. individual customers, and in research objectives, i.e. brand loyalty. The proposed primary research model is presented below: The indicators in the table above show that all scales are relevant with market data and that the scales obtained in the official quantitative study qualify as a basis for testing the research model and testing research hypothesis in the next section. 4.2. Testing research model and hypotheses 4.2.1. Measuring the impact on the overall service quality using the SERQUAL scale

Table 4.5: The result of measuring the impact of factors using the SERQUAL scale

Collinearity statistics

Unstandardized coefficient

Standardized coefficent

Model

t

Sig.

B

Beta

B

Standard error

1

,054 ,054 ,054 ,054 ,054 ,054

,083 ,261 ,159 ,273 ,263

6,256 116,691 1 Constant ,125 Reliability 2,329 7,311 ,392 Tangibles Responsiveness ,239 4,454 ,410 Assurance 7,650 ,395 7,370

Empathy

6,256 Constant ,125 Reliability Tangibles ,392 Responsiveness ,239 ,410 Assurance ,395 Empathy 0,493

R2

10 15

group of customers having deposited savings for 5 years or more accounting for the smallest proportion (18%).

- Number of banks where they deposit savings: Number of customers depositing savings at 1 bank accounting for the majority, 68%, at 2 banks 23%, and at 3 banks or more 8%.

- Term of savings deposit. Term of 4-6 months accounting for the largest proportion (33%), term of below 1 month accounting for the smallest proportion (6%).

- Use of additional bank services. 47% of respondents are not using other bank services apart from account services and savings deposit. 53% of respondents are using additional bank services.

CHAPTER 4: RESEARCH RESULTS

4.1. Testing scales with confirmatory factor analysis (CFA) 4.1.1. Results of scale distribution form evaluation

• Variables in the model include: - Independent variables affecting customer satisfaction include: overall service quality, reputation and social responsibility. - The independent variable of satisfaction affects the dependent variable of brand loyalty. In terms of the average value of the observed variables, all the variables are averaged > 3 except for CP5 (based on Likert scale). In particular, the average values of the observed variables of service quality and satisfaction are relatively high (the highest are CLTC4 for 3.97 and CLPV2 3.93). Accordingly, loyalty, satisfaction, service quality, switch cost, reputation, social responsibility, and commitment from relationships of individual customers in the urban part of Hanoi are relatively high.

- The moderating variables affecting the conversion from customer satisfaction to brand loyalty include: switch cost and commitment from relationships.

• Research hypotheses include: - Hypothesis H1: Overall service quality has a positive impact (+ The results show that the values of the distribution variables are relatively normal, not too much left or right skewed. Consequently, the scales used are normal distributed, ensuring fulfillment of requirements for tests to be carried out in the next steps. 4.1.2. Results of scale testing with Cronbach’s Alpha and EFA impact) on customer satisfaction; - Hypothesis H2: Reputation has a positive impact (+ impact) on Table 4.2: Summary of reliability and total variance extracted from the scales customer satisfaction; - Hypothesis H3: Social responsibility has a positive impact (+ impact)

Conclusion

No.

Scale

on customer satisfaction;

Total variance extracted

Reliability coefficient with Cronbach’s Alpha

- Hypothesis H4: Customer satisfaction has a positive impact (+ impact) on brand loyalty; - Hypothesis H5: Switch cost has an impact on conversion from customer satisfaction to brand loyalty; - Hypothesis H6: Commitment from relationships has an impact on

All scales reach reliability

conversion from customer satisfaction to brand loyalty. -

1. 2. 3. 4. 5. 6. 7. 8.

Loyalty Tangibles Assurance Responsiveness Empathy Reliability Overall service quality Satisfaction

Number of observed variables 10 5 3 4 4 4 3 5

0,902 0,833 0,772 0,782 0,816 0,882 0,915 0,809

69,651% 60,216% 69,027% 61,227% 64,957% 73,917% 85,651% 57,351%

14 11 CHAPTER 3: METHODOLOGY Quantitative research is conducted in two stages: Preliminary quantitative research and official quantitative research. 3.4.1. Preliminary quantitative research 3.1. Overview of research areas 3.1.1. Volume of personal savings mobilization at commercial banks

- Questionnaire design: In addition to the self-introduction and introduction of purposes of the survey, the main part of the questionnaire consists of 7 parts. The questionnaire is designed according to Likert scale of 5 levels. - Sample size for preliminary quantitative research: N = 200 - Data analysis method: Reliability coefficient method of Cronbach’s Alpha and EFA. There is a fairly remarkable difference in the volume of personal savings mobilization from individual customers between public commercial banks and non-public commercial banks. State commercial banks including VietinBank, Vietcombank and Agribank (data from BIDV not available) mobilize a significant amount of personal savings worth 3-10 times as much as that by non-public banks due to their advantages in terms of networking, brand value and reputation. 3.1.2. Proportion of personal savings deposits at commercial banks

- Preliminary scale assessment: Removing 2 observed variables, MQH1 and MQH3, the scale of commitment from new relationships includes 5 observed variables and other scales remain unchanged. 3.4.2. Official quantitative research - Scale of research samples: 650 questionnaires distributed and 602 valid questionnaires collected.

Most commercial banks have a greater proportion of capital mobilized from individual customers than that from corporate customers, except for SHB. Techcombank, Agribank, Maritimebank, and Eximbank are those having large proportions of personal savings deposits, accounting for over 70% of the mobilized capital structure. In particular, Techcombank has capital mobilized from individual customers take up a nearly absolute proportion, 99%. 3.1.3. Number of transaction points of commercial banks - Sampling method: Randomly selecting individual customers on a visit to commercial banks for savings deposit. Bank X on the scale is defined as the bank where the customer has deposited personal savings for the longest time. The interviewer asks 2 exclusion questions and if the respondent answers yes,. - Sampling coverage: Hà Nội – urban part. Customers coming from 2 groups of public commercial banks and non-public commercial banks. - Analysis methods: CFA, EFA and Cronbach’s alpha, regression The Bank's transaction points include: Head Office, Transaction Office and branches, transaction offices / savings funds. The number of transaction points of VietinBank and Agribank accounts for over one third of the total number of branches and transaction offices of the system 3.1.4. Effectiveness of personal savings mobilization at commercial banks analysis. 3.5. Statistics of research samples

- Gender: 232 men (39%) and 370 women (61%). - Age: Age group of 35 to 45 (34%) accounting for the largest proportion, followed by the age group of 22 to 35 (29%). - Qualifications: Group with graduate and postgraduate qualifications accounting for the largest proportion (52%). - Occupation: Group of enterprises’ staff accounting for the largest Vietcombak is the leading bank in mobilizing savings of individual customers at transaction offices with the annual average amount of 743 million dongs in 2916 and 891 million dongs in 2017, the next is Techcombank with 554 million dongs in 2016 and 387 million dongs in 2017, MB with 416 million dongs in 2016 and 458 million dongs in 2017, VPBank with 405 million dongs in 2016 and 423 million dongs in 2017. 3.2. Overview of research methodology and process 3.2.1. Research process proportion (37%). To ensure the accuracy of research results, the author follows the research process below: - Average income. Income group of 5 to < 9 million VND accounting for the largest proportion (34%), income group of over 25 million VND accounting for the smallest proportion (6%).

- Time of starting transaction with bank. Group of customers having deposited savings for 1-3 years accounting for the largest proportion (42%),

12 13

Equation Modeling) linear analysis to test and measure the amount of relevance of theoretical models and research hypotheses.

- Using multiple regression analysis with PROCESS command of Hayes (2013) to test hypotheses H5 and H6 with two moderating variables. - Using multi-group structural analysis to compare the research model by groups of identifiers to identify meaningful differences of some specific public groups. 3.2.3. Formation of scale

The scale is inherited from the following scholars: Lada et al (2009), Kassim and Abdullah (2010), Wang Yonggui (2003), Younes et al (2013), Neeru & Paul (1999), Belén Ruiz (2015), Perez and Bosque (2014). 3.3. Qualitative research 3.3.1. Qualitative research methods 3.3.1.1. Respondents

• Expert interview - Quantity: 10 persons - Subjects: Experts in banking and finance, experts in marketing and branding.

3.2.2. Methods of data collection and processing • Methods of analysis and consolidation of secondary data • Focus group interview - Quantity: 10 persons - Subjects: Individual customers having over 5 years’ experience in using savings deposit products in commercial banks, having experienced savings deposit at more than one bank, living and working in Hanoi. 3.3.1.2. Contents of interview - The interview basically focuses on factors affecting brand loyalty of Collecting, analyzing, comparing and evaluating some research studies on factors affecting brand loyalty of individual customers using banking services from local and international sources available. individual savings depositors in commercial banks. 3.3.1.3. Data construction and processing methods • Qualitative research methods - Expert interview: to check the relevance of draft scale 1 related to - Data are collected through direct interview and collated before potential variables studied. consolidation into the most common viewpoint for the research subject. 3.3.2. Qualitative research results - Checking the variables in the model: the preliminary research model - Focus group interview: Used to check the level of detail and comprehensibility of the statements used in draft scale 1 and modify the language of draft scale 1 to complete draft scale 2 . is preserved and considered as the official research model of the thesis - Evaluating the scales: observed variables and language of some scales are adjusted before applied to quantitative research. 3.3.3. Official research model and draft scale 2 • Quantitative research methods - Developing questionnaires and carry out surveys - Using SPSS 22.0 for Cronbach's Alpha analysis in order to eliminate variables having small correlation with aggregate variables, check EFA to eliminate variables having small factor loading (FL) weight. - Using AMOS 20.0 for confirmatory factor analysis (CFA) to eliminate variables having small CFA weight and for SEM (Structural The official research model is maintained and the draft scale 2 consisting of 62 observed variables are coded and applied to quantitative research. 3.4. Quantitative research