Factors Affecting the Choice of Banks: Do Bank's Interest Rate, Employee Image and Brand Matter?

Abstract

The banking system provides a number of important functions for the economy and is also the lifeblood and financier of the economy in each country. Large amounts of idle money have not been exploited by banks; however, banks still depend on loans, including loans from foreign banks, to meet the growing demand, as such, for banks, the cost of capital is high, the stability and business efficiency are low and banks have not promoted their internal resources to grow steadily. To achieve the goal, this research analyzes the factors affecting the choice of bank for the deposit decisions of customers in Vietnam. The study used a sample data of 250 individuals and SPSS software was used to analyze the data. The results showed that customer policy has a positive effect on customers' deposit decisions in a bank, and this is new evidence regarding behavioral theory in the case of Vietnam. Results further demonstrated that other factors such as employee image, brand, interest rate, relative influencing, and transaction time positively impact the choice of bank for the deposit decisions of customers. However, the bank's promotion strategies had no impact on the choice of bank for the deposit decisions of customers. Besides, employee image is the most influential factor in the deposit decisions, followed by the bank's brand and interest rate.

1. Introduction

The banking system plays an important role in an economy’s performance and is the lifeblood of the economy. Mobilized capital contributes significantly to banks’ growth and efficiency. Specifically, mobilized capital is crucial for banking activities and is the basis for the bank to conduct lending, investment, and reserve activities that bring profits to the bank (Yavas, Babakus, & Karatepe, 2013; Chigamba & Fatoki, 2011; Dao et al., 2020). To obtain this source, the bank needs to conduct capital mobilization activities, in which mobilization of deposits activity is particularly important. However, the deposit mobilization activity of banks now faces many difficulties such as fierce competition from other entities in the economy that also conducts deposit mobilization activities (Denton & Chan, 1991; Yavas, Babakus, & Karatepe, 2013; Chigamba & Fatoki, 2011).

Large amounts of idle money have not been exploited by banks; however, banks still depend on loans, including loans from foreign banks, to meet the growing demand, as such, for banks, the cost of capital is high, the stability and business efficiency are low and banks have not promoted their internal resources to grow steadily (Yavas, Babakus, & Karatepe, 2013; Chigamba & Fatoki, 2011). Attracting capital at a high cost leads to low stability, limits profitability, and at the same time put the bank at risk of interest rate risks. Liquidity risk can lead to instability in the entire financial system (Truong & Pham, 2011; Tran, 2012). In the market today, there are many different investment channels with different levels of profit and risk; however, there are still many people who choose to invest their savings in bank deposits, although the interest rate on such deposits is not attractive (Nguyen, 2011; Moradzadeh et al., 2014). Moreover, there is increasing competition among commercial banks and non-banking entities such as insurance, mutual funds, post office, and treasury bonds in mobilizing capital. Hence, customers have many options to invest their financial resources, as such, to attract capital (mobilize customer savings), banks use appealing advertisements and undertake marketing and promotion programs to influence customers to deposit their savings with them

To achieve the goal, this research analyzes the factors affecting the choice of bank for the deposit decisions of customers in Vietnam. In particular, as a new concept, this study analyzes customer policy as a factor that impacts customers’ deposit decisions in a bank. With a series of economic and political reforms, Vietnam has now become a “socialist- oriented market economy” which is characterized by openness to trade and foreign investment. There are nearly 100 banks including local banks and branches of foreign banks operating in the territory of Vietnam, of which there are four state-owned commercial banks, also known as the “Big 4” banks, including Vietcombank, VietinBank, BIDV, and Agribank. To conduct this research, we selected Vietinbank in Ho Chi Minh City, Vietnam. In the context of the banking system, Vietinbank is one of the major banks, and the bank has continuously promoted its strengths in the currency trading field and well- implemented business strategies. The bank dominates in the corporate customer segment; however, recently with the increasing competition, this segment is no longer attractive to banks. Currently, the people of Vietnam have enough idle money; however, this segment has not been paid enough attention, therefore, banks are now focusing on them.

The remaining of this paper will be organized as follows: In Section 2, we discuss the literature review. In Section 3 and Section 4, we discuss data selection, methodology, and results. In Section 5 we further analyze and discuss the results, and Section 6 includes the conclusion.

2. Discussion on Behavioral Theories

2.1. Theory of Reasoned Action (TRA)

As shown in Figure 1 below, it is Theory of Reasoned Action (TRA) and was developed in 1967 and further discussed by Ajzen and Fishbein (1980). Theoretically, TRA indicates that the trends of consumption are known as the best predictor of consumer buying behavior. Furthermore, the factors that contribute to consumer buying behavior are attitude and subjective standards of consumers. Attitude is measured by awareness of the properties of the products while subjective factors can be measured by people related to consumers such as family, friends, colleagues, who like or dislike them (Keng et al., 2014; Tran and Vu, 2019). The stronger level of intimacy of people involved with the consumer, the greater influence on consumer buying behavior.

2.2. Theory of Planned Behavior (TPB)

As shown in Figure 2, the Theory of Planned Behavior (TPB) was introduced and improved by Ajzen’s Rational Acton Theory. Further, in the consumer buying behavior theory, in addition to the two main factors (personal attitudes and subjective standards) that influence consumer buying behavior, another factor that influences consumer buying behavior is cognitive behavior control. In addition, cognitive- behavioral control depicts how easy or difficult, it is to show a behavior and whether it is controlled or restricted.

2.3. Theory of Buyer Behavior (TBB)

Howard’s and Seth’s (1969) Theory of Buyer Behavior (TBB) also known as the Howard-Sheth model, and to be one of the most comprehensive models of consumer buying behavior. It is likely to use the concept of stimulus-response to explain buyer’s brand choice behavior over the period of time. In addition, the stimulus inputs or input variables refer to the stimuli in environment, and they take the form of informative cues about the product and service offering. These information cues could relate to quality, price, distinctiveness, service, and availability. Other factors that can significantly influence consumers’ buying behavior are cultural, social, and religious factors as well as consumers’ financial position, personality, and lifestyle.

Figure 1: Theory of Reasoned Action

Source: Ajzen (1991)

Figure 2: Theory of Planned Behavior

Source: Ajzen (1991)

2.4. The Consumer Decision Model (CDM)

The consumer decision model is known as the Blackwell, Miniard, and Engel, was developed in 1968 by Blackwell, Miniard, and Engel (2006) and has been revised several times. In the model, the consumer decision-making process includes four components such as input, information processing, decision-making process, and variables affecting the consumer’s decision- making process. The decision- making process of customers is a continuous process from recognizing needs, gathering information, reviewing options, making purchasing decisions, and evaluating after purchasing.

3. Literature Review and Hypotheses

3.1. Literature Review

Determinants of bank selection criteria and bank preferences of bank have been investigated across the world. As shown in Anderson et al. (1976) who use “determinants attribute analysis” in a survey in the USA, stated that convenience customers selected recommendation by friends as the most important factors, followed by reputation, location, friendliness of bank staff, and service charges. Whereas, service customers ranked availability of credit as the main factors, followed by reputation, recommendation by friends, friendliness of the staff, and interest charged on loans. In a study on Polish banks, Kennington, Hill, and Rakowska (1996) investigated the selection criteria applied by Polish consumers, and concluded that major factors such as bank’s reputation, prices, and services should be considered when devising strategies to attract more customers in the new free economy. Furthermore, this study also indicates that Polish banks were conservative when using a new financial product, therefore, they were slowly changing in a positive way. Similarly, Devlin (2005) in a study in the UK, described that UK banks offer a wide range of incentives and product ranges, which are recommended by many customers to their relatives and friends. Moreover, the location of the bank also has a significant influence on the decisions of customers.

Denton and Chan (1991) investigated multiple banking behavior in Hong Kong and found that banking system is widespread and is influenced by factors as convenience in terms of the number of branches, risk reduction, automatic teller machines (ATMs0, the relative advantage of selected banks, need for credit and credit cards, prestige, and special circumstances. Chigamba and Fatoki (2011) studied commercial bank choice among university students in South Africa and showed that the most influencing factors affecting the decision to choose a bank include the quality of service and convenient transaction location. In contrast, the price and marketing strategy of the bank had little influence on the decision to choose a bank. Abbam, Dadson, and Say (2015) empirically analyzed bank choice in Ghana based on 509 customer surveys. The results showed that the quality of service affects the customer’s choice of bank, followed by a convenient ATM location. The customers would like to deal with banks that provide ATMs facilities so that they can enjoy banking services during the late hours of the days and weekends. Similar to Chigamba and Fatoki (2011), factors that have little influence on a customer’s bank choice decision include service fees, opening time of the bank, and the number of tellers (Abbam, Dadson, & Say, 2015).

Finger and Hesse (2008) empirically examined the demand for commercial bank deposits in Lebanon. At the macro level, they found that domestic factors as economic activity, prices, and the interest difference between the Lebanese pound and the US dollar are significant in explaining deposit demand. At the micro level, they found that besides, bank-specific factors as the perceived riskiness of individual banks, loan exposure, liquidity buffers, access to credit, interest rates, and interest margins, bear a significant influence on the demand of deposits. Further, Yavas, Babakus, and Karatepe (2013) in their research on banks in the US and based on a survey of 400 individual customers, demonstrated that the consumer buying behavioral model comprises three components such as research, trust, service fees, and a new banking service. In addition, trust factors include confidential customer information, bank’s adoption of scientific and technical advances, and bank governance while experience factors friendliness, the experience of the staff, speed of decision making, and the bank providing services promptly.

Moradzadeh et al. (2014) ranked main factors influencing the selection of customers in depositing their finance in the bank. In this study, Moradzadeh et al. (2014) categorized the causes and aggravating factors affecting customer orientation to the depository. Using a sample data of 121 observations collected on the primary data and using Friedman test, results show that interest rates, quality of banking services, diversity of services, facilities, the attitude of the bank staff, and electronic banking services, the number of branches, and bank locations, the beauty of the banks have significantly impact on customer orientation to the depository in banks. Adem (2015) studied the impact of bank’s location, work experience, and information technology techniques on the capital mobilization by commercial banks in Ethiopia by interviewing 180 bank clients and according to logit binary regression method, results indicate that work experience of employee had a little bit influence on the ability of banks to raise capital while the investment in information technology had a great influence on commercial bank savings.

3.2. Research Hypotheses

With the aim to consider the factors affecting the choice of banks for the deposit decisions of customers in Vietnam, in this study, we will further discuss the six main factors such as the bank’s promotion strategies, employee image, brand, interest rate, relatives influence, and transaction time. In particular, as a new concept, this study analyzes customer policy as a factor that impacts customers’ deposit decisions in a bank. Figure 3 describes the research model:

3.2.1. Employee Image

According to Anderson et al. (1976) and Almossawi (2001), the most important criteria for bank selection among university students in Bahrain is employee image. Tran (2012) analyzed factors affecting the level of depositors’ satisfaction in Vietcombank, in Long An province in Vietnam. He found that the satisfaction of depositors depends on the employee image and quality of services of Vietcombank. Therefore, to attract more depositors, the bank needs to pay attention to the employee image and quality of services to attract deposits from customers. Nguyen (2011) stated that in banking business activities, customers often interact with and deal with bank employees, as such, the attitude and working style of employees have a robust influence on the reputation, and image of banks. In addition, factors as knowledge, service attitude, experience, appearance, staff, and ability to follow customers can affect the quality of services (Mukherjee et al., 2017; Tran, 2012).

Figure 3: Proposed Research Model

3.2.2. Bank’s Brand

Kennington, Hill, and Rakowska (1996) showed that factors such as bank reputation, price, and service do affect Polish banks’ decision to deposit. They concluded that reputation, and pricing concerns are more important to customers in the case of Poland, who are gradually adapting to the emergence of private banks. Additionally, brand image has a more specific effect on the perceptions of product and service quality of customers. Brand image also indicates the social image, which is the added value that explains why people buy or use that brand.

3.2.3. Interest Rate

According to the studies by Finger and Hesse (2008), Yavas, Babakus, and Karatepe (2013), Moradzadeh et al. (2014), and Tran (2012), the interest rate factor is the most influencing factor on the customers’ deposit decisions in a bank, followed by the convenience factor, bank’s promotion strategies, influence of relatives, bank image, employee image and finally transaction time.

3.2.4. Relatives’ Influence

According to the studies by Abbam, Dadson, and Say (2015) and Tran (2012), the influence of relatives can positively influence the depositors’ (customers’) behavior who are likely to have trust in that bank and decide to deposit their savings in that bank. Further, Abbam, Dadson, and Say (2015) suggested that the introduction by family and friends could easily help customers have a better understanding of the products and other services of the bank.

3.2.5. Transaction Time

As shown by Devlin (2005) when customers carry out activities related to money (their money as savings in the bank), they are also interested in simplicity, quickness of the transaction, and saving time. When a customer visits a branch, they want to be engaged, receive personal attention, and not have to wait around to be noticed by the bank staff. When customers need to carry out some transactions, they may have to follow many procedures. As such, when a customer feels that all these would take a long time, they will feel irritant, thereby creating a bad impression for the bank. If the transaction time is lower, it will save time for customers and satisfy customers, thereby creating favorable conditions for banks in capital raising activities.

3.2.6. Bank’s Promotion Strategies

Tran (2012) had shown that attitude towards marketing and promotion strategies of banks influences consumer brand desire. In addition, customers have a good attitude, and are also interested in the promotion program of either product or brand, they are willing to recognize the presence of that product or brand, and recognize the difference it from other products, brands, and when in demand, the ability to select that particular product of brand is high.

3.2.7. Customer Policy

Theoretically, a new customer is the source of life for any store and business. Therefore, policy of customer has become one of the most important factors and requires lots of investment regarding effort and financial resources. In addition, customer policy is not just about selling to customers the products and services, it requires businesses to create absolute satisfaction for customers with their products and services.

H1: The image of an employee influences customer choice of commercial banks for their deposit decisions (+)

H2: Factors of bank brand influences customer choice of commercial banks for their deposit decisions (+)

H3: The interest rate factor influences customer choice of commercial banks for their deposit decisions (+)

H4: Relatives influence impact customer choice of commercial banks for their deposit decisions (+)

H5: The transaction time influences customer choice of commercial banks for their deposit decisions (+)

H6: The bank’s promotion strategies influence customer choice of commercial banks for their deposit decisions (+)

H7: The customer policy influences customer choice of commercial banks for their deposit decisions (+)

Deposit decision = f (Employee image, bank brand, bank brand, relative influence, transaction time, bank’s promotion strategies, customer policy)

3.3. Variables Selection

Table 1: The Variables Development

Table 1: (Continued)

4. Data Collection and Research Analysis

4.1. Data Collection

In this research, we gather primary information through actual survey results of individual customers at the Vietnam Joint Stock Commercial Bank for Industry and Trade in Ho Chi Minh City. According to Hair et al. (1998), to be able to perform factors influencing choice of bank, we need to collect data with a sample size of at least 5 samples per observation variable, preferably over 10 samples. In addition, the research model has 30 observed variables. For the standard 5 samples for an observed variable, the required sample size is 175. Accordingly, the sample size that we chose is 250, which meets the necessary sample size of 175.

4.2. Research Analysis

4.2.1. Cronbach’s Alpha

Cronbach’s alpha reliability coefficient is a test of the suitability of the scale for each observed variable. It measures reliability or internal consistency of a set of scale or test items, and reliability is how well a test measure what it should. Further, this method allows eliminating inappropriate variables and limiting the variables in the research process. Theoretically, Cronbach’s alpha should be in the range of 0.9 and 1, the scale is good, and the range of 0.7 and 0.8, it means the variables can be used.

4.2.2. Exploratory Factor Analysis (EFA)

In this study, Bartlett’s test of sphericity tests the hypothesis that the correlation matrix is an identity matrix, which would indicate that the variables are unrelated and therefore unsuitable for structure detection. Small values with less than 0.05 of the significant level indicate that factor analysis may be useful with its data. In addition, regarding factor loading, Hair et al. (1998) give rules of thumb for assessing the practical significance of standardized factor loadings as denoted by either the component coefficients in the case of principal components, the factor matrix (in a single factor model of an uncorrelated multiple factor model) or the pattern matrix (in a correlated multiple factor model). Further, factor loading, also ensuring the practical significance of EFA. More specifically, when factor loading is 0.3 and more, it is considered to be a minimum, 0.4 and more, it is considered to be important, and > 0.5 is considered to be practical. For the average variance extracted (AVE), the total variance extracted 50%, and AVE is a measure of the amount of variance that is captured by a construct in relation to the amount of variance due to measurement error.

Additionally, Kaiser-Meyer-Olkin (KMO) test is a measure of how suited the data is for factor analysis. A unit KMO is a ratio between the square of correlation of variables and the square of partial correlation of variables. The value of a large KMO from 0.5 to 1 is significant for factor analysis, but if the value is less than 0.5, factor analysis is most likely not appropriate for the data. Further, the eigenvalue is a measure of how much of the variance of the observed variables a factor explains. Any factors with an eigenvalue ≥1 explains more variance than a single observed variable. Accordingly, only factors with an eigenvalue greater than 1 are retained in the analysis model while factors with an eigenvalue index of less than 1 explain the amount of variance and are generally discarded.

5. Empirical Results

5.1. Descriptive Statistics

Through the survey results, of the total savings account customers, 56.8% are men 43.2% are women. 60.4 % of the savings account customers fall in the age group of 35-55 years. This age group is the middle age where most people have a high and stable income. Of the main target customers of the bank, 41.6% of the savings account customers are officials and employees (working people) and 36.8% of the savings account customers are into business and trading. 72.8% of the savings account customers have an income between VND 9 million to VND 20 million. However, savings account customers who an income of more than 20 million, only account for 4.4%.

5.2. Assessment of Reliability of Scale and Factor Analysis

In this analysis, it is evident that Observed variables with Item-total Correction coefficients less than 0.3 will be disqualified and thus, the selection criteria will be chosen because Cronbach’s Alpha is 0.6 or higher (Hair, 1998). The summary table of the first Cronbach’s Alpha test results is presented as follows:

From the results in Table 3, it can be seen as the most Cronbach’s alpha coefficient is greater than 0.6. However, the correlation coefficients for the factors such as “Employee image”, “Relative influence”, “Transaction time”, “Bank’s promotion strategies”, and “Customer policy” are smaller than 0.3, and does not meet or conform to the requirements. Accordingly, these observed non-conforming independent variables should be extracted from the scale. For this reason, the test will be conducted with the remaining observed variables as follows:

Table 2: Characteristics of Samples 

Table 3: Cronbach’s Alpha Test Results before Eliminating Variables 

Table 4: Cronbach’s Alpha Test Results after Eliminating Variables

Thus, the overall assessment for the 7-part scale, that is, the reliability of the survey data for the 7-part scale ensures reliability. The survey results will be used in EFA in the next step. Approximately 33 observed variables were included in the EFA process (except 5 variables which were excluded in Cronbach’s alpha step). In particular, the 7-part scale components are: “Image of employees”, “Influence of relatives”, “Transaction time”, “ bank’s promotion strategies “ and “Customer policy” include 30 observed variables, the decision to save money consists of 03 observed variables. 

Table 5: The 2nd EFA Analysis

EFA is performed with Principle Component Analysis (PCA) with Varimax rotation (a statistical technique used at least one level of factor analysis as an attempt to clarify the relationship among factors) as well as the indicators as Community standards, Factor loading, Eigenvalue, AVE, and KMO coefficients (Kaiser-Meyer-Olkin) also over 0.5 to ensure consistent data for EFA. The results of the EFA of factors belonging to this second 7-part scale have the following specific results:

Table 5 indicates that the KMO coefficient in the analysis is 0.767> 0.5 which is significant for factor analysis, that is, showing that the factor analysis is appropriate for the data. In addition, the test of Bartlett has a Sig coefficient of 0.000 < 0.005, indicating that the result of factor analysis ensures statistical significance. With the variance extracted by 66.555%, it depicts that the variation of the analyzed factors explains 66.555% of the variation of the original survey data. Further, the Eigenvalue of factor 7 is 1.379 > 1, showing the convergence of the analysis stopped at the factor 7, therefore, there are 7 factors extracted from the survey data. Additionally, the factor loading of each observed variable also shows each of the factors is greater than 0.5, which proves that the influence of the factors will represent.

The seven factors described are as follows. Factor 1 consists of 5 observed variables: LS1, LS2, LS3, LS4, and LS5. These variables constitute the factor “Interest rate” – The symbol is LS. The observed variables have a loading factor greater than 0.7, therefore, all of these observed variables are significant. Factor 2 consists of 5 observed variables: THNH1, THNH2, THNH3, THNH4, and THNH5. These variables constitute the factor “Bank brand” - The symbol is THNH. The observed variables have a loading factor greater than 0.6, therefore, all of these observed variables are significant. Factor 3 consists of 4 observed variables: HANV1, HANV2, HANV3, and HANV4. These variables constitute the factor “Staff image” – The symbol is HANV. The observed variables have a loading factor greater than 0.6, therefore, all of these observed variables are significant. 

Factor 4 consists of 4 observed variables: CSKH1, CSKH3, CSKH4, and CSKH5. These variables constitute the factor “Customer policy” - The symbol is CSKH. The observed variables have a loading factor greater than 0.5, therefore, all of these observed variables are significant. Factor 5 consists of 4 observed variables: HTCTNH1, HTCTNH2, HTCTNH4, and HTCTNH5. These variables constitute the factor “ Bank’s promotion strategies “ – The symbol is HTCTNH. The observed variables have a loading factor greater than 0.5, therefore, all of these observed variables are significant.

Factor 6 consists of 4 observed variables: AHNT2, AHNT3, AHNT4, and AHNT5. These variables constitute the factor “Influence of relatives” – The symbol is AHNT. The observed variables have a loading factor greater than 0.6, therefore, all of these observed variables are significant. Factor 7 consists of 3 observed variables: TGGD2, TGGD3, and TGGD4. These variables constitute the factor “Transaction time” – The symbol is TGGD. The observed variables have a loading factor greater than 0.7, therefore, all of these observed variables are significant.

In terms of analyzing discovery factors in scales of dependent variables, after reaching reliability with the  Cronbach’s Alpha test, the EFA was used to retest the convergence of the observed variables. The scale of the decision of deposit includes QD1, QD2, QD3. The results of the EFA factor analysis for scales belonging to the decision of deposit factor have the following results as shown in Table 6.

Table 6: EFA Analysis Results for Deposit Decision Factor

Table 6 indicates that the KMO coefficient in the analysis is 0.684> 0.5 which is significant for factor analysis, that is, showing that the factor analysis is appropriate for the data. Further, the test of Bartlett has a Sig coefficient of 0.000 <0.05, showing that the result of factor analysis meets statistical significance. The variance extracted by 69.535%, which shows that the variation of the analyzed factors explains 69.535% of the variation of the original survey data - this is a relatively high level of significance.

Table 6 further indicates that the Eigenvalues factor of 1st factor is equal to 2.086 > 1, indicating that there is 01 element extracted from the data. In addition, the loading factor of each observed variable depicts that the factors are greater than 0.7, and influence the factors that these variables represent.

5.3. Multivariate Regression Analysis

We perform correlation coefficient analysis for all the

7 variables which include 6 explanatory and a dependent variable with coefficient of Pearson, and two-sided test with the significant level of 0.05 in the step after of the completion of EFA analysis, Cronbach’s alpha test and before performing regression analysis. The regression model can be employed to predict values of a particular variable according to knowledge of its association with know values of other variables, and it can be used to test scientific hypotheses about whether and to what extent certain independent variables explain variation in a dependent variable of interest. The following table simulates the independence between dependent variables and independent variables. The correlation reaches a significant level of 0.05 (the probability of accepting the false assumption is 5%), then all variables correlate with the dependent variable.

Table 7: Pearson’s Test Result between Dependent Variables and Independent Variables

Table 8: Results of Multivariate Regression Analysis

Adjusted R² = 0.522;

F-test with Sig: 0.000

Durbin-Watson d-value: 1.933

Figure 4: P – P plot of Regression Standardized Residual

Figure 5: Frequency chart of standard residuals

Table 7 depicts a high correlation coefficient between explanatory variables and dependent variable in the range between 0.13 and 0.54. Sig values of all factors are less than 0.05, indicating that the explanatory variables correlate with the dependent variable, therefore, the inclusion of the explanatory variables into the model is correct since it has a certain effect on the dependent variable.

Table 8 depicts that the F test for the value of Sig. <0.05, proving that the model is suitable and, together with that, the adjusted R2 is close to 0.522. It indicates that the regression model discusses 52.2% of the variation of the explanatory variables. Therefore, the model has a fairly high explanatory value. In addition, we realize that the factor “Bank’s promotion strategies “ has a Sig value of 0.669> 0.05, therefore, it can be concluded that this factor does not affect the decision to deposit money. The remaining factors all influence the decision to deposit money since the value of Sig <0.05.

Figure 4 indicates testing the normal distribution of residuals. The results show that the residuals of the regression model are scattered randomly around the 0 axis in a constant range. It means that the variance of the remainder is constant. As shown in Figure 5, the standard deviation is approximately 0.986 and the standard distribution of the mean is equal to zero. Therefore, determining the residuals with standard distribution is accepted. Further discussed on auto-correlation, Durbin-Watson test are used to check correlation of adjacent errors. In the case of evaluating this test, there are two hypotheses as below:

Null Hypothesis: The overall correlation coefficient of the residuals is equal to zero

Alternative Hypothesis: The overall coefficient of the residuals is not equal to zero correlation

Table 8 depicts that the results of the Durbin-Watson d-value in the summary of the model by 1,933. According to regression conditions, Durbin – Watson values must be in the range of 1 to 3. Accordingly, the calculated d value falls into the domain that accepts the null hypothesis. Thus, the model does not violate the assumption of auto-correlation.

About assuming no multicollinearity phenomenon, which is tested in Table 9. With a large Tolerance and a VIF (Variance Inflation Factor) of variables, the regression model does not violate the multicollinearity phenomenon and is consistent with the VIF value of lower than or equal to 10.

- Hypothesis H1: The interest rate is positively related to the decision to deposit. Therefore, the assumption is acceptable in the case of the value of Sig less than 0.05 with a standardized Beta of 0.257. In addition, if the interest rate increased by 1 unit, the decision to deposit increased to 0.257 units. The interest rate is the third most influential factor. This finding is supported by Kennington, Hill, and Rakowska (1996) who stated that a bank with a better interest rate could help its customers with a more competitive advantage in the financial market.

Table 9: Multicollinearity Test

Table 10: Summary of Testing of Research Hypotheses

- Hypothesis H2: Bank’s brand is also consistent with the decision to deposit and positively related to the decision to deposit. Similarly, this finding is supported for the hypothesis in the result of Sig value less than 0.05 and a standardized Beta of 0.292. Therefore, if the brand of the bank increased by 1 unit, the decision to deposit increased to 0.292 units. The brand of the bank is the second most influential factor. This finding is consistent with Kennington, Hill, and Rakowska (1996) who stated that in Polish banks, the brand image has significantly influenced on the perceptions of product and service quality to the customers. Additionally, the brand image also indicates the social image which is the added value and explained why people buy or use the brand.

- Hypothesis H3: Employee image is positively related to the decision to deposit. Therefore, if the employee image factor increased by 1 unit, the decision to deposit increased to 0.321 units. Employee image is the most influential factor. This result is consistent with the studies by Tran (2012), Denton and Chan (1991), and Laroche, Rosenblatt, and Manning (1986), who stated that the employee image, attitude, and employees’ working style have a significant effect on the reputation, image in the banks. Accordingly, with the knowledge, experience, service attitude, ability to follow customers, and appearance, bank employees can affect the quality of services.

- Hypothesis H4: The customer policy is positively related to the decision to deposit. Thus, this finding is accepted. In fact, policy related to customer as one of the most important factors and requires a huge amount of investment in terms of effort and financial requirement. In addition, if the customer policy factor increased by 1 unit, the decision to deposit increased to 0.176 units. Customer policy is the fifth influential factor. This finding is nominated by this study and contributes to the gap regarding behavioral theory concerning the choice of bank for the deposit decisions of customers in Vietnam

- Hypothesis H5: Bank’s promotion strategies do not impact the decision to deposit. However, this assumption is rejected because the value of Sig is greater than 0.05. Therefore, there are no findings to confirm the relationship between the bank’s promotion strategies and the decision to deposit in the case of Vietnam. Similarly, Tran (2012) conducted a study in Vietnam and has a similar finding. had shown that attitude towards marketing and promotion strategies of banks influences consumer brand desire. In fact, if consumers have such a good attitude, and also willing to accept in the promotion program of that product or brand, they are more likely to recognize the presence of that product or brand, and differentiate it from other products or brands. - Hypothesis H6: Relatives’ influence is positively related to the decision to deposit. Similarly, this evidence is acceptable because the value of Sig is less than 0.05 with a standardized Beta of 0.190. Therefore, if relative influence increased by 1 unit, the decision to deposit increased to 0.190 units. Compared to the previous studies, this finding is consistent with Abbam, Dadson, and Say (2015) and Tran (2012), who stated that the influence of relatives can positively influence the depositors’ (customers’) behavior who are likely to have trust in that bank and decide to deposit their savings in that bank.

- Hypothesis H7: Transaction time is positively related to the decision to deposit. It is important to note that this finding can support for the hypothesis because of the Sig value of less than 0.05 and a standardized Beta of 0.130. Therefore, if transaction time increased by 1 unit, the decision to deposit will significantly increase. This finding is consistent with the study by Devlin (2005), who stated that when customers carry out activities related to money (their money as savings in the bank), they are also interested in simplicity, quickness of the transaction, and saving time. When a customer visits a branch, they want to be engaged, receive personal attention, and not have to wait around to be noticed by the bank staff. When customers need to carry out some transactions, they may have to follow many procedures. As such, when a customer feels that all these would take a long time, they will feel irritant, thereby creating a bad impression for the bank.

6. Conclusion

The banking system provides a number of important functions for the economy and is also the lifeblood and financier of the economy in each country. To achieve the goal, this research analyzes the factors affecting the choice of bank for the deposit decisions of customers in Vietnam. The study used a sample data of 250 individuals and SPSS software was used to analyze the data. The results showed that customer policy has a positive effect on customers’ deposit decisions in a bank, and this is new evidence regarding behavioral theory in the case of Vietnam. Results further demonstrated that other factors such as employee image, brand, interest rate, relative influencing, and transaction time positively impact the choice of bank for the deposit decisions of customers. However, the bank’s promotion strategies had no impact on the choice of bank for the deposit decisions of customers. Besides, employee image is the most influential factor in the deposit decisions, followed by the bank’s brand and interest rate.