Which of the following factors does not explain the rapid growth in international banking?

1. Introduction

The advent of digital services has greatly affected the operations of businesses (52% of companies of Fortune 500 have gone bankrupt or have been acquired since 2000) (State Bank, 2019). In addition, the application of the digital banking system was estimated to bring profits from 43% - 48% (State Bank, 2019). In Vietnam, the banking industry has witnessed rapid growth and intense competition among banks in recent years. To increase the competitiveness and attract customers, new utility services such as digital banking is being implemented. This has made the application of digital banking necessary for banks in Vietnam today.

Digital banking is an important service in increasing customer loyalty to the bank. A survey shows that nearly 20% of customers are willing to switch to another financial institution if their current bank does not provide online banking services (Guru, Shanmugam, Alam, & Perera, 2003). Therefore, research on the intention to use digital banking services is considered necessary (Lee, Cai, & O‟Leary, 2006). For new technology services such as digital banking, researchers often rely on the frameworks of theory of reasoned action model (TRA) (Fishbein & Ajzen, 1975), technology acceptance model (TAM) (Davis, 1989, 1993) and its variants (Kulviwat, Ii, Kumar, Nasco, & Clark, 2007; Venkatesh, 2000; Venkatesh, Morris, Davis, & Davis, 2003)

In Vietnam, although there have been many studies related to the intention to use internet banking services. However, no research has been conducted on the scope of digital banking services (digital banking services are more extensive and broader than electronic banking services). In order to ensure that commercial banks in Vietnam can switch to the digital banking business model, it is required to study the factors affecting digital banking development in Vietnam. Stemming from the above reasons, the author have chosen the topic "Factors affecting the intention to use digital banking in Vietnam".

2. Literature Review

2.1. Overview of Digital Banking

Digital banking is a transformation of all traditional banking activities and services into a digital environment (Sarma, 2017). Digital banking is a highly technologically demanding including innovation in financial services for customers and commercial customers around mobile, digital, AI and payment strategies, regtech, data, blockchain, API, distribution channels and technology (Sarma, 2017). In general, digital banking is an operating model based on a technology platform to exchange information and conduct transactions between banks and customers. This process is done through digital devices which are connected to computer software in the internet environment. Customers do not have to come to physical branches of banks to make transactions and vice versa, banks also do not have to meet with customers to complete transactions (e.g., signing documents, tracking records)

According to Davis (1993) and Venkatesh (2000), the intention to use technology services is the awareness of ability to use services of customers. Customers‟ intentions to use the service, will be influenced by several motivating factors leading to the intention (Fortes & Rita, 2016). Research on the intention to use technology services service is often anchored on technology acceptance model (TAM) and the various models developed from the TAM model (King & He, 2006) . In the TAM model, the intention to use through the lens of theory of rational action and theory of planned behavior is affected by factors such as: ease of use, perceived usefulness, attitude to service (Davis, 1989). In addition, the TAM model has also been extended to include a number of new factors such as perceived risk, trust and convenience (Fortes & Rita, 2016). This study is intended to investigate the intention to use digital banking services using the following factors: (1) ease of use; (2) perceived usefulness; (3) perceived risk; (4) trust; (5) convenience; and (6) attitude towards service.

2.2. Hypothesis

Perceived usefulness is a customer's perception of the ability to improve work efficiency, for example by saving time, when accessing services to services in multiple ways (Davis, 1993). When customers feel the service useful they have a positive attitude towards the service (Fortes & Rita, 2016) and directly increase their intention to use the service (Davis, 1993; Pavlou, 2003; Pavlou & Fygenson, 2006). At the same time, the good attitude with service will increase the intention to use customer services (Fortes & Rita, 2016). Therefore, hypothesis is proposed as follows:

H1a: Perceived usefulness has a positive impact on attitude towards the service.

H1b: Perceived usefulness has a positive impact on the intention to use the service.

Ease of use is a customer's level of trust in using the service that will bring freedom and comfort (Davis, 1989, 1993). Digital banking services make it easier for customers to access and use banking services than traditional counter services. Many existing studies show that the ease of using the service affects customers' perceptions of the service's usefulness (Phan & Bui, 2019; Venkatesh, 2000; Venkatesh et al., 2003) and their attitude towards the service (Davis, 1993; Venkatesh, 2000). Therefore, hypothesis is proposed as follows:

H2a: Perceived ease of use has a positive impact on perceived usefulness.

H2b: Perceived ease of use has a positive on attitude towards the service.

Risks are perceptions of the damages that customers may incur when using the service. The risk of losing personal information or transactions creates a barrier to electronic services (Fortes & Rita, 2016; Glover & Benbasat, 2010; Nguyen, Nguyen, Dang, & Nguyen, 2016). Therefore, reducing the perceived risk will increase the positive attitude of customers to the service. Hypothesis is given as follows:

H3: Perceived risk has a negative impact on attitude towards the service

Trust of customers is a factor showing that customers feel secure when using the service without having to care about risks or other issues (Gefen, Karahanna, & Straub, 2003; Nguyen, Nguyen, & Vo, 2019). Page and Luding (2003) think that a high level of trust is an important motivation for using digital banking services (Page & Luding, 2003). Stewart (2003) also suggests that at a sufficient amount of trust, customers find a website or an application more useful. Therefore, Hypothesis is stated as follows:

H4a: Trust has a positive impact on perceived risk.

H4b: Trust has a positive impact on attitude towards the service.

Convenience improves access to services easily with the efficiency equal to or better than other services. Convenience can be demonstrated by saving transaction time and reducing technical errors (Chen, Sun, & Xu,2016; Seiders, Voss, Godfrey, & Grewal, 2007). There are studies that show how convenient it is to find or access services. Increasing convenience will help increase customers' intention to use the service by minimizing time and reducing errors during transactions (Chang & Polonsky, 2012). Therefore, Hypothesis is stated as follows:

H5: Convenience has a positive impact on the intention to use the service.

Attitude towards the service affects the customer's decision to use the service. For digital banking services, customers who have a positive view of the service are more likely to accept it.Various studies have shown that positive customer views or attitudes influence the intend to use (Kulviwat et al., 2007; Nguyen et al., 2016; Nguyen et al., 2019; Sousa & Farhangmehr, 2018; Venkatesh et al., 2003). Therefore this study hypothesizes:

H6: Attitude towards the service has a positive on intention to use the service.

Research model and hypothesis are shown in Figure 1.

Figure 1: Research model

3. Research Method

3.1. Research Design

The questionnaire (see Tale 1) used to measure the factors in the proposed model was referenced from previous studies (Chang & Polonsky, 2012; Davis, 1993; Fortes & Rita, 2016). The time during which the survey was conducted was from 10/2018 to 1/2019. Likert 5-point scale was chosen in which: 1 – strongly disagree and 5-strongly agree.

Table 1: Questionnaire

3.2. Sample and Data

The research sample was identified as customers using digital banking services in Vietnam. Convenient sampling methods was then applied to collect data. The official survey results obtained 201 responses. This sample size was demonstrated to reach reliability according to most sampling rules (Tabachnick & Fidell, 2006). The demographic description of respondents showed that the percentage of women participating in the survey was higher than that of men (113 women made up 56.2% while the number of men was 88 comprising 43.8%). About education, a majority of respondents held college degree (155 people, at 77.1%). In contrast, there were only 39 graduates and 7 respondents of high school level, comprising about 19% and 3.5% of the sample, respectively.

Regarding income, the largest income group included people with monthly income below 5 million, accounting for the largest proportion with 75 people at 37.3%; followed by the group of over 10 million per month the group of over 10 million per month 65 people at 32.3%. The smallest was the income group from 5 to 10 million with 61 people at 30.3%. The frequency of use of individuals was mainly from 2 to 5 hours / day (88 people at 43.8%), followed by less than 2 hours per day (74 people at 36.8%); the percentage of users spending over 5 hours a day accounted for 19.4%. Main occupations of respondents were officer, freelance business and students (each accounting for more than 25%); the housewives / retirement group only made up 6% (see Table 2).

Table 2: Descriptive results

3.3. Data Analysis

Research data was analyzed using multivariate analysis methods. Firstly, scale reliability of the constructs in the research model was assessed through Cronbach's alpha coefficient , the value of which was required to be greater than 0.6 (Hair, Black, Babin, Anderson, & Tatham, 2006) and the item-total correlation greater than 0.3 (Nunally & Burstein, 1994). To assess the appropriateness of the research scales: confirmatory factor analysis (CFA) is used to convergence validity and discriminant validity. Structural equation modelling (SEM) was applied (SEM) to find out impacts of the factors on the intention to use digital banking services at the 5% significance level. CFA, critical and SEM analyses were reliable when the Chi -square / df conditions were less than 3; the value of CFI, TLI, IFI were all greater than 0.9; RMSEA‟s coefficient was less than 0.05 (Hair et al., 2006; Hooper, Coughlan, & Mullen, 2008). A construct with all factor loadings of items greater than 0.5 was considered to have convergent validity and the one having the squared root of the variance greater than the correlation with other constructs was deemed to reach discriminant validity (Hair et al., 2006)

4. Results

4.1. Evaluating Reliability

As mentioned above, the two criteria to deem a construct reliable were Cronbach‟s alpha greater than 0.6 and the item-total correlation greater than 0.3. The items with an item-total correlation coefficient less than 0.3 would be excluded from the scale and considered as unnecessary item. This item would thus not be included in subsequent analyses (see Table 3).

Table 3: Reliability test

PEU: perceived ease of use; PU: perceived usefulness; TRU: trust; RIS: perceived risk; ATT: attitude; CON: convenience; INT: intention to use service.

The results of scale reliability indicated that perceived usefulness (PU) could be reliably measured through the three items: PU2, PU3 and PU4. PU1 was excluded as its item-total correlation coefficient (0.238) was less than 0.3. The remaining constructs all were all deemed reliable using the original items items (Cronbach‟s alpha greater than 0.6 and the item-total correlation coefficients greater than 0.3). Therefore, the items in these constructs would be included in analysis in the next steps.

4.2. Analysis Results

The results of confirmatory factor analysis showed that: Chi-square / df = 1.69, less than 3; CFI = 0.922; TLI = 0.922; IFI = 0.923 greater than 0.9; RMSEA = 0.059 less than 0.08. This suggested that the data was compatible with the proposed model. The factor loadings of items were all greater than 0.5 (Table 4), so it was possible to see the convergent validity.

Table 4: General reliability and Average Variance Extracted

The results of general reliability analysis and average extracted variance (AVE) showed that the factors are the scales with load factor greater than 0.5, reaching the convergence validity. For each construct, the composite reliability was above 0.7 and AVE was greater than 50%. This indicated that the scales used for the constructs achieved the necessary reliability (see Table 4).

Table 5 showed that square root of AVE of each construct was greater than its correlation with one another showing the scales satisfied discriminant validity.

Table 5: Discriminant validity

4.3. Structural Equation Modeling

The analysis results using SEM showed that the model was suitable for the survey data (Chi-square / df = 1.84 less than 3, CFI = 0.915; TLI = 0.90, IFI = 0.916 greater than 0.9, RMSEA = 0.065 less than 0.08). After testing the reliability and validity of the constructs, structural analysis was conducted to find out the factors affecting the intention to use digital banking services. As the significance level of 5% was selected, p-values would be compared with 0.05 to see which factors have significant impacts on the dependent variable. The combined results were presented in Figure 2.

Figure 2: Structural Equation Model

Structural analysis results showed that ease of use had a positive effect on perceived usefulness, which in turn had a positive effect on the attitude towards the service. It was also found that trust had a positive impact on perceived risk and perceived risk as expected negatively affected attitude towards the service. Trust, however, had no direct effect on the attitude towards the service. Furthermore, perceived usefulness and attitude was found to positively affect intention to use, while the impact of convenience was found not significant. In other words, the results accepted hypotheses H1a, H1b, H2a, H3, H4a, H6 and rejected hypotheses H2b, H4b, H5 (see Table 6).

Table 6: Relationship between the variables

5. Discussion

The results showed that attitude towards the service and perceived usefulness had positive effects on intention to use the service, which is in accordance with the theory of reasoned action model (TRA) (Fishbein & Ajzen, 1975), theory of planned behavior (TPB) (Ajzen, 1985). Specifically, perceived usefulness reported a stronger impact than attitude towards the service on customer's intention to use. This implies that banks can take advantage of technological advances to enhance the usefulness of their services, focusing on promoting the development of their digital banking services. Customers increasingly appreciate the advantages of digital services, such as timesaving nature and diversity of services compared to performing transactions at counter.

In addition, the result that perceived risk had an indirect effect on intention to use through attitude towards the service supports the proposed hypothesis and is also found in the study of Fortes and Rita (2016). A high level of perceived risks often leads to a negative attitude towards the service, which means that poor perceptions of the information or transaction security when using digital banking services will make customers have a bad attitude to the service. Fortes and Rita (2016) also suggest customers always react negatively to issues that bring risks or damages to themselves.

In contrast, trust did not directly affect the attitude to the service but had an indirect effect through perceived risk. The negative impact of trust on perceived risk indicates when customers trust the service they feel more secure and their perceived level of risk is reduced. Research by Fortes and Rita (2016) also shows the same result. When banks gain customers' confidence in digital banking services, they no longer feel insecure when using the service, and vice versa if the customer's confidence is reduced due to the information or perceived insecurity or poor service quality, they will tend to increase the level of precautions and think about risks when using the service.

Finally, convenience reported insignificant impact on intention to use digital banking services. This result can be explained by the Kano theory positing that in case of utility services using high technology like digital banking, convenience is considered a must. Customers, therefore, no longer consider the convenience as a criterion when making their decisions to use the service. At the same time, the trust of customers does not directly affect the attitude to the service, meaning that when customers have confidence in digital banking, they will not appreciate or have a good attitude towards the service. They tend to first seek belief that the service poses no risks or harms to themselves and then have their attitude changed accordingly.

6. Conclusion

The study shows the necessity of digital transformation of traditional banking activities today. With the rapid development of technology in the financial industry, especially Fintech businesses, it is imperative that banks develop electronic banking services to facilitate easier management and operation. At the same time, the use of digital banking services enables customers to use more value-added services with flexibility, reducing the technical errors that can be encountered when doing other traditional transactions. With an extensive review of previous studies, the authors have constructed a research model to investigate the factors affecting the intention to use digital banking services in Vietnam. Drawing on the research results, several recommendations are provided to help improve the intention to use digital banking services in Vietnam.

Perceived usefulness has a positive effect on attitude and intention to use the service. Therefore, it is necessary to enhance the sense of the usefulness of customers through media advertising and consulting so that customers fully understand the benefits brought about by using digital banking services. Perceived risk has a negative impact on attitude towards the service. Therefore, banks need to build information security layers to insure customers, but at the same time the services need ease of use to avoid annoying customers.

What factors explain the rapid growth in international banking?

What are the factors that contributed to the growth of international banking over the past few decades?

What are the factors affecting banking industry?

What are the three common challenges in banking?