Bank Customer Loyalty at Post-Implementation of Automatic Exchange of Information (AEOI) in Indonesia

Introduction: Many studies have been conducted massively around the globe. Nevertheless, still, no study has been conducted to explore the impact of banking industries’ AEOI adoption on customer loyalty, Trust, Commitment, and Satisfaction. Purpose: The purpose of this paper is to investigate the effect of trust, commitment, and satisfaction on customer loyalty and, in turn, the indirect effect of satisfaction on customer loyalty at post-implementation of the Automatic Exchange of Information (AEOI). Methodology: A model is advanced and tested using partial least squares path modeling (PLS-SEM), and data were collected from a sample of 200 banking industry customers in Indonesia. Findings: The results indicate significant and positive relationships between trust, commitment, satisfaction, and customer loyalty. Moreover, it is found that satisfaction has an indirect effect in relation to trust, commitment, and customer loyalty. The findings and limitations are discussed, and recommendations for the policymakers and researchers are also provided. The showed that


INTRODUCTION
During the last three decades, we have witnessed a significant expansion of the financial sector (Barattieri et al., 2021). The existence of banks is important of life for economies around the world, many of the economic wheels are supported and run by them. Likewise, in Indonesia, the bank holds a very important control, especially for developing countries that are in the development stage in various sectors. Banks provide services to customers in the financial sector, such as savings, loans/credit, deposits, and others (Roziq et al., 2020).
In line with advancing technological developments, the Indonesian finance ministry has adopted the Automatic Exchange of Information (AEOI) as an effort to overcome the financial crisis. Through the implementation of AEOI, the directorate of taxes has the authority to access financial information for state tax purposes (Darmanti & Mangkan, 2020). The implementation of the Automatic Exchange of Information (AEOI) will benefit countries, which is a form of cooperation between international taxation and banking that has been implemented by several countries (Darmanti & Mangkan, 2020). There are 105 countries at the end of 2020, including Indonesia, that have adopted the AEOI in avoiding financial crises and improving tax administration in their countries (OECD, 2020). AEOI is used as a medium for exchanging information related to banks' customer data around the world periodically, systematically, and continuously to taxpayer residents of countries. Implementation of this regulation, the Government aims to increase tax revenue for state income and to minimize taxpayers who avoid taxes.
Until now, the Indonesian Government has not guaranteed the security and confidentiality of customers' personal information. When the confidentiality of customer data is not guaranteed, the impact that may occur is reduced customer distrust where the customer keeps money because of the basis of trust. The continuing situation will lead to a financial crisis and has a significant impact on the levels of trust that citizens place in the system and its institutions. It is relevant to a finding showing a significant fall in confidence in the majority of citizens in the bank institutions (Roth, 2009). The citizens will rush money as a result of mistrust of banking institutions, which can cause many banks to collapse. When the confidentiality of customer data is not guaranteed, the impact that may occur is reduced customer distrust, in which customers save money because of the basis of trust.
Regarding the confidentiality of customer data that can be accessed by the state, customers have to prepare for data security risks and interest deductions to be transferred to taxes. Based on the results of account growth data by the Indonesia Deposit Insurance Corporation in 2020, the number of commercial bank customers is 344.54 million conventional and 33.77 million sharia, so there are millions of banks' customer information data that need to be maintained security and confidentiality.
Customer loyalty is a commitment to do a business with an organization, to purchase its goods or services repeatedly, and recommend them to colleagues (Mellroy & Barnett, 2000). Oliver (1999) has described loyalty as a deeply held commitment to rebuy or re-patronize a preferred product/service consistently in the future. Customer loyalty is considered the backbone for firms as it is a way of sustaining and increasing sales. As well, it plays an important role in increasing profits, as it is obvious from previous research the cost of obtaining new customers is much higher than to serve current customers (Husnain & Akhtar, 2015). Customer loyalty is considered a key business strategy and the greatest asset of each company that guarantees the company's success and improves its level of profitability. In today's fastpaced world of constant change, creating and maintaining customer loyalty need more attempts (Abtin & Pouramiri, 2016).
Customer loyalty can be achieved by identifying the customer needs, managing them, and fulfilling them (Maikayi et al., 2011). Customer loyalty has been a real concern for banking professionals because of serious rivalry and higher client expectations (Rasheed et al., 2015). Building customer loyalty is the only way of building sustainable competitive advantages. Customer loyalty is one of the most essential for marketing theory and practice because loyalty impacts financial performance, customer attitudes, and customer behavior and repurchase behavior for the product or service. Moreover, based on the previous research, to overcome the decline in performance, constant reviews are done on the creation of customer loyalty (Chiguvi & Guruwo, 2017).
Studies have been conducted to explore factors behind loyalty, such as Satisfaction, Trust, and commitment, but the results were inconclusive (Bricci et al., 2016;Herman, 2017;Masturoh et al., 2020;Zhong & Moon, 2020). It is important for institutions to improve their responsibility in developing team skills, empowering employees and the company in a sense to fulfill commitments, and performing the co-creation of value (Bricci et al., 2016). Sutanto and Djati (2017) found the direct and indirect effects of Trust, Satisfaction, and Commitment on Customer Loyalty at the Alfamart Retail in Indonesia.
Customer satisfaction has become the center of service marketing and operations management because it is a general proposition to be a significant antecedent of customer retention and repeat sales by positive word of mouth. Customer satisfaction is described as satisfying one's desires or meeting one's criteria, or we can say that delighted/satisfied customers have a specific feeling or mentality towards a product or service it has utilized (Maikayi et al., 2011). Consumer satisfaction is characterized as a client's general assessment of the performance of an offering to date (Johnson & Fornell, 1991).
The effect of satisfaction on loyalty is inquired about in wide terms. Numerous discoveries show that if the customer/consumer is satisfied, then they are loyal, and if they are dissatisfied, their loyalty is not ensured. It is utilized to clarify loyalty for behavioral purposes (Heskett et al., 1994). Thus, management ought to pay a unique spotlight on customer satisfaction, and for it, service quality assumes a crucial part. Satisfaction has an important characteristic that must be taken into record when shaping the general customer loyalty towards their service suppliers (Rasheed et al., 20015).
Several research confirmed the existence of the relationship between trust, commitment, satisfaction, and customer loyalty. Firstly, in terms of services, trust is the belief held by a customer that the service provider will provide the service that meets customer needs, and trust consists of two aspects: perceived credibility and benevolence (Mosavi & Ghaedi, 2012). In other words, the belief of one party with the other with sincerity. Trust is important for a bank because, without trust, it is difficult to build a strong foundation. Studying Islamic Banks in Indonesia, Farasky (2019) found that there was a positive effect on trust and customer loyalty. By distributing questionnaires to the local banks in Indonesia, Lestari (2019) confirmed that customer trust has a positive effect on customer loyalty.
Secondly, commitment is maintaining the relationship between the bank and customers, and this is done for the common interest. This condition shows the psychological fundamentals to maintain a long relationship between customers and banks, and here the company must strive to build a mutual relationship. This commitment also leads to the intention of customers to use products or services in the future to maintain relationships (Chai et al., 2015). The empirical evidence showed that there was a positive influence of customer commitment on customer loyalty (Almana et al., 2018;Muniroh, 2018).
Thirdly, satisfaction is a measure of how the products and services provided by a company meet or exceed customer expectations, and satisfaction metrics indicate how successful an organization is at providing products and/or services to the marketplace (Fornell, 1992). Satisfaction is feeling happy or disappointed someone who arises because of comparing the perceived performance of the product (or result) to their expectations. This term is often used as a measure of how the products and services provided by the company meet or exceed customer expectations. In the service environment, customer satisfaction has been seen as a special form of customer attitude (Wantara & Tambrin, 2019).
Customer satisfaction is a crucial factor for bank success, and it has the possibility to influence customer loyalty. From a theoretical perspective, it is very important to investigate which factors influence customer satisfaction. Further, for this reason, academics and practitioners have emphasized the concept of customer satisfaction, defined as the favorability of the individual's subjective evaluations of the outcomes and experiences associated with his or her consumption activities (Al-Msallam, 2015). It was also found that customer satisfaction has a significant and positive relationship with customer loyalty (Rasheed et al., 20015).
The application of Automatic Exchange of Information (AEOI) is expected to provide benefits for Indonesia, such as access to financial information for taxpayers who still invest in the country, as well as all taxpayers in Indonesia (Darmanti & Mangkan, 2020). The Automatic Exchange of Information (AEOI) is a product of a common reporting standard as a result of Indonesia's ratification of the Convention on Mutual Administrative Assistance in Tax Matters. The convention's purpose is to prevent transnational tax evasion and money laundering. With the existence of automatic exchange of information systems throughout, members of the convention can enact and enforce its laws (Tedja et al., 2019). The system implementation's purpose is to increase the effectiveness of tax collection and its utilization. The jurisdiction in which AEOI is being used is also regarded as one effort for international tax compliance and decreasing the number of tax havens in the world (Tedja et al., 2019).
Although many studies have been conducted massively in around the globe, nevertheless still, no study was conducted on exploring the impact of banking industries' AEOI adoption on customer loyalty, trust, commitment, and Satisfaction in Indonesia. Based on the theoretical and empirical background, it is urgent to confirm whether the adoption of Automatic Information (AEOI) in banking industries in Indonesia will have a positive effect on customer loyalty. This research will be focused on the influence of trust, commitment, and satisfaction on customer loyalty in the bank industries in Indonesia.

Research Design
This research was conducted in the form of causal associative with the analytical approach using quantitative methods, which is data collection using survey methods. Associative research is research with the nature of questioning the relationship between two or more variables (Draper, 2004). Meanwhile, quantitative analysis is a measurement and analysis with numbers and statistics to obtain results (Daniel, 2016). The model presented here is in two parts, the measurement model (the outer model) and the structural model (the inner model). The measurement model concentrates on the association between latent variables (constructs) that cannot be directly determined and measurable (observed) indicators (Henseler et al., 2009). The structural model focuses on the hypothesized associations between the constructs (Hair et al., 2013). All of the items used in this research are considered to be reflective measures.
This research was conducted in Yogyakarta Province in Indonesia, involving banking industries' customers. In total, 250 questionnaires were distributed to the customers of banking industries in Yogyakarta, Indonesia, but finally, only 200 questionnaires were completely filled and can be further analyzed. To exclude any potential response bias, the first 61 responses were compared to the last 139 responses using the independent samples t-test (Hair et al., 2010). The analyses showed that for all constructs, the difference in the means of earlier and later responses was not significant.
Commercial banks have the broadest meaning of institutions in charge of carrying out business activities, both conventional and sharia, which is carrying out their activities provide services in all forms of payment (Roziq et al., 2020). The commercial bank also means bank in its business activities carried out conventionally or based on sharia principles. This is stated in Bank Indonesia Regulation Number 11/15/ PBI/2009. Demography has an important role in determining customer behavior in transaction activities. Demographic characteristics that have the potential to influence customer behavior in the Indonesian area are gender, age, and income (Vilèeková and Sabo, 2013;Puška et al., (2018). Based on BPS 2020 data related to the Yogyakarta Province Population Census for the male gender 1.82 million (49.55%) and female 1.85 million (50.45%) where females dominate. Meanwhile, age is dominated by generation X (40 -50 years), Z (8 -23 years), and millennials (24 -39 years).

Measurement of Variables
Data collection was carried out by distributing questionnaires in January 2021. This study used a questionnaire so that the respondent would answer according to the conditions experienced by providing written answers and a checklist ( √ ). Measurement of the independent variable (Trust, commitment, Satisfaction) and the dependent variable (customer loyalty) was used with a 4-point Likert scale based on Table  1. The original Likert scale is a set of statements (items) offered for a real or hypothetical situation under study (Joshi et al., 2015). Here, participants were asked to show their level of agreement (from strongly disagree to strongly agree) with the given statement (items) on a metric scale, revealing the specific dimension of the attitude towards the issue, hence, necessarily interlinked with each other. Repurchase of products or reuse of services 6 The habit of using the products or services Like the product or service

Choose the company or brand
Company or brand is the best

Data Examination
PLS-SEM was used because its formal premises make it applicable to a range of flexible applications (Hoeck et al., 2010). Barclay et al. (1995) stated that the minimum sample size should be no less than ten times the number of indicators related to the most complex construct of the model tested. Therefore, the minimum sample size required for this study is 100 (n = 10 items related to trust).
Normality tests were not necessary since PLS regression analysis does not require the data to be distributed normally (Chin et al., 2003). Two exogenous (independent) variables and two endogenous (dependent) constructs were used in this research. The number of items used to measure each of the constructs ranged from four to ten.

RESULTS AND DISCUSSION
Internal consistency. It shows the degree of interrelatedness between the items, whereas homogeneity refers to the unidimensional of a set of items. It is a crucial process for a researcher in choosing the instrument or considering developing the new one to achieve the research objective as well as the quality of the instrument (Nawi et al., 2020). Furthermore, Nawi et al. (2020) cited that the reliability test is one of the most significant components of test quality which is involved with the reproducibility, consistency, or an examinee's performance on the test. The most common method used to evaluate internal consistency is composite reliability (CR) (Hulland, 1999). The table shows the CR for the reflectively measured constructs used in this study. This shows that the criterion is met, as the CR is above 0.70 ( Table 2).  (Agarwal, 2013). Convergent validity is used to specify that causal indicators from a measurement model should explain a significant proportion of variance from the latent variable that they measure (Wang et al., 2015). To establish convergent validity, most researchers consider the outer loadings of items and the average variance extracted (AVE). (Hair et al., 2014). Table 3 shows the convergent validity for reflectively measured constructs in terms of outer loadings and AVE. Discriminant validity. Discriminant validity is used to specify that these same indicators should explain a much lower proportion of variance from other latent variables. That is, indicators that are associated with the target latent variable will explain much more variance of that latent variable, and those indicators should not explain a large amount of variance of other latent variables relative to the target latent variable (Wang et al., 2015). Further, discriminant evidence is particularly important because it indicates that these indicators do not belong to other latent variables. Discriminant validity depends on whether each construct shares more variance with its measures than with other constructs in the model (Chin, 1998;Hulland, 1999); thus, the square root of the AVE must exceed the construct inter-correlations in the model. As Table 4 reveals, the psychometric properties of the instruments were adequate to support our interpretation of the structural model.  (2009), after validating the measurement model, the structural model represents the hypothesized relationships between constructs. The path coefficient (β) shows the power of the relationship between the latent variables (Idris and Mohammad, 2014). Overall, the results show that all path coefficients were statistically significant at 0.05 levels ( Table 5). As Table 5 shows, Trust, Satisfaction, and Commitment were positively and significantly correlated with Customer Loyalty. This provides support for H1, H2, and H3. To examine the indirect effect, the PLS algorithm and bootstrapping procedures with 1000 resample were run on the full model to obtain the path coefficient and the significance level of these paths. Table 5, all paths were statistically significant, which provides empirical evidence of an indirect effect (Preacher & Hayes, 2008). Therefore, it can be concluded that the indirect effect of satisfaction is statistically significant in the relation between Commitment and Customer Loyalty. The bootstrapping procedure demonstrates that the indirect effect β4 = 0.071 was significant with at-values of 2.221, which confirmed the indirect effect of satisfaction in the relation between Commitment and Customer Loyalty. The same conclusion can be made for the satisfaction in the relationship between Trust and Customer Loyalty with β5 = 0.149 and at-values of 3.273 (Table 5). This provides support for H4 and H5. To assess the amount of variance explained in the endogenous constructs (R 2 ). According to Cohen (1988), R 2 values of 0.2, 0.13, and 0.26 are considered small, medium, and large, respectively. In this study, R 2 values for Satisfaction and Customer Loyalty are 0.576 and 0.582, respectively ( Table 5). This indicates that the ability of the model to explain variance was substantial (Cohen, 1988).

As illustrated in
The results show that trust has a negative effect on customer loyalty of Banks in Indonesia post-implementation of the Automatic Exchange of Information. It supports previous research findings from Afifi and Amini (2018), Wemben et al. (2020), and (Nuryadin et al., 2021). Nevertheless, it contradicts the theory of trust, revealing that one of the factors forming a trust, namely ability, will provide a guarantee of satisfaction and security from the company. This means that building customer trust is important in order to increase customer satisfaction. The existence of a negative influence is due to the fact that some customers have not perceived and understood well, especially in this discussion on indicators of goodness.
In building customer loyalty from a bank, trust has an important role, and customer trust can grow through the quality offered by banks, to build this trust takes a long time because customers need to intensely obtain quality services and products from banks. Besides, high trust from customers will increase their expectations of customers for the quality of service from banks. The results showed that commitment has a significant positive effect on customer loyalty of Banks in Indonesia post-implementation of Automatic Exchange of Information. This is under research by Almana et al. (2018);and Bricci et al. (2016). Commitment is maintaining the relationship between the bank and customers, and this is done for the common interest. This condition shows the fundamental psychological to maintain a long relationship between customers and banks, and here the company must try hard to build a mutual relationship. Where there is a commitment that is held by the customer, there will be a desire for the customer to maintain a good relationship with the bank in the future.
Satisfaction has a significant positive effect on customer loyalty of Banks in Indonesia post-implementation of the Automatic Exchange of Information. It is relevant to the previous research conducted by Wijayanto (2015), Razak et al. (2017), Almana et al. (2018), andBracci et al., 2016. Satisfaction is a form of feeling or fulfilling the expectations of customers for what they get from a product or service. In this study, customers are satisfied with the performance of Indonesian banking, which is obtained from the quality of products, services, brands, prices, and costs.
However, the study has a number of limitations. Firstly, an online questionnaire via google form was used so that the researcher could not maintain objectivity so that the data could be subjective. So, it is necessary to have an interview method for data collection. Secondly, the interview method that was originally planned could not be implemented because it was hampered by Social Restrictions, which were extended until 8th February 2021. Finally, focusing on the three variables trust, commitment, and satisfaction will limit portraying and predicting the loyalty of customers of Banks in Indonesia, especially since the implementation of Automatic Exchange of Information.

CONCLUSION
Based on the results of the research and discussion, in summary, the findings show that Trust and Satisfaction have no significant negative effect on customer loyalty while commitment has a positive and significant effect on customer loyalty. Simultaneously, trust, commitment, and satisfaction have a positive and significant effect on customer loyalty. Meanwhile, the determination coefficient test shows that the adjusted R Square of 0.526 or 52.6% of the loyalty variable can be explained by the variables of Trust, commitment, and Satisfaction. So, it can be explained that the independent variable influences the dependent variable. The results suggest that Commercial Banks in Indonesia need to improve services related to increasing the trust of customers. Trust is important in increasing customer loyalty. For example, from the aspect of the ability, it guarantees satisfaction, security, and comfort in transactions; the benevolence of service quality, safety assurance, and product variety; integrity in maintaining and fulfilling all obligations that have been made; completing their responsibilities not causing a loss to customers, and providing the best advice to customers. Concerning customer loyalty, by building commitment and customer satisfaction, it is necessary to maintain customer loyalty. Because customers tend to maintain good relations and use the services of Banks in Indonesia when banks continue to maintain services that can fulfill responsibility and