Accelerating Digital Banking Transformation through User Acceptance: a Study on the Cultural Impact of Millennials

____________________________________________________________ This study aims to examine a conceptual model that explains the main factors that influence the behavioural intention of the millennial generation in their acceptance of the use of digital banking. The conceptual model is based on the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), with experience as a moderating variable. The modelling was carried out using the Partial Least Square Structural Equation Modeling (PLS-SEM) equation to analyze the data collected from the questionnaires filled out by the respondents. The research results show that behavioural intentions are significantly influenced by performance expectancy, effort expectancy, facilitating conditions, hedonic motivation, and habit. Social influence and price value are not factors that contribute to the cultural behaviour intention of the millennial generation towards accepting the use of digital banking. In line with hypothesis testing, moderation testing with experience as a moderating variable that has no influence is social influence and price value.


Introduction
The Covid-19 pandemic amid the Industrial Revolution 4.0 has caused a slowdown in various sectors of life, including the economic sector so that people's activities and mobility are limited.Various policies are set by the government and have an impact on changes in people's orientation and lifestyle behavior in carrying out economic activities.These conditions encourage the acceleration of digital technology transformation in society from the physical economy to the virtual economy.Information technology was developing through digital transformation has caused to the phenomenon of the digital economy (Tsindeliani et al., 2022).Tight competition, increasing value for customers as well as efficiency and effectiveness encourage banks to use technology applications to reduce human encounters (Alalwan, 2018;Khan et al., 2017).Technology dramatically contributes to the business environment by creating new systems that can improve banking performance by providing high-quality services to customers (Al-Hawari et al., 2009;Yoon & Barker Steege, 2013).
The National Non-Cash Movement (GNTT) launched by the government in 2014 through Bank Indonesia (BI) is aimed at improving the cashless society ecosystem through a savely payment system, effective, efficient, and smooth to minimize obstacles that arise due to payments using cash.Demands for adaptation and innovation of payment systems towards digital technology that BI issues the Indonesian Payment System (SPI) 2025 blueprint and the 2020-2025 Indonesian Banking Development Roadmap (RP2I) which supports banking digitization as a key institution in the digital financial economy sector through open banking and the use of digital technology so that more contributive, competitive and resilient.The use of digital technology will save costs, and add value to customers while increasing banking competitiveness (Owusu Kwateng et al., 2019).Phenomenon effect motivated retail banks to beginner digital banking systems into their business operations to improve performance (Owusu Kwateng et al., 2019;Tsindeliani et al., 2022).The demands for digital acceleration are increasingly emerging because they are driven by public expectations of applicable financial services so digital banking transformation becomes a must (Shoheib & Abu-Shanab, 2022).To support the acceleration of digital transformation, elements of society, processes, and technology are needed.
Indonesia the country branding as the fourth largest population in the world has a population of 270.2 million people (Badan Pusat Statistik, 2020).Of this population, internet user penetration will reach 204.7 million in 2021, an increase of 15.5% from 2020 (Kementerian Komunikasi dan Informatika, 2022).These conditions encourage financial sector innovation to create internet-based financial services.As of the third quarter of 2021, Bank Indonesia has reached a 46.72% (YoY) digital banking has increasing transactions (Bank Indonesia, 2021).However, in developing countries, the lack of public awareness of using internet banking is a major obstacle to using technology (Sharma et al., 2017).The 2021 Indonesia Digital Literacy Status Report released by the Ministry of Communication and Information together with the Katadata Insight Center (KIC) stated that digital banking services are still less popular than conventional services.Of the 10 thousand respondents spread across 34 provinces in Indonesia, ATM and bank account users are 73%, while mobile banking and internet banking users are 13.3% and 7.7% respectively (Katadata, 2021).
Financial Services Authority Regulation Number 12 / POJK.03/2018 states that Digital Banking services allow customers to get various services owned by banks.The forms of digital banking services are mobile banking, internet banking, SMS banking, phone banking, and mobile banking which are used by the public as needed.One of the breakthroughs in digital banking is internet banking, thorugh offers innovative dimension of provide various financial services that replace the role of human meetings ((Akhlaq & Ahmed, 2013).The maturity of digital banking in Indonesia is assessed based on the DMAB (Digital Maturity Assessment for Bank) dimension which shows a ratio of 50% which means that the dimensions of data, technology, collaboration, and customers are not optimal (www.ojk.go.id).
Research by Bain, Google, and Temasek (2019) shows that 47 million Indonesians have limited access to banking services.Therefore, a marketing strategy must be carried out to gain a larger market share through the intention to use digital banking.Digital banking services that are still at the development stage in a country may be because these services are not being used properly by the public (Hilal & Varela-Neira, 2022).The successful use of internet banking requires customer acceptance of the technology.Digital banking does not only depend on the availability of applications but also attracts customers to accept it in full (Al-Hawari et al., 2009).Inviting customers to change traditional behavior to digitally transform is not an easy process, so it is necessary to study and understand the factors that hinder or encourage public acceptance of new technology.
The acceleration of digital transformation is successfully carried out by the banking sector if the acceptance of the technology is in accordance with the needs of business processes as well as the needs and characteristics of consumers so that they can provide benefits that are in line with consumer desires and needs (OJK, 2021).The UTAUT2 model built Venkatesh et al. (2003) is in line with digital maturity (digital maturity level) because the construct built in UTAUT2 also measures various people's behaviors in digital maturity summarized in the customer insight aspect which studies behavior, preferences, and needs.consumers as well as customer trust and perception related to the perceptions and expectations expected by the community.70% of Indonesia's population are in the productive age and are the most prospective consumers (15-64 years).In more detail, Indonesia's demographic structure is dominated by generation Z (8-23 years), the millennial generation (24-39 years), and generation X (40-55 years).This study focuses on the millennial generation of Central Java because of its large number.The millennial generation has adaptive characteristics, is close to the use of technology, and is a productive age so this technology is popular among these people.Previous research has been conducted to study the determinants of the acceptance of technology.Some models are used to predict and explain behavioral intentions in accepting new technologies, one of which is the Technology Acceptance Model (TAM) (Tétard & Collan, 2009;Venkatesh & Davis, 2000).In another theory, the importance of the theory of acceptance and use of technology through UTAUT2 (Unified Theory of Acceptance and Use of Technology 2) is highlighted by (Venkatesh et al., 2012) building seven kinds of constructs.Thus, it is necessary to test the application of UTAUT2 in different research contexts.On the other hand, research on the factors that affecting the millennial generation's intention to accept the use of digital banking has not been studied empirically and extensively.The addition of experience as an additional construct and moderator in this study was carried out to becoming increasingly focused on the focus and scope of its application on technology acceptance.Therefore, this research examine investigate indicate as the factors that influence the area millennial generation's intention to accept digital banking in Central Java.

Hypothesis Development
This research as in UTAUT2 theory will conceptually explain technology acceptance (Venkatesh et al., 2012).Figure 1 describe the main constructs in UTAUT2 are Performance Expectancy (PE), Effort Expectancy (EE), Social Influence (SI), Facilitating Conditions (FC), Hedonic Motivation (HM), Price Value (PV), and Habit (HB) will be used as a predictor for determining the intention to accept digital banking technology.The conceptual model in UTAUT2 is adapted from the concept created by Venkatesh et al. (2012) which consists of:

Performance expectancy (PE)
Performance expectancy is conceptualized as the benefits (effort and saving time, accessibility, convenience, customization) that can be obtained from the use of an innovation.Performance expectancy is considered as one of the most influencing drivers of information system adoption (Venkatesh et al., 2003).performance expectations are considered to have the most influence on behavioral intentions because it is believed that the system will help users to achieve and improve their performance (Venkatesh et al., 2003;Khan et al., 2017).Customers will use online banking when they know that banking will improve its performance.

Effort expectancy (EE)
Effort expectancy is conceptualized as the level of ease of individuals in using the system so that their work becomes more effective and efficient.Effort expectancy is defined as the degree of ease associated with the use of system that elaborated effort expectancy has been based on three existing models (Venkatesh et al., 2003).Martins et al. (2014) concluded that effort expectancy has influence on behavior intention.Therefore, the following is proposed: H2: Effort Expectancy has a positive influence on the behavior intention of digital banking H2a: Experience moderates effort expectancy on behavior intention in digital banking

Social influence (SI)
Social influence is the extent to which an individual feels that significant others believe that he must be use a new system.The impact of this social influence can occur due toward compliance, especially in the early stages of experience to encourage the use of technology (Venkatesh et al., 2003).Martins et al. (2014) concluded that social influence is the most important factor in the influence of behavior intention.

Facilitating conditions (FC)
Facilitating conditions are defined as the extent to which a person believes the organizational and technical infrastructure exists to support the use of the system (Venkatesh et al., 2003).One of the driving factors for online banking users is the ease of conditions.In previous research, facilitating conditions had a positive influence on behavioral intentions and technology adoption.Khan et al. (2017) facilitating conditions are the extent to which an person feel believes toward the organizational and technical infrastructure is in place to support the use of the system.

Hedonic Motivation (HM)
Hedonic motivation is a predictor of behavioral intentions and is defined as the pleasure obtained from using a technology Venkatesh et al., (2012).According to Khan et al. (2017), acceptance of online technology depends on the entertainment provided by the technology.Typically, when facing a difficult work environment, social pressure, and other mental stress, customers are more willing to pay attention to technology that provides fun and entertainment.Hedonic Motivation is conceptualized as a feeling of joy obtained by using technology.

Price Value (PV)
Price Value is conceptualized as a consumer's cognitive exchange, namely the comparison between the benefits obtained with the costs incurred in adopting and accepting new technology.In an organizational context, employees should not pay the price and consumers will pay for the use of online banking services.Consumers will pay a price for the benefits they expect to derive from using a service or product.Price value is defined as a consumer's cognitive tradeoff between the perceived benefits of an application and the monetary costs of using it (Venkatesh et al., 2012).Price Value has been reported to be positively related to behavior intention (Khan et al., 2017).

Habit (HB)
Price Value is defined as the extent to which a person tends to perform behavior automatically because of the learning process.Automatic behavior is formed because of knowledge and skills from time to time.Habit is defined as the extent to which people tend to perform a behavior automatically due to learning (Khan et al., 2017).Habit is one of the predictors that influences behavioral intentions.Habit differs from experience in two ways.First, habit formation does not necessarily require experience, and secondly experience does not provide the same level of habit depending on the level of use of the target technology (Venkatesh et al., 2012).

Methods
This study will test the hypothesis by using statistical data tests involving independent variables in the form of constructs in the UTAUT2 model, namely performance expectancy, effort expectancy, social influence, and facilitating conditions.hedonic motivation (hedonism motivation), price value (price value), habit (habit) with moderator variable experience (experience) and the dependent variable is behavioral intention in accepting digital banking.This research includes quantitative research that uses the type of explanatory research.This study focuses on knowing the factors that influence the millennial generation's intention to accept digital banking in Central Java.
The population consists of the millennial generation of Central Java digital banking service users who come from various educational, social, economic, and cultural backgrounds with a population of 9.1 million people based on BPS population census data in 2020.In sampling, this study uses probability techniques sampling with a simple random sampling method.Determination of the sample is done by using the Slovin formula with a total sample of 100 respondents.The sample in this study has sample criteria, namely the millennial generation who have used one of the digital banking services.
This study uses primary data in the form of online questionnaires distributed through social media.The indicators and items on the questionnaire were adapted from seven UTAUT2 constructs which were measured using a Likert scale of 1-5 from disagree to strongly agree.Data analysis was performed using Partial Least Square Structural Equation Modeling (PLS-SEM) because it can be used to explain theories in exploratory research.The PLS analysis test used includes model specifications, and evaluation of the outer model (convergent validity, discriminant validity, and composite reliability tests).Testing the inner model by looking at the value of R2.Hypothesis testing is carried out by looking at the magnitude of the structural FC1 .I have the resources necessary to use mobile Internet.FC2.I have the knowledge necessary to use mobile Internet.FC3.Mobile Internet is compatible with other technologies I use.FC4.I can get help from others when I have difficulties using mobile Internet.Path coefficient and the stability of the estimate evaluated using the t-statistical test obtained through the bootstrapping test (Hair et al., 2014) MRA Formulation

Respondents Profile and Characteristic
Respondents accepted in this study were 100.With male respondents demography of 68% female and 32% male.This shows that more than half of the respondents are female.From descriptive statistics, it can be seen that the 25-30 age group occupies the largest percentage of 54%, followed by the 31-34 age group 28%.
Where the largest respondents are respondents with the age characteristics of the millennial generation.The distribution of respondents according to the level of income per month is dominated at the level of 4,500 -5,000 as much as 63%, followed by the level of 6,001 -6,500 as much as 11%.For the educational background owned by the millennial generation respondents, most of them are at the bachelor level, while the experience of using internet banking in the millennial generation is quite high, namely at the level of 2-3 years amounting to more than 50%.

Structural Equation Modelling (SEM) Analysis
In this study, a two-stage approach was adopted in the SEM process, namely the evaluation of the measurement model and the evaluation of the structure of the model.

Evaluation of the measurement structure of the model Convergent Validity
In table 2 can be seen that the loading factor value of each question indicator in the questionnaire shows a valid result, namely > 0.7.Statements in the questionnaire can be used.While the value of average variance extract (AVE) as a measure of the variable.Variables that can meet the validity requirements must have a value > 0.5. in this case all variables complete the AVE standard.Table 5 shows the statistical results that the structural model fit index can be seen at the NFI value of 52.9%.The NFI value aims to test how well the model to be studied is.The NFI value has a value above the threshold so that it can be interpreted that the structural model is quite fit for the data.The goodness-of-fit test in this study is 0.067, indicating that this study's model is good (Hair et al., 2014).The goodness-of-fit test is used to examine the inner framework of the model in a study.PR shows a good value if it has a value above 0, in this study the relevant predictive is 0.757.so that it can be interpreted that the evaluation of the structural model in this study supports the existing concept.In addition, the evidence of the main statistical test results that support the conceptual model can be explained by an R Square value of 0.905, which means that it has a value of 90.5% of the behavioral intention variance (BI) is influenced by 8 factors that are endogenous variables.Regarding the path coefficient analysis, and the coefficient value of the path that ends in behavior intention shows the results of Performance Expectancy (γ = 0.32, p = 0.023), Effort Expectacy (γ = 0.30, p = 0.002), facilitating condition (γ = 0.26, p = 0.003), Hedoning Motivation (γ = 0.41, p = 0.013), Habit (γ = 0.22 p = 0.003), shows a value that has a significant influence in a positive direction.In contrast to the path of influence of Social Influence on Behavioral Intention (γ = 0.013, p = 0.243), the effect of price value on behavioral intention (γ = 0.12, p = 0.283) was found to be insignificant.In the experience variable model, the moderating variable is PE, EE, FC, SI, HM, PV, HB on Behavior Intention.In the hypothesis with the moderating variable Experience, from PE (γ = 0.353, p = 0.001), EE (γ = 0.617, p = 0.003), FC (γ = 0.243, p = 0.006), HM (γ = 0.244, p = 0.003 ), HB (γ= 0.375, p = 0.004) had a significant effect on behavior intention (BI), while experience was unable to moderate social influencing (γ = 0.041, p = 0.332) and price value (γ = 0.150, p = 0.375) on behavior intentions.Can be seen in Figure 2 The stages of evaluation of the measurement model in detail can be seen in table 6 (discriminant validity, convergent validity, and construct reliability) have been carried out with consistency results that support the conceptual model under study.In addition, it is also necessary to assess the conceptual model through the determinant factors that affect behavior intention around 89%.In this case, R2 affects behavior intention in accordance with the UTAUT2 conceptual model that was built, with 21% of behavior intention contributed from factors outside the UTAUT2 conceptual model.In this case, it means that the UTAUT2 conceptual model has a very strong influence on the formation of behavior intention.This is in line with Martins et al. (2014) by approving the perceived strength of the UTAUT2 construct (PE, EE, SI, FC, HM, PV, HB) on the extracted R2 value for behavior intention in accepting the use of digital banking.In table 7 there is a summary of the significance of the proposed research hypothesis.In the table, it can be seen that all the hypotheses proposed in this study are supported except the hypothesis H3 (Social Influence on behavioral Intention) and H6 (Price Value on behavioral intention), H3a (Experience as a moderator between Social Influence on behavioral Intention), H6a (Price Value of behavioral intention).
Table 6 shows that the factor that has the biggest influence is Hedonic Motivation on behavioral intention to accept the use of digital banking.Hedonic Motivation in the millennial generation is the most contributing factor in influencing the use of digital banking.Hedonic motivation is a reference for one's enjoyment of using technology (Rodrigues et al., 2016).concept of pleasure becomes a pleasure for the millennial generation with its curious character as an innovation in behavioral intention to use technology in the form of digital banking (Masa'deh et al., 2016).The thing that has a big contribution after hedonic motivation is performance expectancy.PE has a predictor of acceptance of the use of digital banking in the millennial generation.PE refers to how much digital banking activities are useful and useful in daily life (Tamilmani et al., 2021).The millennial generation, who has a pragmatic and realistic life point of view, will be very open to digital banking which has an impact on the benefits and utility of the technology (efficiency, accessibility, adaptability, convenience) (Dwivedi, Y. K.,et al, 2017).The millennial generation will rationally use the level of benefits and utility obtained in the use of digital banking, if it is felt that the benefits and utility are higher, then performance expectancy becomes the main driving Lin and Chang (2011) factor for adopting behavioral intention of digital banking among the millennial generation.Effort Expectacy is projected as the ease of using digital banking.Ease of contributing to behavior intention to use digital banking on the principle of benefits due to the convenience offered (Rodrigues et al., 2016).This is supported empirically by the statement of (Al-Somali et al., 2009)), which states that access to greater benefits is obtained from the ease of technology obtained.Facilitating conditions can be defined as the extent to which a person trusts the infrastructure and technology to support the system.The relationship between facilitating conditions and behavior intention has been proven by many previous studies Migliore et al. (2022) with the statement that with supporting facilities there is a higher interest in behavioral intention to use digital banking.Behavioral intention is higher because of the attributes of convenience, effectiveness and efficiency.While habit has the conceptual meaning of a person in carrying out activities automatically because he is trying to learn.
Habit or habit becomes an adoption of behavioral intention in terms of technology literacy.The millennial generation with its easy-to-adapt characteristics is not surprisingly quick to adopt habituation towards behavioral intention of digital banking (Vinerean et al., 2022).
The social influencing hypothesis seen in table 6 has no significant effect on behavioral intention.Social influencing which is articulated as social interaction with the environment and humans.In this study, it did not affect behavioral intention among the millennial generation.In the characteristics of the millennial generation, they lack a strong social awareness.Lack of social interaction causes predictors of social influence to have less effect on decision making on behavioral intentional adoption of digital banking usage Schmitz et al. (2022) due to the lack of strong social and environmental influences.The millennial generation is more likely to interact with smartphones than to interact socially with the environment.So further steps are needed from the bank to prove the reliability of services or promotions through internet media.The same thing with the perceived value hypothesis which is not significant to the behavioral intention to use digital banking in the millennial generation.This is because the value in exchange for the perceived benefits and monetary value incurred.Perceived Value is presented as time and effort instead of paying a fee.The millennial generation has many considerations of the ratio of utility received to the amount to be paid.When the ratio is not balanced, then perceived value has no significant effect on behavioral intention to use digital banking.Or is it burdensome for the millennial generation?

Moderation by Millennial Experience
The meaning of table 1 regarding the demographics of respondents, there is a distribution of respondents on experience using digital banking.Among respondents with millennial age, more than 50% have more than 2 years of experience using digital banking.Followed by 15% experienced in using digital banking for more than 3 years.This shows that the millennial generation with more than 50% respondents has started using digital banking in the last 3 years.The development of experience has made the digital banking transformation of the millennial generation to encourage the development of digital banking (Vinerean et al., 2022).Experience defines the development of digital banking from time to time.Further analysis in table 7 can be seen that experience as a moderating variable is not able to influence social influence and price value on the intention to use digital banking.In contrast to habit, with increasing experience, user habits will develop in dealing with digital banking.Experience also supports hedonic motivation in the use of digital banking.Experience in using technology has resulted in mastery of technology adding to the perceived benefits for the millennial generation.In terms of facilitating conditions, the more experienced digital banking users are, the more skilled they are in mastering the technology structure, thereby reducing the available external support.Experience is able to moderate Performance expectancy and Effort Expectacy significantly on behavior intention.The millennial generation will be prejudiced against experience, where if experience provides more effectiveness and productivity, the millennial generation's decision making will use adopt behavioral intention digital banking.Experience contributes to convenience, easier to adjust friendly technology (Schmitz et al., 2022).

Contribution
The analysis in this research has contributed to the acceleration and accessibility of digital banking by the millennial generation.The contribution is in the form of the latest phenomenon regarding digital banking and its acceleration.The first is research and empirical, this research examines and evaluates the models and theories used in the digital banking transformation process, namely in the form of technology acceptance.In this case, using the theory developed by the UTAUT2 structural model was specifically developed to clarify the adoption of the new system from the perspective of the customer, in this case the millennial generation .Thus, the major contribution to this study is as a starting point for the validity test for the use of new technology (in this case internet banking for the millennial generation.In theory, UTAUT2 is able to clarify behaviour intentions towards technology, explain the structure of the model that is able to guide behaviour intentions towards adopting technology.Experience is developed with new factors and modification of new relationships using experience as a moderator.The next contribution is regarding the concept of UTAUT2 theory in their model which obtains functional utility aspects, intrinsic aspects, and financial aspects which are directly predict behavioral intentions (Venkatesh et al., 2012).This research offers the banking sector the opportunity to design and market products through technological innovation that can become a competitive advantage thereby increasing digital banking acceptance among the millennial generation in Central Java.

Conclusions and Recommendations
The development of this research focuses on identifying the main factors that influence the behavioral intentions of the millennial generation in their acceptance of the use of digital banking.The main factors influencing the behavioral intentions of the millennial generation are significantly influenced by performance expectations, effort expectations, facilitating conditions, hedonic motivation, habits.Social influence and price values are not factors that contribute to the cultural behavioral intentions of the millennial generation in accepting the use of digital banking.In line with hypothesis testing, moderation testing with experience as a moderating variable that has no influence is social influence and price value.
The first limitation of this research is that the moderating variable experience can still be identified for other influencing moderating variables.The second limitation that can be carried out in further research is the possibility of conducting cross-country research objects so that more comprehensive research results can be created.

Table 2 .
Demografic Characteristic of Digital Banking Respondent

Table 2
Discriminatory testing of validity Fornell Lacker Criteria is useful to see the correlation of one variable with another.In table 4 the results show that each variable meets the standard cross loading value.The value of cross loading can be seen diagonally with the position of the top table being the most valuable.Followed by the table below with lower values.

table 4 .
The value of croncbach alpha is > 0.7 and for composite reliability the value is > 0.7.The illustration from table 4 means that all latent variables have an adequate reliable composite value, with the estimated test value above the standard value of 0.7.From table 4, the highest value for CR is 0.959 by HM, followed by the lowest value of 0.878.Each construct was found to have a value above the standard Cronbach alpha value of 0.7.For the highest value of cronbach alpha of 0.993 attached to EE with the lowest value of cronbach alpha owned by PV with a value of 0.792 even though the value is above an adequate value.