Mobile wallet adoption – Moderating effect of using experience

KỶ YẾU HỘI THẢO KHOA HỌC QUỐC GIA CITA 2020 “CNTT VÀ ỨNG DỤNG TRONG CÁC LĨNH VỰC” Mobile Wallet Adoption – Moderating Effect of Using Experience Ha Hoang1[0000-0002-5630-0454] 1 University of Economics - The University of Danang, Da Nang 550000, VN hahoang@due.edu.vn Abstract. Considered as a disruptive innovation, the development of mobile wallets in Vietnam in recent years has created enormous changes in customers' shopping and payment habits. This study as- sesses the moderating effect of customers’ Using Experience (UE) on three constructs of the research model - Sales Promotion (PM), Perceived Usefulness (PU) and Intention to Use (IU). The results have shown that PU plays a mediating role between PM and IU. When the customer experienced a long period of using the mobile wallet, a high promotion will have the opposite effect of reducing the PU. Therefore, mobile wallet providers need to focus PM on new users because these customers will have a higher PU than long-used customers. Several managerial implications were proposed to increase the effectiveness of PM in enhancing adoption of the mobile wallet in Vietnam. Keywords: Mobile Wallet, Mobile Payment, Digital Transformation, Technology Adoption 1 Introduction In the context that the COVID-19 epidemic is still extremely complicated in Vietnam and around the world, payment tools such as mobile wallets play a very important role in the economy. It allows purchases to take place without violating safety regulations to prevent infection. Mobile wallets could replace physical wallets and even debit or credit cards in online and direct purchases. We see this technology as a major revolution in the digital economy because in the era of technological revolution, the issues of speed, interoperability, security and privacy of mobile technology have been resolved. Although the application of technology has received long-standing attention from the government, it has not developed to achieve the desired scale, required for application in the entire systems payments in Vi- etnam. As a country that has many advantages to develop digital-based services [1], however, Vietnamese clients still prefer to use cash, reflected in the fact that there are still nearly 90% of transactions are paid by cash [2]. This study assesses the moderating effect of Customers’ Using Experience (UE) on three con- structs of the research model - Sales Promotion (PM), Perceived Usefulness (PU) and Intention to Use (IU). We surveyed 315 young people in Da Nang city to measure this effect based on the proposed research model. The moderating effect of UE demonstrated in this study will be the premise for future researches of the adoption of technology and especially researches related to mobile wallet. 2 Literature review and research framework 2.1 Literature review Promotion: Related to the financial and payment sector, Preston et al. (1978) showed the influence of sales promotion to persuade customers to open a bank account. Based on the results of this research, 50% of the increase accounts of a bank is the result followed a promotional campaign. Others researches showed the efficiency of promotional tool to influence the computer purchase and confirmed the positive linkages, promoting the behavior of purchase of financial services [4]. Recent research on omnichannel shopping based on IT has included the promotion in the research model but has not found a meaningful impact of the promotion on intention to use [5]. In Vietnam, based on a recent study on the determinants of the choice of the customer mobile wallet, promotion is in second place among the six elements outlined in the survey. As emphasized by several 122 Ha Hoang authors, there are academic and managerial absences on the deeper knowledge of the relationship between promotion and consumer behavior [6; 7]. Especially in technology adoption, sales promotion still receives very little attention from researchers. That is why the construct of promotion is added to this study. Perceived usefulness: Perceived usefulness was first proposed by Davis in the TAM (technology ac- ceptance model), defined as the ability of users to improve performance by fully using the enhanced system, reflecting users’ willingness to accept [8]. In technology adoption research, performance expectancy refers to an individual’s perception that using a technology could provide benefits to users in performing certain activities [9]. Reflecting a range of at- tributes that a technology could give benefits to clients, performance has been conceptualized by using system features that could enhance speed, productivity, and chances of task accomplishment and perceived usefulness [9; 10]. Unambiguously, in diverse task settings, performance expectancy was affirmed to affect intentions to use technological systems [11]. Consumers’ perception that using mobile wallet would support them to achieve benefits in performing payment tasks could affect the behavioral intention of mobile wallet adoption. Intention to use: According to Davis, behavioral intention is defined as the degree to which an individual believes that they will implement a particular behavior [12]. In technology adoption theories, the relation- ship between behavioral intention and usage behavior has been consistently confirmed [9; 10; 11; 13; 14]. 2.2 Research framework Based on the relationships extracted by hypotheses from previous studies, Figure 1 shows the proposed research model. Fig. 1. Conceptual framework (Source: own elaboration) We proposed this research framework to evaluate the moderating effect of UE to the research’s con- structs. This integrated model can help us have a better understanding of the relationships between con- structs. These relations are already confirmed in the literature review of similar researches, and the validity of model is proven in previous studies except the promotion factor that we added in this study [15; 16; 17]. 3 Materials and Methods This study employed a survey method, using a questionnaire to test the conceptual model and developed hypotheses. The prospective respondents were chosen in convenience sampling method from Da Nang City. 400 young people in Da Nang were contacted by e-mail and social network during the period from January 2020 to April 2020. A link to the survey was included in the messages. 315 valid responses were received. The overall response rate was 78.75%. Measurement instrument A questionnaire-based survey was developed to test the theoretical constructs. The research had three con- structs: perceived usefulness, promotion, intention to use. Constructs and measurement items were adapted 123 KỶ YẾU HỘI THẢO KHOA HỌC QUỐC GIA CITA 2020 “CNTT VÀ ỨNG DỤNG TRONG CÁC LĨNH VỰC” with slight modifications from technology acceptance literature to build the questionnaire. Measurement items for constructs of perceived usefulness were adapted from Venkatesh et al. (2012). The scale of inten- tion to use was employed from Davis (1989). All maim measurement items were measured on a five-point Likert scale, ranging from totally disagree (1) to totally agree (5). The using experience measurement was operationalized by one item that measure consumers’ using experience of mobile wallet usage. Analytical procedures We followed a two-step approach. First, the scales are tested by the confirmatory factor analysis (CFA). Second, we used the Structural Equation Modeling (SEM) method to evaluate the structural relationships between constructs in a theoretical model. 4 Results 4.1 Measurement model: reliability and validity CFA is used to test the overall measurement model and to assess the reliability and validity of the constructs [18]. We based on composite reliability and average variance explained (AVE) to assess convergent valid- ity. By the rule of thumb, composite reliability (CR) should be above 0.6 and AVE should higher than 0.5 for all constructs. As shown in Table 1, CRs fluctuate from 0.856 to 0.898 while the AVEs range from 0.664 to 0.746. These results show that the model meets the criteria for assessing convergent validity. The AVE for each of the research constructs should be greater than the squared correlation between the construct and any of the other constructs [19]. The Table 3 displayed the measurement model satisfy the requirement for discriminant validity. The diagonal elements in bold are the squared multiple correlations between the research constructs. We can conclude that all the constructs in this study have adequate discri- minant validity because the results indicated that the various AVE are lower than the diagonal variables. Table 1. Reliability and validity of the tested model Latent constructs Cronbach's alpha CR AVE MSV FL range PU 0.875 0.883 0.716 0.413 0.775-0.874 PM 0.857 0.856 0.664 0.473 0.743-0.850 IU 0.891 0.892 0.674 0.473 0.692-0.917 Table 2. Factor correlation coefficients and the square root of AVE (shown as bold at diagonal) CR AVE MSV MaxR(H) I_U P_U P_M I_U 0.892 0.673 0.475 0.893 0.820 P_U 0.883 0.717 0.415 0.890 0.644*** 0.846 P_M 0.856 0.664 0.475 0.859 0.689*** 0.588*** 0.815 Significance of Correlations: † p < 0.100 * p < 0.050 ** p < 0.010 *** p < 0.001 124 Ha Hoang The table 2 showed that the CFA model with three concepts also shows a good fit with the data. All the indices satisfied the recommended cut-off point. We conclude that the model fits data well, so this model can be used to test the research hypotheses. Table 3. Indices fit criteria Fit indices Structural Mode Threshold limit value Sources CMIN/df 1.838 < 5 [20] CFI 0.986 > .90 GFI 0.965 > .90 [21] AGFI 0.940 > .80 [22] RMSEA 0.052 < .08 Fig. 2. Structural model Table 3. Testing Direct effects Estimate S.E. C.R. P Label PU <--- PM .704 .075 9.374 *** IU <--- PU .325 .057 5.710 *** IU <--- PM .505 .072 6.996 *** Table 4. Testing Indirect effects of PU Parameter Estimate Lower Upper P PM->PU->IU .229 .125 .379 .001 Table 5. Testing Moderating effects of PU from PM to PU coeff se t p LLCI ULCI constant 2.9281 .4422 6.6219 .0000 2.0580 3.7982 PM .2475 .1088 2.2747 .0236 .0334 .4616 PU -.5420 .1936 -2.7992 .0054 -.9230 -.1610 Int_1 .1493 .0464 3.2206 .0014 .0581 .2406 125 KỶ YẾU HỘI THẢO KHOA HỌC QUỐC GIA CITA 2020 “CNTT VÀ ỨNG DỤNG TRONG CÁC LĨNH VỰC” We see that the interaction between PM and PU was statistically significant (b = 0.1493, s.e. =0.0464, p = 0.0014 < 0.001), suggesting the PU moderates the effect of PM on PU. Table 6. Conditional effects of the focal predictor at values of the moderator(s) UE Effect se t p LLCI ULCI -.9073 .3969 .0723 5.4924 .0000 .2547 .5390 .0000 .5324 .0542 9.8247 .0000 .4257 .6390 1.0258 .6855 .0679 10.0896 .0000 .5518 .8192 Fig. 3. Moderating effect of PU These are simple slopes of the relationship between PM and PU at 3 points along the scale of the mod- erator (Hayes, 2018). At -1SD on UE, the effect was positive and significant (b = 0.3969, p < 0.001). At the mean of UE, the effect of PM was positive and significant (b = 0.5324, p < 0.001). At +1SD of UE, PM was a significant positive predictor (b = 0.6855, p < 0.001). Thus, the influence of PM is proportional to PU, but moderated by UE. Specifically, with low PM, low UE will have higher PU than high UE. At me- dium PM and high PM level, high UE will correspond to higher PU for the same PM level. This means that when customers used mobile wallet for a long time, a high promotion will have the opposite effect of reducing the PU. Table 7. Testing Moderating effects of PU from PM to PU coeff se t p LLCI ULCI constant 2.8123 .1733 16.2297 .0000 2.4713 3.1533 PM .3690 .0459 8.0457 .0000 .2788 .4593 PU .2713 .0420 6.4547 .0000 .1886 .3540 UE .1146 .0275 4.1696 .0000 .0605 .1687 Int_1 .0534 .0348 1.5322 .1265 -.0152 .1219 We see that the interaction between PM and IU was statistically insignificant (p = 0.1265 > 0.05), sug- gesting the PU do not moderate the effect of PM on IU. 126 Ha Hoang Table 8. Index of moderated mediation: Index BootSE BootLLCI BootULCI UE .0405 .0213 -.0004 .0855 This result provides an omnibus test of the conditional indirect effect (Preacher et al, 2007) reflected in the index of moderated mediation (Hayes, 2018) of X on Y. If the null of 0 does not fall between the lower and upper limit of the 95% confidence interval, we infer that the indirect effect is conditional on the level of the moderator variable (w). In this study, we found that PU does not moderate the indirect effect of PM on IU. 5 Discussions PU moderates effect of PM on PU To the best of the authors knowledge, this is the first study confirmed the moderating effects of PU on PM and PU. Specifically, the more PM there are, the more PU will increase, but this relationship is moderated by UE. At the low level of PM, a low UE will have higher PU than a high UE. At medium PM and high PM level, high UE will correspond to higher PU for the same PM level. This means that when the customer experienced a long period of using mobile wallet, a high promotion will have the opposite effect of reducing the PU. Therefore, mobile wallet providers need to focus PM on new users because these customers will have a higher PU than long-used customers. Given the fact that marketing and advertising budgets are increasingly limited, especially in the context of the Covid epidemic that has been going on for almost a year, companies need to focus on the new customers to improve PM's efficiency, thereby increasing the use of mobile wallets of customers. PM have a significant and positive impact on PU. PM has a direct impact on PU (β = 0.704; p-value < 0.001). This result also confirms the separation of PM and PU. Indeed, in the current fierce competition to win customers’ attention, the perceived usefulness has become almost inevitable. PU-free technologies would almost certainly be rejected by customers. There- fore, PU is almost a must for all companies who want to enter the market that need to prove the advantages of their products. However, in order for customers to choose their products, companies need to have appro- priate PM strategies, ensuring long-term maintenance and financial efficiency. Implementing the above strategy would create a long-term competitive advantage for mobile wallet providers in an extremely fierce market with more than 30 mobile wallet providers like in Vietnam. PU have a significant and positive impact on IU. Perceived usefulness in this research has a positive and significant impact on intention to use mobile wallet (β = 0.325; p-value < 0.001). This result is consistent with many studies on mobile payment area which found that perceived usefulness was one of the most influential factors to determine behavioural intentions of users [23; 24; 25]. Usefulness of a new technology is essential to increase intention to use. PM have a significant and positive impact on IU. This is also one of the theoretical contributions of this study. Most studies about the adoption technologies do not take PM as an independent influence on the intent to use the technology. This result again confirms the need to separate PM as an independent factor on researches of adoption technology in the future. Fur- thermore, among the constructs in the model which have a direct relationship with IU, PM was shown to have the strongest impact to IU (β = 0.505; p-value < 0.001). Indeed, in a recent survey conducted in major cities in Vietnam about the key factors influencing consumers' choice of mobile wallets, diverse and regular promotions were assessed as one of the deciding factors to choose a digital wallet. In fact, many people used a technology for the first time because of attractive promotions. Therefore, mobile wallets’ suppliers need to have many attractive promotions to gain consumer acceptance. Compared to other platforms, mo- bile wallets do not have many attractive promotions and that may be the reason that mobile wallets are still not popular in the Vietnamese market. 127 KỶ YẾU HỘI THẢO KHOA HỌC QUỐC GIA CITA 2020 “CNTT VÀ ỨNG DỤNG TRONG CÁC LĨNH VỰC” 6 Conclusion To conclude, mobile wallet is one technology where there is a lot of potential and opportunities to grow, especially for a country that has ambitions to become a cashless economy like Vietnam [2; 26]. In this context, the studies on mobile wallet not only have theoretical but also practical implications. This study is one of the first to discover moderating effect of UE on the influence of PM and IU. When the customer experienced a long period of using the mobile wallet, a high promotion will have the opposite effect of reducing the PU. The results confirm that promotion and perceived usefulness of mobile wallets are determining factors in user adoption. To get users to adopt a mobile wallet, application developers must emphasize the benefits of usefulness and attractive promotion related with this new financial service. Furthermore, in this study, scales measuring construct sale promotion in the model were developed based on the literature and qualitative methods such as expert panel review and focus group interview. These items provide the foundation for further research in consumer behavior concerning technology adoption. Beside of our study’s major contribution that adds into the existing body of knowledge, we also recog- nize its limitations, mostly regarding the sampling with typically young, highly educated people as respond- ers. In addition, our results have limitation regarding the specific context of Danang city. Danang is an average city with a relatively limited market as its population is approximately one million people. This sample only covers with young people in this city. In order to further enhance its generality, future research could extend the study to more cities or conduct in more countries in Southeast Asia to improve the gener- alization of the study. Although this research contributes more empirical results in this area, it also has some limitations and therefore further studies in this area will be needed. References 1. Uyen, T. D., & Ha, H. The sharing economy and collaborative finance: The case of P2P lending in Vietnam. Jurnal Ilmiah Ekonomi Bisnis, 22(2) (2017) 2. Ha, H. The Cashless Economy in Vietnam-The Situation and Policy Implications. Journal of Reviews on Global Economics, 9 (2020) 3. Preston, R. H., Dwyer, F. R., & Rudelius, W. The Effectiveness of Bank Premiums: Stoneware calculators and Mercedes Sport Coupes–do they pay their way? Journal of Marketing, 42(3) (1978) 4. d’Astous, A., & Jacob, I. Understanding consumer reactions to premium‐based promotional offers. European Journal of Marketing (2002) 5. Lee, Y., & Cheon, H. (2019). A Study on the Factors Affecting the User Intention of Omnichannel Shopping Based on Information Technology. Paper presented at the Proceedings of the 2019 5th International Conference on E-Business and Applications. 6. d'Astous, A., & Landreville, V. An experimental investigation of factors affecting consumers' perceptions of sales promotions. European Journal of Marketing (2003) 7. Haans, H., & Gijsbrechts, E. “One-deal-fits-all?” On category sales promotion effectiveness in smaller versus larger supermarkets. Journal of Retailing, 87(4) (2011) 8. Davis, F. D. Perceived usefulness, perceived ease of use, and user acceptance of information technology. Mis Quarterly (1989) 9. Venkatesh, V., Thong, J. Y., & Xu, X. Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. Mis Quarterly (2012) 10. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. User acceptance of information technology: Toward a unified view. Mis Quarterly (2003) 11. Baptista, G., & Oliveira, T. Understanding mobile banking: The unified theory of acceptance and use of technol- ogy combined with cultural moderators. Computers in Human Behavior, 50 (2015) 12. Davis, F. D. (1985). A technology acceptance model for empirically testing new end-user information systems: Theory and results. Massachusetts Institute of Technology, 13. Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. User acceptance of computer technology: a comparison of two theoretical models. Management science, 35(8) (1989) 14. Venkatesh, V., & Davis, F. D. A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management science, 46(2) (2000) 128 Ha Hoang 15. Kim, C., Mirusmonov, M., & Lee, I. An empirical examination of factors influencing the intention to use mobile payment. Computers in Human Behavior, 26(3) (2010) 16. Liébana-Cabanillas, F., Marinković, V., & Kalinić, Z. A SEM-neural network approach for predicting antecedents of m-commerce acceptance. International Journal of Information Management, 37(2) (2017) 17. Shaw, N. The mediating influence of trust in the adoption of the mobile wallet. Journal of Retailing and Consumer Services, 21(4) (2014) 18. Garver, M. S., & Mentzer, J. T. Logistics research methods: employing structural equation modeling to test for construct validity. Journal of business logistics, 20(1) (1999) 19. Fornell, C., & Larcker, D. F. Evaluating structural equation models with unobservable variables and measurement error. Journal of marketing research, 18(1) (1981) 20. Bentler, P. M., & Dudgeon, P. Covariance structure analysis: Statistical practice, theory, and directions. Annual review of psychology, 47(1) (1996) 21. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. Multivariate data analysis 6th Edition. Pearson Prentice Hall. New Jersey. humans: Critique and reformulation. Journal of Abnormal Psychology, 87 (2006) 22. Hu, L. t., & Bentler, P. M. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural equation modeling: a multidisciplinary journal, 6(1) (1999) 23. Liébana-Cabanillas, F., Sánchez-Fernández, J., & Muñoz-Leiva, F. Antecedents of the adoption of the new mobile payment systems: The moderating effect of age. Computers in Human Behavior, 35 (2014) 24. de Luna, I. R., Liébana-Cabanillas, F., Sánchez-Fernández, J., & Muñoz-Leiva, F. Mobile payment is not all the same: The adoption of mobile payment systems depending on the technology applied. Technological Forecasting and Social Change, 146 (2019) 25. Madan, K., & Yadav, R. Behavioural intention to adopt mobile wallet: a developing country perspective. Journal of Indian Business Research (2016) 26. Kapoor, K. K., Dwivedi, Y. K., & Williams, M. D. Examining the role of three sets of innovation attributes for determining adoption of the interbank mobile payment service. Information Systems Frontiers, 17(5) (2015) 129