The emergence of stablecoins is a growing concern for authorities worldwide including Indonesia as it could affect financial stability. Thus, if a central bank chooses to develop a central bank digital currency (CBDC) to tackle this problem, the design should conform to the country’s characteristics and consumer needs. This study draws on experts’ opinions from various economic agents and utilises an amalgamation of the analytic network process (ANP) and the Delphi method to show that the cash-like CBDC model is the most appropriate digital currency design for Indonesia, since it could enhance financial inclusion and reduce shadow banking in Indonesia.

We examine the preferences of respondents for six types of payment instruments, namely cash, debit and credit cards, card and server-based electronic money, and internet or mobile banking. By applying a nested logit model to 500 household data covering six provincial capitals in Indonesia, we find that the decision to choose payment instruments is made sequentially. Socio-economic characteristics, including education, age, income, and transaction objectives or functionality have a significant effect on the probability of using non-cash electronic payment instruments. We find a substitution pattern between payment instruments, not only between cash and non-cash instruments but also between non-cash instruments. In light of these findings, appropriate payment system policies are in order to hasten the use of non-cash payment.