Privacy Matters for CBDCs

Uniquely amongst payment options, cash offers people payment privacy. A key part of the debate about Central Bank Digital Currencies (CBDCs) is about the extent to which a CBDC can mimic the privacy achieved by cash. A new Bank for International Settlements (BIS) working paper considers designing CBDCs for privacy protection and data governance ¹.

The paper makes the point that, while for cash anonymity and privacy are inherent in what cash is, CBDCs have to comply with anti-money laundering (AML) and combating the financing of terrorism (CFT) regulations. For a CBDC, the degree of anonymity and privacy protection are determined by the policy judgment of the central bank.

A representative sample of 3,500 people in Korea were surveyed to understand how their willingness to use a CBDC changed as the degree of anonymity and privacy protection varied. The survey also measured whether people’s willingness to use a CBDC changed as they became aware of the privacy benefits that could be designed in to a CBDC, such as preventing the commercial use of personal identity information and transaction data.

Privacy options in the trial

The first part of the randomisation process provided participants with one of the three versions of the CBDC design that varied the extent to which privacy was preserved when using a CBDC as a means of payment.

• A combined repository (CR) stores both personal identity information and transaction data in a single institution, which can be combined arbitrarily to identify an individual who pays for the associated transaction using CBDC.

• A separate repository (SR) stores personal identity information and transaction data separately in two different institutions, which makes it practically impossible to combine them for individual identification.

• A small-amount CBDC Voucher (Voucher) allows its users to have cash-like anonymity for small-amount transactions without revealing either personal identity information or transaction data to the AML authority.

For the second part of the randomisation, half of the participants were provided with the information that CBDC can prevent the commercial use (PCU) of its users’ personal identity information and transaction data by private financial institutions and BigTech companies. This information was not provided to the other half of the survey participants.

Findings

Privacy – preserving variations of the CBDC design significantly increased the willingness to use CBDC when respondents purchased privacy-sensitive products (eg. psychiatric services, adult products). Compared to the CR treatment as a baseline, the SR treatment and the Voucher treatment increased the willingness to use CBDC by 7 and 5 percentage points, respectively, in offline purchasing situations.

The impacts of the CBDC privacy- preserving design appear to be even larger in online purchasing situations; the SR design and the Voucher design increased the likelihood of choosing CBDC by about 11 and 9 percentage points, respectively.

Considering the average frequency of choosing CBDC in the CR treatment, the magnitudes of the effects of the privacy- preserving variations (ie. SR and Voucher) in the CBDC design ranged between 19% and 29%.

The study found that the increase in the willingness to use CBDC due to the privacy-preserving variations of the CBDC design was mostly accompanied by decreases in the usage of cash and credit and/or debit cards, but not in the demand for mobile fast payment.

When purchasing privacy-insensitive products (e.g., food, office supplies), there were negligible treatment effects in both offline and online purchasing situations.

The study confirmed that when CBDCs are designed to preserve privacy, there is a significant increase in people’s willingness to use CBDCs. For sensitive products, for example the afore-mentioned psychiatric services and adult products, the increase was up to 60% more. The increased willingness to use CBDCs was slightly greater for online purchases than offline, 2%.

The willingness to use CBDC increased significantly with the provision of information about the privacy benefits of using it. Information about the PCU increased the willingness to use CBDC by about 6 percentage points in the offline purchasing of privacy-sensitive products, amounting to a 20% increase relative to the usage of CBDC in the baseline CR treatment, and by 4% in online purchasing, amounting to a 9% increase relative to the baseline treatment.

When purchasing privacy-insensitive products, information about PCU had weak effects on respondents’ willingness to choose CBDC.

Compared to the baseline CR treatment, when respondents were provided with the privacy preservation of the CBDC design and the PCU information jointly, their willingness to choose CBDC increased by 14 percentage points in the SR and PCU treatment in both offline and online purchases, amounting to a 60% increase relative to the baseline treatment. In the Voucher & PCU treatment, willingness to choose CBDC also increases by 11 and 12 percentage points in offline and online purchases, respectively.

Finally, the study found that the treatment effects are larger for those who are more responsive to privacy protection activity by using cash in offline purchases, and for those who show less concern over privacy invasion by government or private institutions such as traditional financial institutions and BigTech companies.

In addition, it found heterogeneous treatment effects across demographic characteristics; for example, females are more responsive to the extent of privacy preservation associated with using CBDC and the provision of PCU information than males.

Summary and implications

When CBDCs are designed to preserve privacy, there is a significant increase in people’s willingness to use CBDCs, particularly for sensitive products. When people were informed about the privacy measures in place, the willingness to use CBDCs significantly increased.

Respondents’ trust in public or private institutions and demographic characteristics are important determinants of adoption. These, of course, vary by country.

The implication is that if CBDCs provide sufficient anonymity and protect privacy, they are more likely to substitute the existing payment instruments provided by the private sector, including commercial banks’ demand deposits. This means that CBDC could enhance the public-good role of money in providing anonymity and protecting privacy in the digital age, but at the expense of credit intermediation.

1 - Central Bank Digital Currency and Privacy: A Randomized Survey Experiment. BIS Working Papers, No 1147, Syngjoo Choi, Bongseob Kim, Young-Sik Kim, Ohik Kwon.