The European Central Bank (ECB) recently announced that it will start an investigative phase of the digital euro project in Q4 2021. While this was not a big surprise to many who closely follow the worldwide discussions about CBDCs and digital money, the ECB released a very interesting report on experiments that were conducted over the past 9 months. This report and its results caught many by surprise. It shows that the ECB has already tested different technologies as the core infrastructure for a potential digital euro and investigates how to mimic cash-like privacy and offline payments in a digital world. This article provides a brief overview about the background of why the ECB looks into the digitization of physical cash, which experiments it has already conducted, and which results these experiments produced.

Background

For years, cash has been losing importance as a means of payment worldwide. Scandinavia is at the forefront of this development, with surveys by the Swedish Riksbank showing that in Sweden less than ten percent of transactions are carried out in cash. As a result of the Corona pandemic, the trend toward digital payment methods has also increased further in the rest of Europe. For example, according to representative surveys by the German Bundesbank, the proportion of cash transaction value in 2014 and 2017 was 53 and 48 percent, respectively, but by 2020 it had fallen to only around 32 percent. This trend is expected to continue in the future, resulting in a further decline in the share of cash in payment transactions. 

Cash plays a special role in today's monetary system. It is the only legal tender. Money created by commercial banks represents a claim on the respective bank, where the bank customer holds his current account for the disbursement of cash. Because central banks, in contrast to commercial banks, cannot become insolvent and can even operate with negative equity, central bank money is a very safe form of money. Central banks around the world are increasingly emphasizing the need to ensure access to a public and secure means of payment also in an increasingly digitized world. 

The decline of cash as a relevant means of payment is, therefore, one of the main reasons for central banks in advanced economies to consider developing their own digital currencies, so-called central bank digital currencies (CBDCs). To some degree, CBDCs have the potential to become “digital cash” and mimic a digital version of “physical cash”. The ECB announced on July 14 that it will launch a two-year investigation project in which the Eurosystem, consisting of the ECB and the national central banks in the euro area, will examine the framework conditions for a potential launch, different design variants, and the possibilities of issuing a digital euro. Following the investigation phase, a decision will be made on the start of a further development phase lasting up to three years.

Experiments of the last nine month

In the investigative phase starting in Q4 2021, the ECB will build on the findings of previous experiments conducted over the past nine months. In the experiments, the ECB tested various technical design variants of possible technical infrastructures for different functions in four workstreams:

• Centralized systems: Connection of the CBDC prototype with the TARGET Instant Payment settlement (TIPS) system via the Single Euro Payments Area (SEPA) payment system, via payment channels at stores or at online-merchants (point-of-interaction), and via PSD2 interfaces. In particular, the scalability of the TIPS system was investigated.
  
  
• Interaction between centralized and distributed systems: Enabling innovative functionalities while building on existing infrastructures. Tests were conducted on how an account-based and a distributed ledger-based (DLT) digital euro could complement each other. In the experiments, the centralized TIPS system was used as a settlement layer, while liquidity was transferred via "payment channels" on distributed networks. This interlinkage enabled innovative payment options while respecting privacy and programmability of transactions. However, it remained unclear whether liquidity transferred via payment channels can be a claim on the central bank or must ultimately be a claim on the payment channel operator (so-called "synthetic CBDCs").
  
  
• Distributed systems: Use of a DLT-based platform and fixed-denomination tokens ("digital currency bills"). The experiments explored and demonstrated the scaling potential of distributed platforms, as well as how such a design could protect privacy without hindering compliance policies such as anti-money laundering (AML) and combating the financing of terrorism (CFT) regulation. It further tested the combination of a distributed system with digital identities (e-ID).
  
  
• Hardware and offline systems: Hardware payment solutions with bearer instruments (personal payment devices) that could facilitate the use of a digital euro as an offline means of payment. Solutions from a total of six companies were tested and research reports were prepared by the companies on predefined questions. Issues addressed included the transfer of an offline token between different parties, enabling different types of privacy, geographic limitations, cyber-security aspects, usability, and transaction costs.
  
  

Results of the experiments

The four technological designs were examined for different functions: 1) the scalability and flexibility of the technical designs, 2) privacy while complying with AML and CFT regulation, 3) the possible quantity limits of the digital euro, and 4) end-user access.

1. Scalability and flexibility of technical designs:
   
   
   • Both the centralized and distributed infrastructures were capable of processing 40,000 transactions per second. The distributed systems tested even achieved a transaction rate of up to 325,000 transactions when focussing on “core components” which makes them significantly more scalable than the TIPS system.
     
     
   • Transactions could be settled in 0.8 seconds (TIPS) or 1.3 seconds (DLT-based).
     
     
   • The energy consumption of the centralized and distributed systems was very low. The energy required to run the systems was equivalent to the consumption of a single electric car on a highway or the carbon footprint of a few European households.
     
     
   • The potential extension of a digital euro to include programmable transactions was successfully tested and positively evaluated in distributed blockchain-based systems using payment channels. 
     
     
   • The feasibility of offline payments via hardware- and token-based systems was confirmed from a technical point of view. However, they did not answer all questions on how to fully control the "double-spend" risk.
     
     
2. Privacy and compliance:
   
   
   • Both the centralized settlement system (TIPS), the decentralized blockchain-based solution, the combination of both, and the hardware solutions were examined for different levels of privacy and the respective fulfillment of compliance requirements such as related to AML, and CFT.
     
     
   • In the TIPS system, privacy could be strengthened via the use of pseudonyms. Higher privacy could be established via "anonymity cards". These cards allow a special TIPS account to be loaded with cash, thus eliminating know-your-customer (KYC) procedures.
     
     
   • Different levels of privacy could be achieved via the DLT-based systems tested: i) using one time pseudonyms could prevent the many pseudonyms from being associated with the payer, ii) "mixing" transactions could prevent the pseudonyms of senders and receivers from being associated, iii) using bilateral payment channels could achieve different levels of privacy. 
     
     
   • Hardware systems enabled complete anonymity by transferring the token offline.
     
     
3. Limits on a digital euro in circulation:
   
   
   • All technical variants were examined for limits on the digital euro in circulation with a focus on a quantity limitation of a maximum amount held and a two-tiered interest rate.
     
     
   • An automated transfer to a bank account after exceeding a defined maximum amount was evaluated positively and technically feasible.
     
     
   • Tiered remuneration was also successfully implemented on the various technical systems. However, according to the report, a tiered remuneration could potentially generate euros of different qualities which must not happen since a euro should always have the same value and should always be fungible – regardless of the payment method. 
     
     
   • For offline bearer instruments, a capped amount for individual transactions or the balance held offline would be possible. Imposing limits on transactions in terms of time and remuneration, however, presented significant challenges as this would need an online-connection at least from time to time to check potential validations. 
     
     
4. Access for end users:
   
   
   • The experiments show that the use of existing infrastructures and technologies will facilitate the adoption of the digital euro as a payment instrument. Point-of-sale card processing systems, smartphone applications as well as web applications were tested.
     
     
   • Hardware-based systems that transmit the digital euro via near field communication (NFC) and Bluetooth were found to be promising in terms of transaction speed. However, limitations were encountered with large amounts of data to be transmitted, which need to be investigated further.
     
     
   • Tests based on different electronic ID (e-ID) solutions showed that it is possible and very helpful to combine a blockchain-based digital euro with a digital user authentication solution. A centralized and a federated identity solution connecting multiple data sources were tested. Tests with fully distributed solutions (Self Sovereign Identity, SSID) will be continued.
     The experiments showed that linking e-ID with digital euro wallets/accounts, maximum amounts, and tiered remuneration can be enabled in a relatively smooth way. The combination with e-IDs could also facilitate switching between digital euro wallet/account providers and potentially reduce KYC and AML costs. However, a successful implementation would first require widespread adoption of e-ID solutions in Europe.
     
     

This article is based on the report "Digital euro experimentation scope and key learnings" by the ECB