Wholesale CBDCs: Lessons Learned and What the Future Looks Like

Central banks have had real-time wholesale payment systems in place for sometime. These CBDC systems help settle cross-border transactions, large inter-bank transactions, and serve as the backbone of a nation's economy. In many cases, a wholesale currency needs to operate without fail for global systems to function correctly. But wholesale CBDCs have yet to be perfected. Recent projects have shown that we still have a lot to learn about architecting the right solution.

In this article we’ll talk about the goal of wholesale CBDCs and what lessons we can glean from some recent projects. We’ll also talk broadly about what might need to be true for a single solution to properly power an increasingly global economy.

Recent Wholesale CBDC Efforts

The Reserve Bank of Australia (RBA) recently completed “Project Atom”, an effort to create  a wholesale CBDC used to settle interbank transfers and transactions and documented their findings in a comprehensive white paper report. The digital currency allowed banks to eliminate manual, paper-based processes typically associated with data maintenance, transaction processings, and loan payments and settlements. Before we talk about lessons learned from Project Atom, let’s take a look at how the CBDC was built.

Technical architecture

The foundation of the project included two sub-layers:

• The Hosting Infrastructure Layer that used Amazon Web Services cloud infrastructure. This layer hosted off-chain application data and services. Namely, databases used to simulate bank accounts and transactions.
• Above that lived the Blockchain Infrastructure Layer. This layer validated and recorded transactions processed on the blockchain network. Here Project Atom maintained a ledger of all token transactions and balances held in each wallet. Kaleido maintained the blockchain network layer that managed tokens and wallets.

Blockchain infrastructure

The blockchain layer consisted of five nodes on an Enterprise Ethereum network. To facilitate a high throughput of transactions and because of its immediate finality, a Hyperledger Besu client was used as it leverages an IBFT2 (Istanbul Byzantine Fault Tolerance 2) consensus algorithm. This voting-based proof-of-authority consensus mechanism allowed approved nodes, or validators, to validate transactions and blocks. The immediate finality is achieved because the creation of a new block rotates between validators, meaning only one block is proposed at any given block height.

According to the white paper, further development of the blockchain layer would include more validator nodes distributed across more participants and multiple chains to handle data, connected by bridges. These solutions are represented in the images below.

‍

Challenges Facing Wholesale CBDCs

The white paper points out a handful of issues related to digital currency. There are challenges presented by the technology itself, like how does one construct a solution robust enough to handle a high number of transactions and remain online around the clock. And also practical challenges related to data and information, like where data is stored and how messages are passed.

Let’s dive deeper into these challenges and talk specifically about wholesale CBDC requirements as it relates to efficiency, throughput, programmability, and privacy.

Enterprise-grade up-time

If a commercial bank collapses, it’s possible that an individual would lose a portion of their savings. Central banks offer a stronger layer of protection. The value of the currency central banks issue is backed by law. This makes a lot of sense for consumers. But how do we ensure the security of a system this important?

Wholesale CBDC systems require a truly enterprise-grade solution. Any banking tool central to the economy is only as strong as its weakest link, and a bad node could crash the system. Risk reduction in CBDCs starts with ensuring the system is live. Uptime in banking is liquidity, security, and access. If a central bank cannot guarantee uptime, it should be designed to empower offline use, according to Visa.

Standardization

As we look at the intricacy of the Project Atom architecture we see another challenge facing wholesale CBDCs. For systems to be truly scalable, able to connect to other central systems, we need some measure of interoperability. Many distributed systems begin to look much like the siloed landscape of the past if our central banks can’t communicate with each other.

According to a report published by the Bank of International Settlements, “for CBDC systems, domestic interoperability would need to be sufficient to achieve an easy flow of funds to and from other payment systems and arrangements.”

Programmability

The ability to automate financial instruments is one of the great advantages of digital currencies and also one of the challenges we face in architecting a solution to match our complex economy. Any system we build needs to come with the ability to do business via smart contracts. Replacing goods-for-services business with atomic swaps, where events trigger payments, is the wave of the future.

“The programmability of CBDC would allow retailers and banks to impose restrictions on transactions or trigger additional validation requests,” according to Accenture.

Privacy

Privacy remains an issue for wholesale CBDCs. And further developments in zero-knowledge protocol (ZKP) is one answer. A zero-knowledge proof is a method where one party can prove to another party that a value or statement is true without conveying any information about the statement besides the fact that it is true. Zero-knowledge proofs are a rapidly developing space, and any wholesale CBDC architecture should be flexible and intelligent enough to incorporate new developments as they come online.

According to a recent publication from the Bank of Canada, zero-knowledge proof systems are in their infancy and not quite ready to deploy. This immaturity creates vulnerabilities, but central banks need to be ready to implement these protocols as soon as they’re ready.

Looking Forward

The interesting thing about central bank applications of blockchain technologies is that in many ways they are developing with the same technology, but remain independent of the ever-expanding world of digital finance. Central bank applications, unlike those in DeFi, cannot be open to the public, they require consortium management capabilities. They require a higher level of enterprise security, as global systems depend on these currencies to operate. Proof of work and proof of stake aren’t realistic, as they need more flexibility. And they have to be resilient, always on.

Central banks will continue to experiment with wholesale CBDC applications and platforms like Kaleido will evolve with those learnings to ensure a future built around security, efficiency, and interoperability.