The Future of CBDCs: An In-Depth Interview with Jim Zhang

The world of central bank digital currencies (CBDCs) is rapidly evolving with numerous countries experimenting with the technology. From the People's Bank of China's Digital Currency Electronic Payment (DCEP) to the European Central Bank's Digital Euro, central banks are trying to bring digital versions of their national currencies to life.

The drive towards CBDCs is being spurred on by a number of factors, including the increasing use of cashless payment methods and the emergence of cryptocurrencies. Central banks are looking to ensure they can take advantage of the latest technologies in the digital age and are able to meet the changing needs of their citizens and businesses.

While the specifics of each CBDC experiment differ, they all share a common goal of providing a secure, convenient, and efficient digital alternative to cash and ensuring they can work more efficiently with commercial banks. As more and more countries explore the potential of CBDCs, it’s been interesting to see how these experiments shape the future of money and financial systems around the world. We sat down with Kaleido Head of Protocol and co-founder Jim Zhang to talk about developments in the CBDC space, what challenges remain to be answered, and where we’re headed next. The conversation was edited for clarity.

For people investigating CBDCs and the underlying technology, what are some of the big questions we’re still trying to solve?

There are a lot of early phase experiments. A majority of CBDC projects are still trying to figure out if it makes sense to build on a distributed ledger or should they use a more traditional centralized system. I think both types of architecture are still being actively used.

MIT’s Digital Currency Initiative published a paper from the project Hamilton that presented two designs of centrally operated architectures. The paper based the designs on the findings that the trust assumptions of a CBDC do not require using a decentralized architecture. Project Hamilton is a collaboration with the Federal Reserve Bank of Boston. Just based on that, I think there's a lot of interest in that type of architecture.

China has rolled out a production quality CBDC based on a centralized system. They've been working on that for 20 years. So that predates blockchain.

However, they are very interested in using blockchain for peripheral use cases around CBDCs, such as cross-border settlements, or foreign exchange. The country has been collaborating with the Hong Kong Monetary Authority, Bank of Thailand, People’s Bank of China, and Central Bank of the UAE for cross-border payments in the project “m-Bridge”, which is built on DLT technologies.

Does blockchain only make sense in a decentralized system? Would it make sense to use it in a centralized operation?

At Hyperledger Global Forum, I asked Dr. Imre Kocsis this question. He is a professor in Romania and his team has done work with the Central Bank of Romania. He said it's very legitimate to have a single-party blockchain-based CBDC system where the central bank is the only party looking after all the blockchain nodes. So then I asked him, how do you categorize the value of the blockchain if you operate it in a more centralized way? He said it's simply more robust, security wise than what currently exists. It’s a much more robust system. If you hack one node, it doesn't put the whole system in danger.

With Delivery vs. Payment, for example, you have a security settlement that needs to happen every day and institutions want to use CBDC to do the payments. We’re figuring out how DLT makes sense for these scenarios. On the other hand, if you have a higher performance benchmark to hit, you have a gap to fill using DLT.

What are some other interesting projects around the globe?

In terms of other DLT based projects in production, there is one in Nigeria and one in Cambodia. In Cambodia, the system is built on Hyperledger Iroha. Iroha is a DLT stack that has focused on digital assets from the start. Bank of France is using Hyperledger Fabric and they’ve experimented with a number of protocols, I think. They had a very good presentation at the Hyperledger Global Forum about their advanced privacy approach using zero knowledge proof to hide transaction content.

With so many countries experimenting with CBDCs, is there a vision for how these systems communicate or are we not there yet?

Swift is working to define how CBDC systems, whether based on DLTs or not, communicate in the future, much like how they’ve designed ways for our current banking systems to speak to one another. We worked with them to build a sandbox that will allow large institutions to collaborate and find answers about how best CBDCs can work together. That’s one example.

Elsewhere, I think a lot of the central banks are still figuring out how to run the technology and meet technical criterias. They are trying to prove that they can support the security requirements of both wholesale and retail use cases. Will I have enough privacy for all the participants? These are complex topics and most banks are still trying to get their effort off the ground so they can prove the potential.

And I think what Dr. Kocsis said makes sense here. IT security-wise, blockchain is a more secure system. Central banks, I think, in particular, are by nature going to be very concerned about hacks. If you have a centralized system, it's a single point of failure. DLT removes some of this risk. It gives you a much better security posture, and maybe that's enough value to justify using blockchain even with the challenges it presents.

Reading Project Atom, wholesale seems easier to implement where retail seems like you have to get a whole bunch of people on board and give them the tools to use the system. Is there a world where wholesale works and retail remains a challenge?

Wholesale is an easier starting point. It deals with a much smaller number of players in the system and they are players with good behavior. There are clear regulations that all the players abide by. It's also a much lower volume of transactions. So I think a smart way to roll out a CBDC project is to focus on wholesale first.

Still, even with wholesale digital currencies, you will have privacy concerns. Banks shouldn’t know what other banks' balance with the central bank is, which means a regular token on Ethereum, for example, just won't work. You can’t use typical privacy approaches like private transactions either, or with Hyperledger Fabric channels. None of them will work because you are by design running a token economy where all the players must have the single state tree. Private transactions or channels are basically segregated state trees.

Read more about why private transactions don’t work in a token economy.

Wholesale has the problem of privacy. Each participant is typically a large commercial bank that wants to run their own node in the network, so you have a privacy concern. With retail, it's different. Participants are individual users, they're not going to run any nodes and will access the system through API's, right? This makes it easier to control the privacy aspect and the issue is more the volume of transactions.

Another issue with retail is key management. Retail is going to require a smart wallet. Today’s smart wallets ask a user to download an app and then the key is in the app itself. This means the user is responsible for looking after the signing key(s). If somehow they misplace that phone or can't unlock it, then they’ve lost their key and they’ve lost their money. We have to find a way that the key can be recovered, which by itself is a pretty complex issue to tackle. Recoverable keys or keys that are looked after on a server somewhere adds a layer of centralization and risk to the equation. Beyond key management, what if we're in a place where there's no cell phone signal? Can we get to a place where I can pay you $100 just as long as your phone and my phone are turned on and nearby without needing a network connection? How smart wallets will work in a blockchain-based system still needs to be thought through.

Finally, does the news around digital money, much of it negative in the crypto space, dampen or accelerate innovation from institutions to find a digital monetary solution?

As far as CBDC goes, I don’t think the crypto downturn has an impact. Of all the negative news we've been hearing, it’s not about technology. It’s about how crypto businesses have been set up and run. It’s plain old fraud, right? So most of this news has nothing to do with technology.

One exception to this is with bridges. There's been a stream of bad news about bridges getting hacked and people losing money. A bridge is a very sensitive technology that still needs improvement. It’s a critical component in crypto because assets need to flow from one chain to another. In most cases these issues aren’t related to the bridge itself but rather to how the connection from bridge to the public chain was built. Bridges that connect one permissioned ledger to another permission ledger, versus those connecting to public networks, are stronger because each player's identity is known and we can leverage that additional security. CBDCs to this point are almost exclusively built on permissioned ledgers, which in turn means the bridges here are less sensitive than in the crypto space.