Central Bank Digital Currencies: How Should Privacy Be Built In?

Hi everyone, welcome to Unchained, your no-hype resource for all things crypto. I’m your host Laura Shin.

This panel, from a panel for the fifth anniversary of Hyperledger, features Rob Palatnick, managing director of global head of technology research and innovation at the DTCC and chairman of the Hyperledger board, Matthieu Saint Olive, Codefi payments product manager and CBDC advisor at ConsenSys, and Robert Bench, assistant vice president at the Federal Reserve Bank of Boston. In this discussion on the current outlook on central bank digital currencies (CBDCs), they cover: 

• what main problems CBDCs can solve
• whether CBDCs should be open sourced
• why building a new technology for CBDCs is preferred over using existing tech
• how concerns over CBDCs and their privacy implications differ across countries
• what possible pain points or opportunities CBDCs pose for central banks
• whether CBDCs should be blockchain-based
• to what extent CBDCs will be distributed and open networks, and whether fees would be charged for transactions
• how central banks are thinking about methods of adoption, like whether they will bank directly with retail customers or still use commercial banks 
• how developers balance the drawbacks and benefits of blockchain-based CBDCs with different stakeholders
• whether stable coins will be replaced by or coexist with CBDCs
• and what the future holds for the continued development of CBDCs

Thank you to our sponsors!

1inch: http://1inch.exchange 

Episode links: 

Rob Palatnick: https://www.dtcc.com/our-experts/robert-palatnick

Brian Behlendorf: https://twitter.com/brianbehlendorf?lang=en 

Matthieu Saint Olive: https://twitter.com/msaintolive?lang=en 

Robert Bench: https://www.bostonfed.org/home/people/bank/robert-bench.aspx

Transcript:

Laura Shin:

This week, on Unchained, we’re featuring a panel discussion from the fifth anniversary of Hyperledger on central bank digital currencies. The speakers are Rob Palatnick, managing director of global head of technology research and innovation at the DTCC and chairman of the Hyperledger board, Matthieu Saint Olive, Codefi payments product manager and CBDC advisor at ConsenSys, and Robert Bench, assistant vice president at the Federal Reserve Bank of Boston. In this discussion on the current outlook on central bank digital currencies, or CBDCs, we cover how and when privacy should be built into CBDCs, the role of commercial banks in distribution, whether transaction fees would charged in CBDCs, how to structure a CBDC keeping in mind the constraints around scaling and security, as well as how they’ll coexist with Stablecoins. It was a great discussion, and I hope you enjoy it as much as I did! And now, here’s the panel discussion on CBDCs from the Hyperledger conference.

Hi, everyone. Welcome to our panel. So, we have three great speakers for you. One is Bob Bench, who’s the assistant vice president at the Federal Reserve Bank of Boston. We also have Rob Palatnick, managing director of Global Head of Technology Research and Innovation at the DTCC, and who is also chairman of the Hyperledger board, and then Matthieu Saint Olive, the Codefi payments project manager, and CBDC advisor at ConsenSys. Welcome to our panelists. 

Why don’t we start by having you each go around and stating what it is that you and your organization do, and how it is that you’re either working on central bank digital currencies or other blockchain-related initiatives? How about we start with Bob?

Rob Palatnick:

Sure. Thank you very much, Laura, and thanks to the panelists. And first, as chairman of the Hyperledger board of governors, I would like to congratulate Hyperledger on five years. I think the accomplishments just in helping push software, push the ideas, push the concepts, and push the discussion, and bring together the community in sessions like this has been phenomenal, and part of, you know, the great value that Hyperledger is bringing to the world, as we embark on this journey.

So, I run research and innovation for DTCC. DTCC is Depository Trust & Clearing Corporation. We’re one of the premiere financial industry post-rate infrastructures in the world. In the United States, we’re the primary mechanism for clearing and settlement of equities, cash trades. Most of the non-futures and non-option markets all settle through DTCC. So, we’re not a payment organization. We’re primarily a clearing and settlement, and regulatory reporting, and data organization. 

But we move assets around, and having a new and consistent set of payment rails would certainly benefit our industry.

Laura Shin:

Great…

Rob Palatnick:

Now, over to Bob.

Bob Bench:

Thank you for having me, Laura and Brian. So, my name’s Bob Bench. I’m an AVP at the Federal Reserve Bank of Boston. So, what our team does is we, in association with the DCI at MIT, we are working on a general-purpose central banks digital currency research project. The main idea here, as opposed to Rob’s organization, which handles the securities markets, we are trying to understand what are the trade-offs in trying to build a fiat alternative, fiat-alternative digital currency. And we’re starting at the core layer. 

We think that there are unique design requirements to central bank’s fiat currencies, and we think that that needs to be built, and we’re going to try to build it for research purposes. What we think about is almost a Linux for central bank currencies, and so, we are in year one of this project. We plan on releasing open-source software in summer 2021, along with a white paper, and we look forward to people from the community testing our product, playing with our product, much like Hyperledger and Linux did over their history, and we’re excited to release this and get your feedback.

Laura Shin:

That’s interesting. Yeah, I’ll have some questions about that, but before that, why don’t we turn to Matthieu.

Matthieu Saint Olive:

Hi, so, Matthieu Saint Olive. I’m leading ConsenSys work on CBDC globally. So, briefly, ConsenSys is a software company, a pioneer in blockchain. Our founder and CEO Joseph Lubin co-created Ethereum six years ago, and then created ConsenSys to build a community around Ethereum, and we are really in between public blockchain innovation and enterprise, and strongly believing in this as a convergence of the two, and CBDC is a great example of this.

We are really close to Hyperledger, and really grateful to be part of this community. We have contributed to Hyperledger with Hyperledger Besu, which is an Ethereum clients, open-source Ethereum client, which we submitted almost two years ago, and we strongly believe that Hyperledger brings a lot of value by being this greenhouse with 15 projects, which allows for collaboration, creation of standards, and all of this is really critical for enterprise adoption.

And then, more specifically on CBDC, so, we are really committed to support central banks on CBDC. We are working with six different central banks on CBDC pilots, wholesale, retail, and closed-border pilots. And yeah, right now we believe that central banks do not yet plan to go live with CBDC. They are cautious by nature, and so, they need experimentations, they need to understand, as you said, Bob, what are the trade-off, what are the capabilities, and how to implement it very concretely.

And so, it’s a really exciting time for us, to see this technology being used by such high-profile institutions.

Laura Shin:

So, as I’m sure everyone’s well aware, when it comes to blockchain technology, people always make a joke that the solution to everything is to just put it on a blockchain, but obviously that’s just a joke, because that’s not the case. So, when it comes to central bank digital currencies, what are the main problems that could be solved or alleviated with central, sorry, the problems with central bank money that could be alleviated or resolved with blockchain technology? And this is just an open question, anyone can answer it.

Matthieu Saint Olive:

I think that’s, if I can start, briefly, I think there’s three main problems, in terms of, which are wholesale, retail, and cross-border. So, wholesale, there are already-existing RTGS systems and interbank settlement which work pretty well, but there are delays and costs associated to this, and more importantly, there is lack of programmability, and programmability will not solve a problem, but provide additional value. For cross-border payments, obviously we are still relying on our correspondent banking infrastructure, that is completely outdated, very slow, very costly.

And here, many people believe that CBDC can significantly facilitate cross-border payments. And then for retail CBDC, I’m not sure we solved the problem, because in the developed countries, we have great means of payment, but it can be more relevant for the developing countries, and even for developed countries, it will actually not solve problem, but create new ones.

Laura Shin:

And Rob, do you want to add something for the wholesale side?

Rob Palatnick:

Yes. So, from a wholesale perspective, I guess Matthieu touched on two aspects, to the digital currency, that if realized in a central bank digital currency would be of tremendous value. One is that they would be digital, and they would be a consistent methodology for recording those payments and tracking where the money is moving, and second, that they’re programmable. So, those two things in combination is what really brings that value.

Within the wholesale ecosystem of the financial industry, there are many different process flows that are really unique to every asset class. So, equities has a particular process flow. Various fixed-income transactions have process flows. Money markets have their process flow, and all of these have their own settlement work flow, their own margin models. So, it seems fairly obvious that if we adopt a more efficient payment rail that consolidated all of the cash movements for our clients in the industry, that it would be much easier for our clients to manage their liquidity needs. So, the math of it just seems much simpler.

Laura Shin:

And Bob, did you want to add anything?

Bob Bench:

Sure. I mean, I think, you know, we’re certainly not wedded to a blockchain architecture, but certainly there are advantages there, resiliency across distributed networks. There’s benefits there, and again, our research is primarily around trade-offs, right? So, the challenge with distributed networks is they’re really complex, and complexity adds attack surfaces, and those things can cause challenge to one of our most important goals, which is security.

But…and certainly, the traceability of certain blockchains is very intriguing from an AMLCFT standpoint, right, the ability of firms like Chainalysis and Elliptic to track funds globally fairly instantly, as long as you have some KyAC somewhere along that trail, is really unique and interesting and new for the AMLCFT world. So, again, I think there’s a lot of positive aspects, and we’re looking at those, and what trade-offs exist with those distributed systems.

Laura Shin:

Yeah, I imagine that there’s a segment of the leadership that already has their hackles raised about what you just said, but actually, before we get to that, because obviously that raises privacy issues, I do want to just ask you about something that you mentioned in the beginning, about how you’re working on the solution and you’re going to release the code for it in the summer. So, I’m just curious, you know, if you are working on solutions for central bank digital currencies, then do you see that as something that would be an open-source technology, and a government would build that kind of solution using open-source code?

Bob Bench:

Of course, and governments currently use a plethora of open-source code. Much of the government architecture is built off Linux, and you know, we at least, at least our research team, in concert with MIT, we don’t believe in security by obscurity. We think that if code is going to be secure, you need a lot of people looking at it, and a lot of people working on it, and it’s very hard for that to be done when it’s a very small, classified team.

Again, you know, the Federal Reserve and treasury have not made any announcements regarding central bank digital currency plans to go live in any sense of the word, but we think the best use of our research time is to build something open-source, where we can build off the minds of the women and men who take part in the open-source community.

Laura Shin:

And why are you building something, as opposed to maybe using one of the existing technologies?

Bob Bench:

One is, and to be sure, our division will be evaluating public and private platforms to test our own in-house build, but one, you know, I think it’s really interesting to build something from the ground up. You learn a lot about your platform if you build it yourself. You learn a lot about the difficulties, you learn a lot of the trade-offs that we may not have learned if we exclusively used a provider. I think that’s critically important. 

Two is, you know, I think joining with MIT, we are really building on some of the better brains in this space. Neha Narula, who runs the DCI, has put together a fantastic team of some of the top cryptographers and developers in the world in this area, and we want to see what they can do with the very unique design requirements for central bank digital currency. As far as I understand, there is not a single platform that has exclusively been built for the design requirements for central bank digital currency. 

There’s been a lot of enterprise chains for value, so to speak, but we think central banks have unique design requirements, and we’re going to try to build for that purpose.

Laura Shin:

And Matthieu, you’ve worked with a number of different governments, or ConsenSys is working or has worked with South Africa, Singapore, Thailand, Hong Kong, Australia, and I just wondered, how do concerns differ across countries, and what factors tend to affect their concerns? Is it, like, the size of the country, or are there cultural factors, or is it about kind of the existing penetration of fintech technology, or, you know, how are different governments thinking about this?

Matthieu Saint Olive:

Just before I answer the question, I just want to react on what Bob said, about open-source, which I think is quite important, that one of the great benefits of DLT, with open-source, is that it’s a transfer of values, of value capability, is unbelievable. And so, it’s much different from any other technologies, like database or whatever. Here you have, the protocol itself allows transfer of value, which, and then on top of this open-source software, people will build additional proprietary applications, and make a business out of it.

And these, we see this as the opportunity, also, to foster digital innovation, and the Bank of England has forecasted a boost of GDP of three percent with the general-purpose CBDC, which I think is quite exciting. Then, when it comes to, in our discussion with central banks, there are several things. So, first of all, this technology is new to them. So, the first thing they need is to basically get their hands dirty, and start to understand exactly what are the capabilities, what are the trade-offs, and how they can implement it.

When we look at the technology layers, they have some concerns about how we can manage privacy, how we can manage transaction throughputs, so how many transactions per second you can manage. So, and how will it be interoperable with other systems, whether it is legacy systems or new DLT platforms that might be built with the same protocol, or with different protocols. So, they have this need of understanding the technology, but what’s, as we see the markets becoming more and more mature, central banks are also more focused on what the use case is, what the application.

And here, collaboration between the public, so, the central banks, and the private sector is absolutely key, because they need to understand not only how the individuals will use a CBDC, but how the enterprises, how the private sector will be able to use it, how it will solve their pain points, and how it will allow them new opportunities. And so, there’s obviously a lot of collaboration, very useful with the financial institution, for the wholesale layer, but also for the, I mean, any merchants.

Every merchants need a means of payment, and so, understanding how they can manage their subscriptions, their invoices in a smarter way and a better way, it is very important for the central bank to understand proper designs.

Laura Shin:

And Rob, why don’t you maybe give more insight to what Matthieu just said, and tell us how the DTCC currently works with central banks, and maybe mention some of the pain points that he discussed, as well as new opportunities that you see could come with CBDC?

Rob Palatnick:

And yeah, I guess the way we look at the entire ecosystem in the financial industry, and in the financial markets across the globe, basically, is that they’re helping people manage their retirement accounts, their pension funds. You know, the financial industry serves a purpose that benefits everyone. So, being able to move assets and move value and exchange for those assets, with a complete confidence in the integrity of the markets and in the integrity of your, that your asset is accounted for, or your money is accounted for, is critical.

And DTCC’s primary focus has long been on that resiliency, on the safety and soundness of markets, and that when you make a trade of a security, you buy a stock, you sell a stock, you sell a money market instrument, that when you do that transaction, you get your money or you get your stock. And it is unquestioned that that accounting exists, is reliable, has integrity, is verifiable. So, the markets that exist today have instrumented that and built that through decades of incremental advancement, incremental advancement in the technology, in market practices, in regulation, in oversight.

The Federal Reserve Bank in the United States and central banks around the world have been critical components of making sure that those markets have that safety and soundness, that they can scale when scale and performance is necessary, that in the face of market volatility and disruptions, that they’re resilient, and that everyone always can be confident that those markets work. So, when we start looking at, so, all of that is kind of table stakes. That’s the foundation.

You know, when something goes wrong at 2 in the morning today, everyone knows exactly what happened. That’s the point in time where we’re processing variable annuities, or, you know, there’s something in the process that already is going on there that we’ve been working on for decades. It’s been instrumented. It comes up on your alert screen, you know exactly how to respond. 

As we start moving into replacing components of the infrastructure, and especially the payment rails, you’re going to have to have that same degree of confidence that when something goes wrong, you know exactly how to fix it, you know exactly where to find that problem, and that the system itself is resilient in the face of security attacks, of unexpected, you know, natural disasters, of any kind of disruption, and that the markets can keep on operating.

So, our focus in the wholesale markets has long been making sure that train works, and that all of it works without problem, without fail, and that we can test failure scenarios, that we can try things out. Payments is a critical part of how DTCC processes. The equity markets have a cycle. Trading occurs on exchanges. Today’s cycle is at the end of the day. All day long, those trades are submitted into a organization, central security’s clearing corporations and depositories. 

Then, at the end of the day, traditionally, we have various netting cycles. Some activity settles on a two-day basis, and everyone knows that in two days, all of your settlements net down, and you have to settle, you either have to send a security or receive a security, and then you have to either send a payment or receive a payment, and that all nets. So, 200 million trades net down to one million payments. 

So, in this world with central bank digital currency, making sure that that payment is as reliable and as durable, and has integrity, it’s non-refutable, it can’t be duplicated, it can’t be denied that you actually made that payment, and that it has the same integrity as payments do today, and better, because it’s traceable to the degree privacy and regulations want it to be traceable, and that you can have confidence that once it’s executed, it’s immutable. All of those would add tremendous value to the existing system.

Laura Shin:

So, we’re going to get a little bit more into kind of the technical details around how to structure CBDC on the back end, but I actually want to draw in one of the questions that came in through Q&A from Jacques, and hopefully I’m not going to butcher this name too much, Bicondeau, it looks like, who wrote, should CBDCs be blockchain-based, why or why not? I don’t know if maybe Bob or Matthieu want to address this, but maybe we should just get that question out of the way before we talk about more technical details.

Bob Bench:

Again, I think one thing that we try to reiterate is, in our research, we find that while I think there’s a great value to interoperability among CBDCs to solve some of the cross-border issues. Every country needs to decide for themselves the design requirements, right? Like any good product manager, you’re focusing on your customer, and the customers of Sweden may be very different than the customers for the Eastern Caribbean central bank. And so, you need to solve the problems that your customers bring to you. 

And so, there might be economies that the blockchain may help out, and there might be ones where a centralized database is a lot more appropriate, and I think that’s something that needs to be discussed, is every central bank needs to solve the problems of their customers, like every software company needs to solve the problems of their customers. So, you know, again, like we discussed earlier, there’s upsides, but you need to understand your customer first.

Matthieu Saint Olive:

Yeah, and to emphasize on this, so, we also believe that some CBDC will be built with DLT, and particularly interoperable with hyperledger, and others won’t. And we also strongly believe that there is no one-size-fits-all solution. It’s just impossible. Each country, each jurisdiction have their specific requirements, whether it’s in terms of number of people in the country, or the attachment to privacy. 

And so, each approach will be quite different, but we believe as a core layer, which is brought with DLT, which prevents all the double spending, which is really the core of the technology, to prevent double spending without a central party to maintain a central ledger is absolutely key. So, this is one of the key benefits. The second one is programmability. So, when we imagine programmability, it’s already possible somehow with centralized systems, where one centralized party will say, okay, if you do this action, I will do this output.

What is, again, quite exciting is that with DLT, you can create those Smart contracts that can execute themselves to, depending on an input, they provide an output, and all of this without any central party, and with a significant trust, because anyone can look at it, verify that it does what it say it does, and that’s really bringing a lot of value. And then, what’s even more exciting is the next step, is when we start to have an ecosystem, have such an open ecosystem, which is very different from our walled-garden systems we have today, when we have this open ecosystem, we have composability, which means that products built by one company can leverage the product built by the others.

And it goes really much further than the open API or open API system, where people can connect through APIs, because here, it’s native in the protocol, and that’s what we have seen recently over the last year, with the emergence of decentralized finance, where a lot of protocols build on top of stablecoins, and then the Hyperledgers, and we create this ecosystem that can just benefit from each other’s innovation, and really allows us to create new application and new services for the end users.

1inch is a decentralized exchange aggregator that sources liquidity from the top DEXs and liquidity sources to save users money and time on swaps. 1inch is capable of finding the best possible trading paths and splitting them among multiple market depths. Recently, the 1inch team unveiled 1inch version 2. The main highlights of v2 are Pathfinder, an API that contains a new discovery and routing algorithm, and a new intuitive, user-friendly UI. The v2 improvements ensure the best rates on swaps, while dramatically cutting response time.

Laura Shin:

So, I actually want to ask, this is a structure question, and this kind of goes to what is perhaps, at least to my mind, an inherent tension in the technology. As Matthieu was saying, for a distributed ledger system to work well, it needs to be distributed, and obviously, central bank digital currencies has the word “central” in it, and normally that does come from, you know, what we think of as a centralized institution. So, for a central bank digital currency, what is the back-end structure, in terms of, like, who’s running nodes? 

Is it all, like, a permission chain? How do you spread that out to make it secure, and yet, also, I’m sure, you know, the more spread out, as we all know, the kind of greater limitations on scaling, at least as far as I understand. So, how are you guys thinking about how to structure, given those constraints?

Bob Bench:

So, that’s something we think about a lot. Certainly, I don’t necessarily…a lot of times, we think about whether or not there’s a malicious actor intent, right? And that’s, I think, a lot of what Bitcoin and Ether were based upon, was the assumption of a malicious actor, and the removal of a trusted party, but it’s, certainly there’s a lot of gains to be had from a resiliency standpoint, having a distributed system, and that’s kind of where we look at the upside for distribution in our initial research.

But again, and again, we are exclusively focusing on the retail use case, if you’re going to have throughput that’s comparable to existing retail payment methods, you’re going to have impact there. And so, understanding how you can maximize resiliency but also throughput and settlement finality for the retail use case is one of the most critical tradeoffs. And so, again, we think less about centralization and more about malicious actor intent, when we look about, you know, centralized versus distributed systems.

Laura Shin:

But can you, I’ll answer what I was asking about, in terms of, like, who would be running nodes, and how do you resolve that issue, because…is it just a permission chain, or would there be any way to kind of open it up to other people, or is that, you know, one of the potential attack vectors for, like, some kind of, yeah, unfriendly actor on the network?

Bob Bench:

So, that’s something we’re still looking into. I don’t think we’ve gone that far in our research. You know, we’re still working on our core-level code base. So, that’s something that we’re currently mapping out, but I don’t have a good answer for you on that yet.

Laura Shin:

And Matthieu, is this something that you guys have thought about?

Matthieu Saint Olive:

Yeah. So, from our perspective, so, today, central banks are really focused on permissioned blockchain, where the central bank and the regulated financial institution, like banks and payment service providers will run nodes, and this, consensus will emerge through those regulated financial institution. And the fact that the central bank is, as you say, central, but the fact is, they are the only entity allowed to issue the CBDC, and that’s fine, and that’s, again, one of the benefit of the technology, is that you can program into the protocol itself each roles and responsibilities of different stakeholders.

So, yeah, really, they’re focused on these, and then those regulated financial institution, for general purpose, will be responsible to distribute it to the end users. But still, in this open ecosystem, where there is trust and where the transfers are so simple between one PSP and the other, and this would be seamless. But we also believe that, and I’ll finish with this, so we all…is that we still believe that central banks will, in a couple of years, be actually quite open to public blockchains. 

We always do this comparison with the internet. People initially, when they see a new technology, they do not understand it. They see the problems, and indeed, there are some limitation at the beginning, and they do not trust this technology. But as it is being implemented, as it is being used, initially as a niche, and then by more and more companies, and we already see large banks, large financial institution using public blockchain. 

We do believe that some CBDCs will be either directly shared on public blockchains, or at least that there will be some bridges between the permission network built by the central banks, and the public blockchain, in which other companies are building their own applications.

Rob Palatnick:

I’d like to add a little, just, I guess, from a Hyperledger perspective, and I’ve been around the industry over 40 years, and the pace of technology and the way technology moves is not in a straight line. And everyone kind of wants the answer immediately when an idea came out, but it’s like finding a piece of metal and imagining an iPhone. I mean, the gap between those two is a whole lot of element exercises. 

So, I think, you know, it’s a tribute to the ConsenSys, Matthieu and his company, and Joe Lubin, who’s a colleague of mine on the board of Hyperledger, that the Ethereum community started out as very public-focused and public chain. But with Besu and their contributions to the community, they’ve made this tremendous investment in private enterprise models as well. It’s a tribute to Bob Bench and Neha, and the work they’re doing on evaluating new security models, because security and threat vectors are ever-present and ever-increasing.

And the role of Hyperledger, with its greenhouse and its many projects, is to bring together lots of different dimensions. The game isn’t over. It’s like, you know, when the book on computers was written 50 years ago, and saying, all right, that’s going to be the model, and it will have vacuum tubes, and you flip these switches, where the first book on relational databases was written, and then that was the end of it. It’s a progression, and we’re all learning from each other.

The fact that we have these conversations, and DTCC apparently is the only one on this panel that is not writing its own blockchain software, but we’re investing through our, we have a couple of prototype projects. Project Ion, that’s doing clearing and settlement on DLT, is a fully functional prototype, and we have this Project Whitney, which is a private-market securities issuance, and we’ve been able to write those on multiple blockchains, both public and private, to understand them, to see, just the way Bob was explaining, to see what their scale dimensions are, what it means to put out different nodes, what it means from a resiliency perspective. And we’re all sharing, and we’re all learning from each other. So, it’s a progression.

Laura Shin:

And something else that I was curious about, before we dive into what I think is probably going to be a somewhat meaty topic, which is the privacy issues that came up earlier. I was just wondering, you know, for normal, public blockchains, transactions are typically paid for with a fee. Would that also be part of a central bank digital currency blockchain?

Bob Bench:

So, I guess I’ll jump into that question, because that’s an important one. There are very specific laws, I should say, around how a central bank can charge for its services. Right, so, that is a constraint on a central bank build, at least with regards to United States. There might be other central banks that may model it differently, but I think questions about fee, and I guess the later question about privacy, are really important when you think about things like spam on network, and DDOS attacks.

That’s something that isn’t really talked about in CBDC conversations a lot, is the simple question of spam, but stopping spam is really important. And fees is one way to do it, and having really robust identity structures is one way to do it. Those are two of the easier ways to fix that problem. And so, understanding legal and policy constraints with regards to spam and DDOS attacks on one of these platforms is something we think a lot about, but as far as the US is concerned, there are constraints around fees for any central bank payment process.

Laura Shin:

Okay, well, let’s segue now to the privacy, or Matthieu, did you want to add something?

Matthieu Saint Olive:

Yeah, just wanted to add one thing, is you are comparing the transaction fee in public blockchain and CBDC. If it is built on private blockchains, even if we still have what we call gas to pay for the transaction, it might not need to have a price on it. And so, the reason why it has a cost on public blockchain is that you have to pay for the validators who participate in consensus. And so, it is slightly different, and might not be required for CBDC.

And I also wanted to emphasize on what Rob said, and finish with this, that all DLT platforms, all blockchains are not created equally. And so, picking the right platform that fits your specific use case will be very important, and even if at ConsenSys we are really focused on Ethereum, we also strongly believe that there are great alternatives out there, and that will emerge and be widely adopted. It just depends on what use case you are focused on.

Laura Shin:

Okay. Yeah, so, now let’s talk about what Bob mentioned near the beginning, which I think is going to be of interest to a lot of people. He talked about how CBDCs have the potential to have built-in KYC, which is know your customer, and AML, anti-money laundering, processes built into them. And I think that, obviously, you know, people have a lot of privacy concerns about CBDCs, particularly, maybe because one of the first, or the first to really be rolled out is the DCEP in China, which, there’s, I think, a surveillance culture over there when it comes to their technology.

So, how are different central bank digital currency players thinking about concerns over privacy? What are some of the options for managing privacy when it comes to CBDCs?

Rob Palatnick:

So, I guess I’ll hop in there. I think, with regards to our research, you know, I think the question of privacy in the United States remains one that needs a lot more discussion, and is a critical policy question, but certainly a question that informs technology. You know, when we started our project, our main focus was, how can we make something useful for retail purposes, primarily meaning extremely secure, with throughput and finality equivalent to leading retail payment systems?

But one of our learnings has been that you need to start thinking about privacy early in the stack. You know, one thing that we’ve been educated by the MIT folks is that the later you add privacy to a system, the less private it becomes. And so, that’s something that we are thinking a lot about, and certainly, we are not a policy team. We’re exclusively a technology team, but understanding where a policy needs to enter into the platform is absolutely critical, and our, I think more advanced discussions need to happen to get a better understanding of sentiments around privacy, with regard to payments, because I think it’s still largely undefined, at least over here.

Laura Shin:

Well, one thing that I was wondering about is, you know, Zcash and Monero have their viewing keys, or ways to take a shielded transaction and have, and I guess prove certain aspects of that data to specific people. Is that an option that’s being considered?

Bob Bench:

We’re not, like, going directly into, as they say, Zcash or Monero. We have cryptographers that have worked on both platforms deeply, that understand those methods very, very, very well. But that’s not, like, we’re not modeling any of our systems to date off of Zcash and Monero, any platform like that. You know, without question, the state will have to perform its financial intelligence functions. I think any institution, whether it’s us, whether it’s DTCC, any large banking entity, those rules are not going away, barring a change in legislation.

So, we need to understand all of our customers’ needs. So, that being the uses of the dollar, and the members in the treasury department that have jobs to do, to stop financial crime. So, that’s a balance that has to be done. The key thing for our platform is understanding where do we put that privacy, because I think regardless of how the privacy works, we need to make sure that if data is collected, only the data should be seen by the people responsible for seeing the data, and no one else.

And I think that’s the hardest thing, is once you collect data, it’s on you to secure it. And so, that’s the most important thing for us, is making sure whatever we’re required to collect is absolutely secure.

Matthieu Saint Olive:

I think that I agree, privacy, there’s a lot to be discussed still, and I think technology is not the bottleneck. What’s more important is the decision of the central bank of how they want to design it, but when we look at the wholesale layer, so between the central bank and the financial institution, I think it is commonly agreed that the central bank, the banks do not want other banks in the network to see their transactions, the value, the quantity of activities.

So, this has to be made private to the other banks, but for the central bank, the central bank wants to have access to this information, whether natively or thanks to a lawl or whatever. They want to be able to have access to this information. So, I think that’s one point, and then, for retail application, there are different approaches, as you say. As far as I know, in China it will be quite open, so even if the identity of the users are not directly visible on the CBDC platform and the CBDC ledger, you will see all the transaction.

You will see history of transaction, and so, you could deduce some of the identities. So, this is one option, and then, there are honestly a lot of technical option to abstract or to create additional layers of privacy. Just to mention it, there’s the most adopted ones, there are ideas around implementation, leveraging zero knowledge proof, or privacy groups. So, there are basically different approaches, which, yeah, will depend on the use case, and the jurisdiction, and the institutions.

Laura Shin:

And so, now, let’s talk about adoption, because there, you know, just with the current system, this will be quite a shift if something comes to pass with CBDCs. So, in general, how are a lot of CBDs, sorry, how are a lot of central banks thinking about how to get CBDCs adopted, in terms of, you know, whether or not they’ll use commercial banks the way that they’re used now, or whether they’ll have their own retail accounts, or just, what are they thinking? Or is it just too early to even ask this question?

Bob Bench:

Yeah, sure, I’ll start. Starting with the banking intermediation, that’s a critical question, right? There’s really, really important policy implications to change in the current intermediary model. Some countries, such as China, have, are going with a two-tier model. There are certainly advocates for continuing the two-tier model here in the United States. That’s something that has, like we said, tradeoffs. There’s cost and benefits to that.

Certainly, a direct model to a central bank requires a material operational uptick by the central banks’ activities. Doing retail compliance, retail servicing is hard. It’s a lot of work. It’s very different than wholesale, and I think that’s something central banks, would be fairly new to most central banks. I think there’s a world, some central banks are looking at payment service providers, and the technology community has gotten very good at directly handling retail customers.

That’s something that we need to learn from as a central bank, of how they do it, and what makes them so good at it, but again, I think that’s something that is what we call a Phase II question. We don’t think the core codebase really changes much, if you go through a two-tier model, or a direct model, or a payment service provider model, but I think what’s critical is institutions like the Federal Reserve understand the impacts of any change to the model, right?

So, what does that mean for small and community banks, if their deposit base erodes, to some extent, because of CBDCs? That’s something really important for, say, community banks, and the communities they serve. So, again, that’s something we’re looking at closely, because we need to understand those tradeoffs.

Laura Shin:

But just out of curiosity, so, I understand the discussions may not have gone too far in this, but if central banks were to deal more directly with retail customers, does that kind of enable them to do things that they can’t currently do with the model, using commercial banks at the moment? Like, you know, are there sort of, like, pros to switching things up?

Bob Bench:

I think there certainly could be, right? I think there’s…so, the FDIC does a really great research every year on the unbanked and the under-banked. That is something that is a really thorny question, and has been really hard to fix. What we do know from the FDIC study is that certain people are unbanked because they choose not to enter into relationships with banks. An open question is, would they enter a direct relationship with a central bank? Maybe. Maybe not. We don’t know that answer.

But that is something that, it’s seen at least by some research as a positive. Are there certain unbanked persons who would rather work directly with the government? That might be true. And so, those are the kind of things we look at. Certainly, there have been advocates if they had direct accounts. So, again, another issue where government benefits could be more quickly received, but a lot of research has to happen, and a lot of tradeoff research has to happen, because there’s a lot of stakeholders.

I mean, you know, you look at how many, Rob, I’m sure, with DTCC, the amount of stakeholders you have in your environment is enormous, and getting all those people at the table to agree on a new model is extremely hard. And so, you really have to understand all the issues when you bring all those people to the table, because it’s a dramatic change.

Rob Palatnick:

Yeah, I mean, I think one of the general first laws that we all adhere to is do no harm. So, I think there is a lot of different ways of looking at that, the alternatives that technology can bring, but there are definitely some capabilities, like the comments we’ve all made about resiliency and security, that have been quite durable over many decades, and we don’t want to lose a lot of those values. 

I am partner and advisor to the Digital Dollar Project, and that does, is advocating for a two-tier system, a wholesale system that involves interactions between the central banks and the banking tier, and a second tier that’s retail-oriented. I think there are a lot of different models, and I’m sure a lot of the work that Bob is doing with MIT and some of the other academic institutions that we’re all speaking with, that are part of Hyperledger as well, are going to contribute in a great way over the next few years, to advancing thought leadership on this.

Laura Shin:

Well, Rob, I actually wanted to ask you a question about what you said about how, you know, you’re not trying to do too much damage with whatever changes you make, or do no harm. But you know, this technology, obviously, could bring a lot of benefits if it were to do more than simply just keep the existing processes in place. So, in that regard, like, how do you kind of work with the different stakeholders to talk about what different benefits you could get if you were to embrace more disruption?

Rob Palatnick:

I completely agree with you. So, let me just make that clear, and I think most of us that are in, certainly in this conversation, but in this ecosystem, and that are trying to push the technology, agree with that idea as well. I just, I think we’ve got this tool, and we’re a little bit afraid of the power of this tool. It’s like a lightsaber in the hand of a three-year-old. And there’s a bit of, you know, to some of the questions that are on the Q&A and the earlier question before, do we need DLT, distributed ledger, to answer central bank digital currency, or even increasing or improving the efficiency of equity settlement, or any asset settlement?

The fact that the entire world is basically spending a tremendous amount of energy on reconciliation of information constantly, that everyone on this call has a different view of what a ledger is, of how much money or how many assets they have on the ledger, and that we have to go through a variety of different protocols to reconcile what we all should know as a central truth, is something this technology can answer. Does it need to answer it in a broad, public, either proof of work, proof of stake model?

Is it able to trust certain entities, like federal banks, like governments, like certain institutions that were created specifically for trusts? How do you trust them in the, you know, event of security and malware onslaughts, and various actors looking to subvert that? These are all questions that need to be answered, but the basic premise that this technology can enable much more efficient interactions, much more efficient tracking of sources of truth, absolutely true.

So, I think there’s a willingness on everyone’s part to take a fresh look at those, you know, long-held beliefs that our traditional siloes on mainframes were, that needed to be reconciled with other mainframes, was the way everything had to work. We’re ready to throw that out, and say, let’s move to something new, but we need to do it with, you know, with all the concerns about safety, soundness, and protecting everyone’s retirement account, everyone’s investments, everyone’s pensions.

You know, so, there’s that balance of, how do you disrupt, but address the concerns of people with real assets, that want to make sure that they can pay their rent.

Laura Shin:

And I actually want to call another question by Jacques Bicondeau, which is, should CBDCs coexist with stablecoins or replace them? I don’t even know if this is something that this, you know, that you are considering, but maybe, I would be curious to just hear what your thoughts are in that.

Rob Palatnick:

I think that’s more towards Bob, and the Fed. I mean, I think the aspect of stablecoins right now is a way of creating equivalence with the crypto ecosystem, and what that evolves into once there’s additional, once there’s viable fiat currencies issued by governments, and viable other types of currencies issued by organizations like Facebook, and consortium that support that. How that balances out with the values underneath cryptocurrencies will be an interesting thing we’re all interested in seeing. 

Matthieu Saint Olive:

I can jump in. So, I think there’s two different things with CBDC and stablecoins. One, and it’s about the aspect of money, either it is a store value and a means of payment. And I believe that it’s unlikely that stablecoins will become widely adopted as a store value. For instance, even Libra as a Facebook project, people will pay with this, but you will not keep your earnings and your salary in your WhatsApp wallet. It just doesn’t make sense.

And so, the way we derive this is that we believe that some stablecoins or CBDC will be more appropriate to specific use cases. And so, for instance, we see on the public blockchain MakerDAO, which is a stablecoin which worked very well, but…and the way it is built and organized, it’s fully decentralized, it’s done by a community, and it will probably not make sense for the DTCC to use it. The DTCC will very likely prefer a CBDC issued by the Fed, but it could make sense for a number of projects that happen in the web3.0.

And similarly, we believe that some stablecoin will be preferred for niche use cases, and will basically coexist with CBDCs.

Bob Bench:

Yeah, so, with regards to stablecoins, you know, we think about this a lot. I spent my prior career helping build one of the largest stablecoins out there…

Laura Shin:

USDC.

Bob Bench:

…yeah.

Laura Shin:

At Circle.

Bob Bench:

Yes, and I think most stablecoins right now, aside from Dai, are commercial-bank deposits on a new rail. It’s commercial bank money, and I think that’s fundamentally different than central bank money, and I think that, you know, for the countries that bring out CBDCs, I think that they’ll continue to coexist with stablecoins. I think the stablecoin use case will continue, and I think that they will coexist. I think there is a use case for direct central bank money in a digital form for retail persons.

Whether or not the United States chooses to do so is a major policy decision above my pay grade, but I think, certainly, countries will, certain countries are going to go forth with retail CBDCs, and enable retail persons to have central bank money, but I think stablecoins will continue to exist with, through the commercial banking system, moving commercial bank money on new payment rails, and are going to achieve significant innovation through that way.

Certainly, there’s brilliant women and men working on the problems that can be solved with commercial bank money on these rails, and certainly our team looks to learn from them, but I think there are fundamentally different use cases for commercial bank money on these rails and central bank money on these rails.

Laura Shin:

All right, so, we’re going to have to wrap up, but before we go, why don’t you all each say kind of what you expect going forward, in terms of the next steps on this journey, or, you know, what questions you’re looking to resolve as these different efforts continue?

Rob Palatnick:

I’ll start. So, we’re actually building a fully functional prototype we call Project Ion, working with our financial industry clients, to simulate clearing and settlement on a distributed ledger, in a accelerated settlement-type model, where you can optionally say, I want to settle this trade immediately, I want to settle this trade in a hourly net, I want to settle this trade in two days. So, something that combines the current ecosystem and a digital asset ecosystem.

We’re expecting to implement a cash token as part of that prototype build, but we’ve been talking to the Digital Dollar Project about making this an eligible pilot for central bank digital currency and the wholesale markets. So, we expect to be active, working with our industry, certainly working with the Fed, our payment rail, on different models, and if we can help advance the thought process on, you know, how this could work in the wholesale markets, we’re all in.

Matthieu Saint Olive:

May I say, I think we are kind of at a turning point, where in the last couple of years, central banks have been testing the technology to understand whether or not it’s technically possible to create a CBDC. And I think when you look at all the reports, all conclude that it is possible. There are some things where the technology has to be pushed further, but basically, it’s possible, but now, what’s really exciting is central bank will start to work with the private sector to identify clear use cases, clear needs of how this CBDC can be really used and adopted, and that will make things quite interesting.

Bob Bench:

I think over the next couple years, you’re going to have a lot deeper understanding of the use cases here. I think the technology certainly makes this possible. Transaction throughput isn’t really an issue, but understanding why people want this, and convincing them, and particularly the wholesale, the world that Rob and Matthieu have been dealing in recently, I think that’s really compelling, because you don’t have the thorny issue of privacy and identity.

And I think there could be compelling, compelling developments in that front, if you get the right stakeholders in the room. But I think, you know, policymakers are going to start catching up with technologists in this area, and I think that’s going to be really, really helpful, folks like the Digital Dollar Project, ConsenSys, Hyperledger. Educating policymakers and getting them up to speed is really going to enable this technology to be better, and understand how we can use it to help all the economies who want to approach this in a thoughtful way, I think is going to be very interesting.

Laura Shin:

Great. Well, this has been a fascinating discussion. Thank you all so much for participating, and thank you to Hyperledger for hosting, and yeah, we’ll have to resume this conversation at some point in the future.

Rob Palatnick:

Looking forward to it.

Bob Bench:

Thank you, Laura. Thank you to all the panelists.

Laura Shin:

Thanks, Brian.

Rob Palatnick:

Thank you, everybody.

Brian:

Bye.

Laura Shin:

Bye.

Thanks so much for joining us today. To learn more about CBDCs, Rob, Mathieu and Bob, check out the show notes for this episode.

Don’t forget — you can now watch video recordings of the shows on the Unchained YouTube channel! Go to youtube.com/c/unchainedpodcastand subscribe today!

Unchained is produced by me, Laura Shin, with help from Anthony Yoon, Daniel Nuss, Bossi Baker, Josh Durham and the Team at CLK Transcription, Inc. Thanks for listening!