Episode 230 – Talking Central Bank Digital Currencies with Jim Cunha, Former EVP, Federal Reserve Bank of Boston

Yvette Bohanan

January 17, 2024

POF Podcast

Launching a payment product requires big moonshot thinking alongside a lot of agonizing over the details. Understanding the market, the customer journey, and the technology – from every angle requires patience, persistence, and a beginner’s mindset to learn from every success and failure. Thinking about launching a new form of currency, however, is on another level altogether.

There’s an entire population of a country to consider – what does each segment of the population require to be able to use the currency? Then, there is the currency’s role in supporting the economy. How will a new currency be protected from counterfeiting, and what controls will be required to prevent its misuse? How will it work with legacy payment systems and commercial banks? And how will it adhere to laws in place today and be prepared for tomorrow? Last but certainly not least, which technologies will enable all of these things and more?

These were just some of the thoughts bubbling through my mind when I sat down with Jim Cunha, former EVP at the Federal Reserve Bank of Boston, to discuss the capstone project of his career – Project Hamilton – and the questions he and his team sought to answer as they embarked on their CBDC experiment. I hope you enjoy hearing Jim’s insights on digitizing the world’s reserve currency – I sure did.

 

Yvette Bohanan:

Welcome to Payments On Fire, a podcast from Glenbrook Partners about the payments industry, how it works, and trends in its evolution.

Hello, I’m Yvette Bohanan, a partner at Glenbrook and your host for Payments on Fire. In this episode, we are exploring the work that the US Federal Reserve has been doing in creating a central bank digital currency. Central bank digital currencies or CBDCs are the digital representation of government issued fiat currency. We have all used physical fiat currency, cash and coins, and we simply trust that it works. I hand you $10, you use it to buy lunch. The restaurant deposits the physical fiat at their bank and their account balance goes up. Most likely though, throughout our lives, we have been giving very little thought as to how it works. And in our workshops, many frustrated participants will ask, “I use digital wallets and credit cards all the time. Why do we need CBDCs? Aren’t payments already digital?” That’s a great question.

The answer is this. Traditional payment systems such as wires, cards and ACH are electronic payment systems that use accounts. The key here is that the value is incremented or credited and decremented or debited on account ledgers at a bank, or in the case of a closed loop system, a payment provider like PayPal. And digital wallets like PayPal or Venmo use digital technologies for initiating account-based payments. Building a new system from scratch, like CBDCS offers new opportunities to rethink how the system works. The system may be account-based or token-based. That’s an interesting point and one that we’ll be diving into with our guest.

Central banks around the world have been investigating CBDC in wholesale and retail payments. The technology works. Most pilot results for domestic systems have had positive outcomes. Cross-border pilots have already been conducted, but technology is not the only issue being explored. CBDC’s are forcing a rethink. They’re asking us to dig deeper into the role of fiat currency in an increasingly digital world. Trust has to be established between the end parties in the system. This requires regulations, rules, and risks to be examined in a new paradigm.

Big questions are surfacing. Will introducing a CBDC create tangible benefits for a country such as greater resilience for domestic payments, or is it simply redundant, an expensive technical novelty looking for a problem to solve? What about energy consumption for minting, managing and destroying digital fiat? And then there’s the question of visibility. Who can see our payments? How does this impact privacy? How does it help fight money laundering and crime? Are CBDCs better or an alternative to stablecoins, or will we need both?

In this episode, I’m joined by Jim Cunha. Jim was an Executive Vice President at the Federal Reserve Bank of Boston. Among his many contributions over the years, Jim led Project Hamilton, the work that the US Federal Reserve has done to explore a US CBDC. Jim, thanks so much for being here today, and welcome to Payments on Fire.

Jim Cunha:

Thanks. It’s a real pleasure, Yvette, and I’m really looking forward to the session.

Yvette Bohanan:

Me too. This is going to be a fun conversation. I want to start just by acknowledging that you were at the Fed for 38 years and congratulations on a terrific career and on your recent retirement, although it seems like you haven’t really retired because you’ve been pretty active on podcasts and panels from everything I’ve gathered.

Jim Cunha:

Yeah, still doing a little bit. I’d like to give back and I also want to just keep vital and see how things unfold. As I said to my wife, if I do some work, great, if I don’t, great. So it’s really a nice balance of trying to keep busy and stay connected, but not overdo it.

Yvette Bohanan:

Yeah, it sounds perfect actually. And from your vantage point, which we’re going to get into here a little bit in your understanding of the space, it’s probably hard to walk away cold turkey because there is so much going on right now, so it’s a long-

Jim Cunha:

Yeah, it took me a long time, but I figured my passion is the nexus of technology and money. My whole career has been there, and I didn’t realize it at first until I started really veering off into various topics, but that’s it. It’s hard to not pay attention and I try to contribute.

Yvette Bohanan:

Tell me a little bit more about how you found yourself at that nexus of technology and money. Can you take us through your career journey and how this all sort of evolved? Because you ended up in a very interesting spot.

Jim Cunha:

Yeah, very exciting spot. So I graduated with two degrees, accounting and philosophy, and so my first job out of college was working for a bank in Rhode Island for four years, and I realized that the banking environment and the commercial side of things, this wasn’t what I wanted to do. I wanted do something more in the public sector. So I moved to the Fed in 1984 and I was a technology auditor, and then eventually moved into technical support the old mainframe days and then get drafted into what’s called Retail Payments Office, which you’re familiar with. So I get drafted into that after being a technologist for a long time, and we were running ACH and check. So I really get deep into payments during that time working on those payment systems.

And then I started doing sort of creative work for the US Treasury, piloting new technologies. We piloted, for instance, something called an eCheck, which was a digital representation of a check that was emailed digitally signed, and this is in 1996, so this is quite a while ago. Then moved into testing a prepaid card for the military back in 1999. And so I really get into sort of the leading edge of more of an entrepreneur inside the Fed building systems out and then fell in love with in the ’90s Mondex. That was a digital e-money.

And so I actually started studying researching digital currencies for the Federal Reserve unofficially on my own back in 2002. We didn’t actually end up doing work as a result of that, but I’ve always been at the leading edge with technology and money. To me, it’s fascinating to try to understand technically what’s possible and then what kind of issues they create. So that’s really been my career, is creating things with the treasury and then at the tail end creating this digital currency, central bank digital currency research work. So it’s a great way to have ended my career.

Yvette Bohanan:

Wow, that’s a huge arc and a journey in touching so many different systems. ACH, checks, this, that’s quite amazing. And Mondex, that’s like the Mondex, what was it? e-gold, e-money?

Jim Cunha:

Yeah, back in that day. So it stuck with me that that’s why I really started understanding my passion was technology and money coming together back around the late in the mid ’90s, et cetera, and I was able to continue to do that for the next couple of decades. So very lucky.

Yvette Bohanan:

Very cool. Very cool career journey. Okay. So the Fed has been investigating Central Bank digital currencies or CBDC, both the wholesale and the retail version of them. And when did you officially, or when did the Fed officially start looking at this seriously? What year was that?

Jim Cunha:

So we started in 2015. We started looking into blockchain and decided to, this is the Boston Fed. This is not an official policy move. This is really just the Boston Fed deciding we wanted to do some technology research. So we just had a volunteer army and we created what’s now called a wholesale digital currency, but basically we put the books of the Fed onto a blockchain, specifically Ethereum. This is back in 2016 when Ethereum was just created.

Yvette Bohanan:

Wow, that’s really early. That was really early. Yeah.

Jim Cunha:

Yeah. We found out that the Bank of England, Canada, and Singapore were doing the exact same experiments as we were doing at the same time. And so we started communicating with those central banks, but our goal was not to really see if blockchain could be used for a wholesale digital currency. We wanted to understand an experiment with the blockchain itself. And the project we picked just happened to be, we think made more sense to be able to do a proof of concept. And then in 2019 we decided to pivot to something else, and because of a number of different things going on around the world, we decided to look into retail central bank digital currency. So we pivoted and I got a team of 10 in a real budget in 2020 to do that particular work, because we were seeing what’s happening around the world and where the pressures were on to better understand what can happen with a retail CBDC from a technology perspective.

Yvette Bohanan:

Okay. So let’s double click a little bit on some of the things that have sort of materialized since then. There was Project Cedar that was published in 2022, and then in July of 2023, the most recent results of what’s referred to now as the Regulated Liability Network or the RLN.

Jim Cunha:

RLN. Yeah.

Yvette Bohanan:

We have CBDC, RLN.

Yeah. How many more acronyms?

Jim Cunha:

Tony McLaughlin’s baby. Yeah.

Yvette Bohanan:

Okay. So that proof of concept, those results were announced. What are things, as I was reading through some of the work that they’ve published here on this was that the RLN proof of concept, they kind of started upfront saying payments are already fast, and ISO 20022 carries a lot of data, a lot of information with the payment transaction, which is super valuable and really cool in and of itself. And so the thesis that RLN was really exploring was improving settlement. And from that respect, what were the real significant learnings coming out of Cedar and of RLN proof of concept?

Jim Cunha:

So they’re trying to solve both domestic and international cross-border, but in different ways. So Project Cedar, and I’m not an expert, but I work with the folks there, Project Cedar was trying to interconnect two different disparate ledgers in a way that allowed you to do atomic swaps across those ledgers. So think about moving the money on two different ledgers and with the atomic swap is the money moves on both ledgers or neither. So they were interconnecting ledgers. What the RLN is really doing is having a single ledger where you can have multiple assets, commercial bank or the tokenization of the commercial bank money, central Bank money, and even securities theoretically.

And what they were trying to do is to even speed up the time and reduce the risk of a settlement problem, because there still are settlement issues, especially if you think about something like delivery versus payment of securities where the security and the money has to move at the same time. So there are still risks out there and what they’re trying to do in both projects was trying to solve for both speed and settlement risk, but also the atomic nature of the fact that all the money has to move at the same time or none of it moves. So they’re really trying to solve.

And then what’s valuable about the atomic swap side of it is most of the experiments around the world outside of even these have proven that just having a wholesale central bank digital currency, you don’t really gain a real lot, but if you have a wholesale digital currency and can move that currency and the security at the same time, delivery versus payments, that’s the real business case, or moving to currencies on different types of technologies as some of the experiments through the innovation hubs out there, that’s where the real value is, is trying to do multiple currencies. So both projects are trying to reduce that risk and also speed up. And while they work well, when you go across-border, things don’t work as well, especially with the cost, the opaqueness, the potential of data loss, et cetera. So they’re really trying to get rid of some inherent problems with some cross-border work that because of the way those systems are built are inherent, I would say.

Yvette Bohanan:

Right. And so let’s sort of take that to the next logical step, which was getting to that retail space and the retail digital currencies and Project Hamilton. So just to clarify one small point here, a lot of people think that that refers to Alexander Hamilton, is that the case?

Jim Cunha:

Well, partially. So my teams, Bob Bench was my second in command, and so when the team was trying to come up with a name, because you need a cool name for a project.

Yvette Bohanan:

Absolutely.

Jim Cunha:

We thought this was a moonshot because we really thought this project was going to contribute to something important. So we thought it was a moonshot. We were working with MIT as our partner, and so they did some research and they found Margaret Hamilton was a technology leader in the ’60s at an MIT Lab, and her team was responsible for some of the guidance system of the Apollo space program. So it was really a nod to Margaret Hamilton as much as the obvious Alexander Hamilton. So that’s the history of the name.

Yvette Bohanan:

A little bit shared lineage there in the naming.

Jim Cunha:

Yes. Yeah.

Yvette Bohanan:

That’s very cool, very cool. So how did the Project Hamilton thesis differ from the RLN and proof of concepts and Project Cedar and the settlement and asymmetry risk that they were going after?

Jim Cunha:

So we were trying to, one, prove whether the technology could be used at scale. So when you’re talking about retail, first of all, you’re talking about whole different constituent. With wholesale, you’re talking about very sophisticated banks and large financial institutions of other types. And so they’re used to these types of risks that they’re dealing with, and it’s a small number of very educated institutions. Whereas with retail, one problem you have is that your grandmother’s going to hold it and have to figure out how to make it work for an individual, are you got people that worry about different kinds of things like privacy in a different way. Scale is a big problem because instead of doing thousands of transactions per second, we were looking at a goal of 100,000 transactions per second if it was going to have uptake and it needed to be very fast.

So for instance, I think it was in the RLN or maybe the Project Cedar, I apologize, I was boning out this morning. They had settlement under 30 seconds. Under 30 seconds is impossible for retail payments, you need under five or two seconds in order for retail. So you get speed, you’ve got the audience you’re working with, you’ve got regulations are different, and especially if you layer in cross-border, now you have a whole more complex sort of thing. So wholesale is not easy, but retail is a lot harder for many reasons than the wholesale front as far as the types of technology and even the policy issues that you have to grapple with. It’s just different in a retail system. And questions like, if I lose my money, do I get it back, or someone steals it, do I get it back? Those in a retail sense, that’s a very different, and then there’s the volumes of data in the speed.

So you mentioned ISO 20022, you would never put ISO 20022 on a retail payment system in my mind because it would just slow it down too much. I’m not talking ACH, I’m talking about something that as you said, is a calf replacement and the volumes and it has to work offline when two people out in the desert. You need and it has to comply with ADA. So all sorts of things. Really complicated retail versus a wholesale system.

Yvette Bohanan:

Wow. Yeah. Okay. So you broke this big problem down into some tracks and some phases. So you had track A with two phases and track B. Can you sort of describe what you were trying to get at with each of the tracks and phases?

Jim Cunha:

Yeah. And we didn’t go as far with all of these as we thought we might, but so the first phase of work was to build a core system. If we can’t build something that can do 100,000 transactions a second, settle under five seconds, be resilient. In other words, if we lose a data center or a cloud environment, it has to keep going. Your money can’t be down, and we had to maximize privacy. So the first phase was to match all those things because if we can’t build a system that’s fast or secure or robust enough, then we should stop. There’s no reason to try to go further. So phase one, which was published I think in February of 2022, did that. And we had two systems, one that went 170,000 transactions per second, one was 1.2 million transactions per second. So we accomplished everything we were meant to and we did it in a way that was in the cloud such that we could take down two data centers and still operate. So we think we’re very successful in that.

And the second phase of work was supposed to be looking at various really hard problems and see how many we can solve. So for instance, the quintessential problem with retail is privacy versus compliance. You want to maximize privacy but also catch the bad guy. So how do you think about that and how do you think about moving that money on a ledger in a way that allows you to ensure that transactions are done and done completely and look at privacy? So we had a long list of things we were going to get done, because of timing, we started getting into just a few of those particular problems sets.

And the reason we’re doing it is because we thought new technology could allow us to allow better and different policy outcomes. Right now, if you look at how privacy and compliance is done, it’s messy. It doesn’t work great. Banks hate doing it, and there’s a lot of both false positives that aren’t caught. So we’re thinking if we could possibly solve that with brand new cryptography and new ways of trying to examine the data but keep it private, that we could offer a new solution for the future.

So we started looking into, so I won’t get too deep, it’s called zero-knowledge proofs, which is a way to keep data private, but still examine it for say an OFAC list. So we ended that work at the end of the very beginning of 2023 with MIT. That was all in partnership with MIT. They were partners with us. Then the Fed, and I was not part of this next phase as I retired, they’re now looking at the next phase of work like how do you think about the next phase of experimentation that has to be done? And I don’t believe they publish yet what their game plan is for that. We also were looking at all of the, and this we did from scratch. We designed and built it ourselves. We also looked at platforms that were out there by known actors just to see what’s already in the market that we may be able to use. And when we did that research, it wasn’t for publications just to understand if we quickly had to build a system, what might be a potential buy versus build type of option.

Yvette Bohanan:

You went with build, like seriously build like C++ programming-level build?

Jim Cunha:

C++ Nacora. Yeah. And we had engineers, we had architects, we had this brilliant engineer who graduated MIT, then joined the Fed, James Lovejoy, and he’s just amazing. He did most of the coding. And then the people at MIT we work with, we got time from various people, including some developers that actually developed Bitcoin or created what’s called the Lightning Network and someone who actually created zero-knowledge proofs with Zcash. And so we had some of the best minds in the world at MIT and I think some of the best minds on my team who all came out of the crypto world working on this, and we built from scratch, because you learn a lot more when you build from scratch. And that’s what our goal was, is to learn and to think about things differently. We didn’t want to just build a payment system that looked like a regular one. We would have FedNow and call it money. If we did that, we don’t think we’d be potentially solving some thorny issues if we just did that. As good as Fed now is.

Yvette Bohanan:

Yeah. And we’ve been talking about FedNow, so we’re going to stick to our knitting here today. But when you’re bringing up some really interesting things around use cases and how people could apply this to different use cases like a person payment in the desert and at point of sale in remote commerce scenarios and income, and each one of these use cases has little wrinkles to it. So maybe talking a little bit about some of the nuances there that you were recognizing and telling us a little more about if I lose my digital cash, is it gone forever? Because that’s a big problem, right? For people.

Jim Cunha:

Yeah.

Yvette Bohanan:

Where did you land-

Jim Cunha:

Yeah. The challenge we have with Hamilton was we didn’t have those policy decisions, so we’re trying to build something that was flexible enough that we could alter it for those decisions. I remember talking to my president at the time and I said, “Well, it’s a question, if you lose your money, am I supposed to get it back for you?” And he said, “Of course.” I said, “Why? That’s a policy question. Maybe yes, but I wouldn’t assume that.” And so as we looked at use cases, we said, well, the most likely one is a person-to-person. But then you get into, okay, well, what if there’s no cell coverage? What if I’m offline in a way, will I still do a transaction? And then I means, do I have to, like an ATM offline, who’ll give you a hundred who in a buck, but I won’t give you the maximum?

So we weren’t totally solving those problems, we were more identifying the issues that were raised knowing that down the road we solve. So offline is a big one. Then if you look at, I think that’s the most likely case. The next one I think is either B2C, C2B, and by that I mean business and government. So I walk both online commerce, I walk into a store where I go to pay my taxes. All those are really good cases for central bank digital currency. And then you think about, well, what’s the form factor? Is it only in a phone or are there cards or different ways of thinking about it? So once again, we started on these paths, but with just two years in the budget we had, we didn’t go too far.

The one use case that I don’t think would be useful is probably business-to-business. You mentioned ISO 20022, you need the remittance information, and if we want to go hundreds of thousands of transactions per second, that data is going to bog us down. And there’s a lot of attempts to separate the data from remittance over the years and it’s a thorny issue when you separate them. So we were not thinking about B2B as much because there are some good platforms out for there. And guess one more point is I believe a digital currency should be a public good, so it should be free and really fast and it should try to solve some social issues.

So for instance, I think it would be a goal to try to solve for financial inclusion with the CBDC. And so now you’re thinking about, okay, think of a single mother of two who works in a hotel and gets a paycheck. She’s got to go check cashier, she’s got to get a money order, she’s got to possibly go in-person to pay a bill at the telephone company, whatever. All that can be done with the CBDC and if it’s built right without fees. So I think that’s when you get into the interesting stuff is what are the public policy, what are the goals of a CBDC? And if you create the right goals, you have the opportunity to solve some problems and use innovation to improve the overall market. So I think the use cases are pretty well-laid out that way, but I think it’s only the beginning to think about what innovation may happen along the way after it is launched by the private sector.

Yvette Bohanan:

Yeah. And I loved when we were getting ready for this episode, you said you didn’t want to pave the cow paths. It’s an analogy that is a very pragmatic take on a very thorny set of topics. What prompted you to draw that analogy when you were doing this?

Jim Cunha:

Well, first, I’m from Boston, born and bred, you can tell because I have no accent.

Yvette Bohanan:

That’s right.

Jim Cunha:

Anyways, so there’s a saying in Boston, don’t pave the cow path, so you get Boston roads. So to me, the cow paths are how we build things today. Like I mentioned, compliance. Why is it that banks have to do the compliance? With new technology and new partnerships, isn’t there a better way of doing compliance? So a cow path is how it’s done today. A cow path is how financial inclusion issues create those problems, how cross-border works.

I think you would have to think differently about how the money may move to a person to be able to think about what’s possible. That’s why we created Hamilton, my whole team came from outside the Fed. They were all crypto people because I just wanted to think differently. If I’m creating money digitally, I want to have a whole new fresh look at what it could look like. And that’s one thing that’s interesting in Project Hamilton, so I’ll go back to, is the system looks a lot more like Bitcoin than it does a traditional payment system as far as how it operates. And so that allowed us to also do things differently and get the speed we get because we weren’t sort of tied by the same types of controls and the same types of ways of thinking.

Yvette Bohanan:

I have to ask because I’ve listened to some folks when they were discussing Project Cedar and that and hearing lectures and reading and some of this stuff is like, it looks a lot like Bitcoin more so than anything else. But because of the pseudonymity and because of some of the characteristics of true Bitcoin, you have to sort of throw a lot of that out when you’re talking about a government-issued form of fiat or massive wholesale payment transactions. People want to know who’s validating and ideally the government’s the validator, right? So in for a CBDC.

At what point is it something totally different? Because we have two basis of comparison. We have account-based technology and sort of the cow paths of how we do everything today. And there are many very well-ingrained cow paths in payments because it’s hard to move this stuff. And you have all this new sort of greenfield environment out there that’s evolving all the time, even zero-knowledge proofs and all of that. It’s going so fast and there’s so much new stuff all the time around that. Do you think CBDCs end up being something in its own sort of third category because you’re kind of cherry-picking the technology to create what you really need out of this, right?

Jim Cunha:

Yeah, absolutely. First of all, not all cow paths are bad. Some of them are there for a good reason. And so we have to understand is what should not be tinkered with that works today or creates unintended consequences and what can be tinkered with. And it’s not easy. That’s why we did experimentation was to try, and it’s just from a technical perspective, once you get into the policy side, it gets much harder. But I do think you need to, one, sort of throw it assumptions out and experiment and try different things and say, okay, that didn’t work. What’s my problem? What’s my goal? We were trying to get to 10 million transactions per second and just didn’t have the time to do that. And someone said, why? I said, because every time you break something, you learn something. So we might have learned something about a bottleneck that wasn’t there.

So the goal really is to be smart and not throw out the things that are really important that you need, but extinguish assumptions about how things work. And don’t start off by eliminating all these different possibilities because in your mind they don’t make sense. It’s a hard thing to get to, which is also great why we work with MIT, these were Bitcoin development. These are hard forward crypto people and we always didn’t always think alike, but that’s great. We really needed diversity to be able to challenge each other. And they had Narula who runs the MIT digital currency initiative was really good about questioning assumptions and asking really good questions and vice versa because they don’t know banking.

Yvette Bohanan:

Yeah, they don’t know banking. They’re kind of in this sort of experimental mindset anyway, first principles and question everything. And so you showed up and you said, “Hey, I have this project funded, Project Hamilton, do you want to come and join?” How did you pull this team together exactly?

Jim Cunha:

Well, we actually, so they were in Boston and so we were chatting for the last couple of years because they just wanted to run various things by us and I was curious about it. They had an informal network of technology people and central bankers that they would just talk about their projects with to try to understand it better. And one of my architects teaches at MIT. So we had a lot of relationships there.

And then when the team was formed, it was really the team’s idea to bring MIT together on this as the best minds in the world. They were the leading academic institutions and they were hands-on, they were developing. So once we started looking at it just was unnatural to try to do it. And then because we did it, we also were able to create open-source license for our code and publish the license openly because that’s what they do. Everything they do in that lab is open-source. So it really brought two worlds together that worked well. In my career, a lot of what I’ve done has been the middleman between say economists and payment people or technologists and business people. And so I’ve tried to talk in two different languages and try to bridge that understanding. So it was really wonderful working with this team because of the disparity of knowledge. But in total, it was just amazing.

Yvette Bohanan:

And was there anything that they threw out and you were like, no, it’s a little too far afield, we’re not going to go there?

Jim Cunha:

Yeah, but they were really good about throwing it out. I mean, these are people whose world is very, very different than ours. And was there some things that we said, no, that’s a non-starter, in some cases, I can’t imagine. I can’t think of an example of it. In some cases I’m sure there were things I said “no”, and then I go, “Oh, okay, let’s not dismiss it that that quickly.”

I’ll tell you one example of something that, not exactly on that, but one of the best learnings I had was I’ve been payment systems forever. I’ve made an assumption that we had to process the transactions in the order they came in. Makes sense, right? That kind of makes sense is that we do with payments. And we’re building a platform. And I literally woke up and talking to the team said, “I don’t know why we’re doing that. I don’t think we have to process them in the order.” And what that did is allow us to run multiple copies of our system. That’s how you went from 170,000 to 1.2 million. You’d spin up new systems because the order doesn’t matter, you have to correlate the data. But that one assumption was just profound as far as assuming something or what you asked the question is they have to say it’s too far afield or just impossible. And very few things they say that about, because I knew that I would be quelling the innovation that this team had in a way that was not constructive.

Yvette Bohanan:

There’s a level of humility and a level of brilliance in what you just said that’s quite remarkable and hard to achieve. So that’s very cool. One of the things, you’re mentioning a whole bunch of really interesting stuff, the fact that this was coded in C++, the fact that it’s available out there, I’m assuming still right now and is open-source.

Jim Cunha:

Yeah, it’s called OpenCBDC. Yeah, you can get it on MIT’s website of the figure.

Yvette Bohanan:

Yeah. One of the other things we were talking about was this notion of UTXO, and you said this is a lot more like Bitcoin. There’s sort of this notion of tokens versus accounts in payments. And tokens, a lot of times when you read about CBDCs and things, they’re saying the value moves with the token. It’s a unit of measure and so forth. And then you have UTXO. Can you kind of break down what UTXO is for everybody? And then how you sort of landed on that choice in some of the experiments and what you learned from that that might be different than some of the stuff if you read out there today in the literature about this space.

Jim Cunha:

Yeah. So for one thing, I mean when you get deep into anything, you realize that terms really matter and get slippery the more you get deeper into them, the people have assumptions. So when you say token versus account, people have ideas about what you mean, like what is a token, what is an account? And first of all, they mean different things to different people. They have different even meanings for different disciplines like a legal person versus a technologist. And so the simple way of thinking about it is in your bank there is a computer system with a record in it. It says Jim Cunha and it has a balance of $100, and if you give me 10, they just update the balance. So that’s an account. That’s a simple way of thinking about an account.

I like to think of properties or characteristics versus lump them into a token versus a company. People think, oh, token, it’s anonymous. Or token, the token moves and the money’s actually moving, like it’s actually physically going someplace else, like a dollar bill moves. And that doesn’t have to be true. Our system was based on UTXO on spend token outputs, but it was a ledger. The money is sitting on a ledger at the Fed. It just happened to be in a particular format that’s called the UTXO. And that’s what Bitcoin uses. Oversimplifying it, there’s a record in our ledger that may say this public key owns $77, and then there’s another record someplace else, it says this public key, which is the same one, it’s mine, has $80. So really it’s just data that represents different hunks of money I’ll say. And so if I want to spend $50, the system will find a token or multiple that add up to that and change who owns them.

So I may have 100 records in there that have different pieces of money. So that’s in a nutshell what UTXO is. And what the system does is it takes my $77 one and destroys it and creates a new record with $27 with my number on it and 50 with your number on it. And so what that does, and I didn’t understand this going into it, but my technologist did, is you can do that very fast. You can update those records faster than you can update accounts. And one of the problem with accounts is a quintessential computer problem where if I want to update a record, I have to lock it so nobody else can touch it while I try to update it. And all those locks on money create bottlenecks and contention that if someone else is trying to give me money, it has to wait. It has to wait because my record is locked with my balance. And that actually is not intuitive, but UTXO could be done much faster.

And because the team that we created and the team at MIT were working in that world with crypto and stablecoins, et cetera, that was one of the constructs they gravitated towards. So that’s sort of how we got to that particular model. But as you mentioned, we threw out a lot of Bitcoin and just kept what we thought was the good step that we needed to be able to create a system that the Federal Reserve would run. So we wouldn’t need proof of work and all the energy consumption, et cetera.

Yvette Bohanan:

Yeah, that makes a lot of sense. And then you let go of those things that you don’t need and it sort of exposes new opportunities every time you do that, like you’re saying. And then you learn from that and you keep going. So you must have made many, many passes in each phase of experimentation, how to land on something that you would sort of say, this is our learning here, or this is the conclusion we can draw now.

Jim Cunha:

And a lot more happened than I was in the room for because my team were, it’s during COVID, so a lot of this was online. But because my developer James Lovejoy actually worked at MIT and graduated undergrad and grad from there, he already had relationships with these people. So they worked well together because they had been for years.

And sometimes it would be that someone at MIT would come up with an idea and they would jawbone about whether it made sense or not. Maybe he would run off and he would try to start to code it and see if it made sense. And so some of it was just, will this work, and it was obviously paper debates and paper talks about it. And then we had to hone in on something to focus on. And we had been fully building out two really three systems as part of the project and the few others we did some work in and then didn’t go further with. So it was really the teams, and like I said, Neha was the leader of the DCI trying to get these minds to congregate around a solution or multiple because we didn’t want this one because this experimentation. So some things just died early, some things went a little way and then died on the way and still could potentially be viable further down the road.

Yvette Bohanan:

And you alluded earlier, you had to sort of think about how you could enable something or disable something based on how policy and regulations sort of plays out and give them some guidance, I suppose, in the technological feasibility of some sort of regulatory idea that they’re thinking about.

When you think about privacy, I just want to come back to that for a moment because how we think about privacy today as individuals is our accounts at work and at our bank and how we manage money at home and maybe the relationship we have at work with our employer. And one of the examples that I was thinking through was an income payment in the income domain. Today, my paycheck is private, it gets deposited in my bank account, but only my employer really knows how much money I make, and my banker. It may be third party that has access to my account data or something like that out there. But now you have open, you have a blockchain, you have, it’s auditable essentially in its purest form, everybody can see the transactions. And you were talking about this zero based proof and how did you kind of land on that for, what does that do for you in terms of showing what’s going on and preserving anonymity? Can we go into that a little bit? Just for-

Jim Cunha:

Yeah. And once again, this is where there are a lot of assumptions made. We did not have a blockchain. We did not use blockchain. We use some concepts of blockchain, but this was not like Bitcoin blockchain or Ethereum blockchain or anything else. So it wasn’t a true blockchain, but it was data using cryptography that was in blocks, and there was Cedar. So there’s some of the attributes were blockchain we used, but you can have blockchains that nobody can see. A blockchain doesn’t mean it’s Bitcoin and the whole world can see it. So if you wanted to use the Bitcoin blockchain as a base, we would do it in a way that no one could see it. The Fed would not allow the whole world to look at it. And you can do that just by how you protect the information. So that’s one point. And you have to think about privacy of what from whom.

Right now I’m able to figure out your pay because I know it went through ACH, it’s a PPD, which is an ACH transaction, maybe your name’s on it. So now the Fed doesn’t do that. We have all that data because we won half of the ACH transactions in the US through our system. So can the Fed see it? Are you worried about that? So that’s all the questions. Privacy of what from whom, and we’re trying to do is maximize privacy. So in our transactions, there’s basically almost no data. We had hashes of information. Like I said, my guys know this a lot better than I do, and we would just keep certain pieces of data in hash form and throw it away as we didn’t need it to maximize the privacy and minimize the amount of data we had to store itself. And so it also depends on where is the transaction going through? Is it going through your bank before it comes to the Fed? So does your bank see it and then they can strip off stuff so the Fed doesn’t? So this is complicated stuff.

And how we thought about zero-knowledge proof, so zero-knowledge proof it’s a type of cryptography and processing where you don’t get to see the data, but you can test it. And always simplifying this dramatically, but you could say is this field, which is the name, is this field on the OFAC list? And the person that’s asking that question doesn’t get to see the data, all they get back is a yes or no. So what we could do, and this is not policy at all, we could build a utility that does all of the OFAC checking as part of our application and only kick it out back to the bank or whomever if there’s an OFAC hit, and we don’t see it, they don’t see it unless it’s a hit and then they can examine it.

So there are ways of protecting the data, like I said, way oversimplifying it, and asking, querying it without actually ever seeing the data. In one of the gentlemen at MIT was actually one of the originators of zero-knowledge proofs with Zcash. And so he was actually starting to go down that path. The problem with that is it doesn’t scale well, so you can’t go very fast in doing that type of cryptography and that kind of checking. So he was starting to look into can we speed it up and could it actually be useful in a CBDC context to check for the bad guy stuff without exposing the data? And that’s what the new technology potentially informing different policy choices comes to a head when you start to think about applications like that.

Yvette Bohanan:

And that’s sort of a great example too of if you solved for speed, it’s like telling a startup, just get profitable. It’s so easy to say 12 years later maybe, but if you can solve for the speed, then that’s a great example with the screening, that would just completely change the way we do things now. You would not be following the old cow path in that case, and you’d actually probably be a lot more accurate and informed at all levels if you could-

Jim Cunha:

Yeah. And this is just technology conversations, but you immediately get into, okay, how about legal? Will the lawyers at a large bank allow a utility that’s maybe a public private partnership to check for OFAC when they’re ultimately responsible? That’s a big problem. That’s a big potential legal issue. So it’s not all technology working, it’s also all of the business processes and the legal then regulatory that are, I’d say the more challenging part of it. But if you can solve the technology project, at least make it an option then down the other paths.

Yvette Bohanan:

Exactly. Exactly. So we’ve spent a lot of time talking about Hamilton, Cedar a little bit, RLN. Before we finish up here, and I hate to finish up because there’s a whole lot of stuff I want to ask you, but we’ve only got so much time. Tell me a little bit about your perspective on what’s going on in the rest of the world. I hope this podcast is sort of shedding a little more light for our listeners on how advanced and how deeply the United States has been considering this, but you also have this global view of what’s going on. What’s going on around the world that you think is most interesting right now outside of the US?

Jim Cunha:

Well, just one, how many institutions are working on this and how fast things are going. Like I said, from 2017 to now is not a real long time. There’re, if I get them right, I think 130 central banks are experimenting with either wholesaler or retail or both. And some of these experiments are really advanced. You’ve got Christine Lagarde, the ECB talking about potentially launching in X number of years. I think October is their timeframe for identifying what their actually plans are for digital euro. You’ve got China with their digital yuan. There’s just so much work going on in both wholesale and retail. It just blows my mind that most of this has really been initiated in the last five years. And even with all that, there are very few large at all, large scale real live things. So people are still being appropriately cautious about when do you launch, but they’re trying to experiment in ways and do real deep research both in legal and technology side.

I find it fascinating just the amount of work that’s going on and then some of the more detailed projects like, how do you solve for offline? Japan’s looking at that. How do you think about accessibility? Canada’s looking at that deeply. And how do you bring it all together, because the potential is, if you think about the potential is we’re starting from scratch, so it wouldn’t be great if these things could interconnect both in wholesale and retail and solve some real big problems because we’re starting from zero. So that’s the real challenge now is how do you keep going fast, experimenting, breaking things and doing it in a way that there’s still some coordination so hopefully we get some consistent standard or interoperable ways of doing this. So that’s hard. But I think so far there’s been a lot of cooperation.

So we started in ’15 and immediately started meeting with these other central banks, just a bunch of technology folks getting together and sharing experiments. And that’s continued for six, seven years, COVID really. So that excites me the most, just the amount of what’s going on. And with the private sector, innovation, stablecoins are fascinating. We didn’t really talk a lot about stablecoin, but they’re fascinating in that they’re trying to solve the problem of volatility, but still keep things more in the crypto realm as far as how they work and who operates them. And as you’re seeing with FTX and a few other things, there are some major issues still that have cropped up, including if you have a stablecoin, I mean Circle is a great organization, but Circle has the second largest stablecoin in the world, and they had some money that was tied up in Silvergate or Silicon Valley Bank, I forget which one it was. And because of that, people were worried about their deposits and their dollar coin dropped down to I think 88 cents and then quickly recovered.

This is risky, complicated stuff. And they’ve always tried, I think, to be designed in a way that would eventually be regulated as an organization. And even they had an issue that who would’ve thought those banks failing as they did at the time that were front to stablecoin? It’s very complicated. When it’s really hard and really complicated, then that’s where the fun is with a chance of actually solving something.

Yvette Bohanan:

Always. Always.

Jim Cunha:

I had a great quote, I had this great quote and I forget where it’s from. I heard an NPR, which I’ll say, we don’t do things because they’re easy. We do things because we thought they were.

Yvette Bohanan:

That’s great. That’s great.

Jim Cunha:

It’s a startup talking on that point.

Yvette Bohanan:

That’s exactly.

Jim Cunha:

It’s just fascinating how this stuff works and how it’s all starting to really gel and get momentum, but we’re still a long way away in my mind.

Yvette Bohanan:

What are the guideposts that would say we’re getting pretty close here to actually having something that would be, for lack of a better word, launchable or production ready, we’re out of alpha-

Jim Cunha:

Yeah, but once again, it’s not the technology, it’s the policies makers making the decision to do it. It’s really understanding all the risks. Look at the US. If we launch a CBDC retail, it’d be the most attacked platform in a financial services world, and you’d have to deal with really, really challenging issues.

So I forget who it was that said, if we decide to do it, I think it’s either ECB or England, we decide to do it, we’re a decade away. Because it’s just really, I’m from Boston, it’s wicked hard. But if you decide you do it and you really want to invest in the technology and invest in the debate, then that’s the only way of getting there. But there are forces that are on both sides of this here in the states that simply don’t agree on whether we should do it or not. Just look at some of the bills in Congress, in March of 2020, there was two that would’ve forced the Fed to do this and they would’ve given us one year, which is part of the justification for my work was to just stop learning. But the tipping point I think really has to be that there is a concerted champion and the right political will and the right decision that this should be done worth spending real money on and make the right policy choices so that the work can get done in earnest.

Yvette Bohanan:

Exactly. Out of everything you said in the last hour, the only thing that made my heart go up to my throat was that you would’ve had a year to do this, I think I would’ve been, oh my goodness. Sometimes it’s good that things don’t quite go through, but yeah, that’s really a great point. I mean to see sort of the tea leaves here of what’s going on, it is in the regulatory space, how that’s shaping up, what’s being proposed, how things are changing on that front. Because every paper you read, the conclusion often starts with the technology is capable of doing this, right? But there are all of these other extringencies that you have to deal with.

Jim Cunha:

Yeah. And they’re incumbents, and incumbents by their nature want to preserve what they have and how they operate. And that makes, if I was one, I’d be the same way, but it’s just how to make progress without creating too much risk and without disenfranchising some of the important institutions like banks. You don’t want to hurt banks, but it doesn’t mean that you don’t do it. It’s not a zero or one. There’s a way of doing it in a way that tries to create a good product, but also minimize the risks in that particular community. Every new system that convenes is competing with all the other ones. So it’s not like, and if you freeze things the way they are now, you become sears. You have to change and evolve or else someone’s going to eat your lunch. And it’s hard to figure out how to do that in a way that is safe and sound.

Yvette Bohanan:

That’s right. And it’s a tall order. And I think we are all really lucky that you were driving a lot of this research and experimentation and it had the benefit of your career in all of these other systems and your curiosity to guide that team. So hat’s off.

Jim Cunha:

Thank you. I appreciate that. And as we always say, there’s one person on it with a team that I try to get a vision and get out of their way, let them work really hard. And it was a great way to end a career though, I must say. I could not have been happier with going out with Hamilton being one of the last things I’ve done.

Yvette Bohanan:

That’s wonderful. And you still get that seat to sort of watch it continue on, which is very cool.

Jim Cunha:

Oh yeah. Yeah. It’s a fast moving train, believe me.

Yvette Bohanan:

It’s a bullet train. Jim, thank you so much for joining us today. It’s been an absolute pleasure talking with you, getting to know you a little bit and hearing about this sort of from the inside view and tap into your knowledge and your expertise in this really interesting and evolving, rapidly evolving topic. So thank you so much for being with us.

Jim Cunha:

Oh, thanks, Yvette. I appreciate it. And as you can tell, I’m excited about this stuff. It is my passion, and thanks for giving me the opportunity and one little soapbox here to talk about it.

Yvette Bohanan:

You can come back anytime, Jim. Anytime.

Jim Cunha:

Yeah.

Yvette Bohanan:

And to all of you listening, thanks for joining us. And until next time, keep exploring this industry and asking lots of questions. Keep up the good work. Bye for now. If you enjoy Payments on Fire, someone else might too. So please feel free to share this podcast on your favorite social media outlet. Payments on Fire is a production of Glenbrook Partners. Glenbrook is a leading global consulting and education firm to the payments industry. Learn more and connect with us by visiting our website at glenbrook.com. All opinions expressed on our podcast are those of our hosts and guests. While companies featured or mentioned on our show may be clients of Glenbrook, Glenbrook receives no compensation for podcasts. No mention of any company or specific offering should be construed as an endorsement of that company’s products or services.

Recent Payment Views

Payments Post #12: Lessons from Change

Payments Post #12: Lessons from Change

In this month’s Payments Post, we want to draw your attention to several recent fraud incidents that underscore the criticality of effective risk management to your business and the safety and soundness of the payments industry.

read more

Glenbrook Payments Boot CampTM

Register for the next Glenbrook Payments Boot CampTM

An intensive and comprehensive overview of the payments industry.

Train your Team

Customized, private Payments Boot CampsTM workshops tailored to meet your team’s unique needs.

OnDemand Modules

Recorded, one-hour videos covering a broad array of payments concepts.

GlenbrookTM Company Press

Comprehensive books that detail the systems and innovations shaping the payments industry.

Launch, improve & grow your payments business