Dominic Hobson 00:13: Hello, I’m Dominic Hobson, Co-founder of Future of Finance. My guest today is Rosario Ingargiola, who is the founder and CEO of Bosonic, the San Francisco-based decentralised financial market infrastructure company, whose mission is to eliminate counterparty credit and settlement risk in digital asset trading. To that end, Bosonic has developed a cross-custodian net settlement service, which it is looking to develop with a group of custodians via the Cross-Custodian Net Settlement (CCNS) Working Group. Rosario, thanks for joining us.
Rosario Ingargiola 0.46: Thanks for having me, Dominic; I appreciate it.
00.49: What problems does the Cross-Custodian Net Settlement (CCNS) service aim to solve?
Dominic Hobson 00:49: I hope you’ll forgive me being a bit pedantic here but I’d like to understand exactly what is going on in this marketplace. What’s the problem, which you’re trying to solve, with Bosonic in general, but with this CCNS service in particular? Now, if I could share with you what I think is going on in the marketplace, I’d like you to give me a sense whether you think it’s a reasonable description. So what I think is going on is this: You’ve got sophisticated firms trading in the crypto markets – hedge funds, for example. Now they’re looking to profit from staking their cryptocurrencies – their most obvious source of profit is validating transactions and so on. They’re also looking to do a bit of yield farming, i.e., lending their cryptocurrencies, their DeFi tokens, into these DeFi protocols. But they’re also looking to leverage their positions, in order to profit from swings in these in these highly volatile markets. They’re using Stablecoins as kind of on-off ramps. They’re also using Stablecoins as a convenient way to move between the various crypto exchanges and the blockchain protocols, the automated market makers. They’re also using those Stablecoins as a kind of refuge from volatility – it’s a stable store of value. But it’s also another source of profit, of course, because they can then arbitrage the Stablecoins that drift away from their currency peg. Indeed, that is one of the mechanisms by which algorithmic Stablecoins actually are meant to function. Now you, as a sophisticated hedge fund, you’re active on multiple exchanges simultaneously. My understanding is that they’re finding the post-trade, the operational side of this, very cumbersome. They have to be fully pre-funded on every cryptocurrency and DeFi exchange where they’re active in order to settle trades. And they have to settle those trades directly onto the blockchain. And there’s lots of different blockchains where these digital assets are issued and traded as well as different exchanges. And all of that is very slow. It’s very expensive in terms of transactional “gas” fees. So, this explains why those professional trading firms are very interested in buying services that can reduce those operational costs and operational complexities. Is that the service that you’re offering to sell them? Is that a reasonable description of what’s going on and the problem that you’re looking to solve?
Rosario Ingargiola 03:13: Yeah, it’s a good description of what’s going on in the space, for sure. All of those activities you described – staking, you know, involvement of various DeFi protocols – are obviously all happening by all of the different participants that are trying to get yield from the space. But a lot of the activity is trading-related. So if you think about trading-centric entities, like a hedge fund, for example, if they’re going to do anything at scale, they’re going to need to access liquidity on multiple exchanges. And they’re generally going to access liquidity with multiple market makers or OTC desks. And what a lot of people don’t realise is that, obviously, as you pointed out, the exchange side is fully funded, which is problematic from a capital efficiency perspective and other perspectives which we’ll talk about in a moment. But the market making and OTC desk side of things is actually done 100 per cent on unsecured bilateral credit lines. So what that means is, you quite literally call up any one of the big players in the space, which now is basically all of the major TradFi market makers [and other assets]. They’re now generally all making markets in the digital asset space, and you literally establish an unsecured credit line – neither side posts any collateral anywhere. And you’re doing that across a range of those participants in order to really achieve a sort of a nice aggregation of global liquidity that you can then execute trading strategies on at scale. And so the problem that Bosonic is solving is not just the capital efficiency and the operational side of that, but it really is around the counterparty credit and settlement risk side of it. Because if you think about how you just … You have to kind of look at the different counterparties, sort of independently, right? So if you think about exchanges, for example, the issue that you have there from a counterparty perspective is that, as was recently acknowledged by Coinbase, and pointed out by [Gary] Gensler, the chairman of the US Securities and Exchange Commission (SEC)], assets that you put on the exchange are quite literally on the balance sheet of the exchange, which is fine until something goes wrong. And I think what we’ve seen, in the light of Three Arrows Capital, and some of the fall-out from that, you have exchanges that lost significant … nearly US$100 million of client assets. The way those client assets were lost is because these exchanges are, generally speaking, giving out unsecured credit lines to various market participants that they think are a good credit risk. And so when those market participants fail, because of something like a Three Arrows Capital event, and they don’t show up with the amount they owe to settle, you basically end up having losses that, if the exchange doesn’t have the capital to themselves cover that loss, it gets neutralised against client assets that are on the balance sheet of the exchange. And so really, if you think about that, just at any level, if you’re if you’re coming from the TradFi side of things, the notion of putting assets on a retail exchange in the crypto space and having your assets on the balance sheet or, put another way, having your clients’ assets, your investors that are trusting you as a fiduciary, putting those assets on the balance sheet of a retail crypto exchange is a bit of a … It’s not really fit for purpose, if you really think about it. The other issue that you have is on the market maker side – those unsecured bilateral credit lines – you’re quite literally, you know, on behalf of your clients, if you’re a fiduciary, taking those direct credit risks to your downstream counterparties. And then the bigger issue .. I don’t know if it’s a bigger issue – it’s probably the lesser of two evils – is you then have a settlement problem across all of these parties. There is nothing that nets the transactions that you’re doing between the exchanges and the market makers and OTC desks, let alone other buy-side counterparties that you might want to interact with, in a way where you have capital efficiency. So, for example, to make it very concrete, if you put an asset … if you have an asset, if you trade on an exchange or go long on an exchange to buy Bitcoin, you sell that Bitcoin to a market maker, you now have a position that you have to net settle with the market maker at the end of the day, potentially multiple times intraday, depending on your P&L, and you have a position that’s still open on the exchange that you have to somehow neutralise. There’s nothing that is cross-margining that. There’s nothing that is multilaterally netting that. And so on Bosonic you not only have a mechanism – which I’m sure we’ll get into, around eliminating the risk by not having to physically put your assets on the exchange, and eliminate the risk of market makers by not having to actually have a bilateral credit line or bilateral settlement where somebody has to go first on the settlement, meaning you send them dollars and hope they send you Bitcoin in return – [but] those problems all go away through what we refer to as atomic exchange. And then, ultimately, you get all the capital efficiency of having everything multilaterally netted. So when you go long on an exchange, and short on another exchange or short with a market maker, on Bosonic you’re net flat instantly.
08.25: How are users of the CCNS service able to keep digital assets at their chosen custodian?
Dominic Hobson 08:25: I’d like to come back to you about that atomic settlement and netting in more detail in a second. But if we just focus here on the nature of the solution that you’re offering, you’ve been very clear about the sort of what we might call the Three Arrows risk, if you like, that actually you have this direct exposure, nothing is collateralised. Everything’s on the balance sheet. And so it cascades through the system when something goes wrong. Now, as I understand the solution which you’re offering, it’s a network in which you’re going to have this white-listed group of digital asset custodians and cryptocurrency exchanges, and they’re going to be a closed group. And the trades are not going to be settled by the movement of assets from one balance sheet onto another, if you like, including the Stablecoins, between these digital wallets. So actually the assets themselves are not going to move at all. But instead of that you’re going to kind of in some way delegate those digital assets to members of this group of white-listed cryptocurrency exchanges and custodians, and then they’re going to settle, as I understand it, atomically, as you’ve just mentioned, i.e., in real-time, you know, cash versus assets, if you like, on your network, and then at some point later on, there’s going to be a final reckoning, a settling up between the relevant Layer One blockchains, where those assets were originally issued – on a net basis, which I think you also mentioned a minute ago. Now, is that rather crude description of what the Bosonic network is doing accurate?
Rosario Ingargiola 09.57:Yeah, it’s pretty good. It’s pretty close. I mean, the main thing is that we don’t delegate assets to exchanges. So really clients keep their assets in their own account at their own custodian, where we do represent basically a multi-custodial blockchain network, right? So the solution that we’re giving to our custodial partners allows those custodians to re-ledger those assets or tokenise those assets onto that Layer Two multi-custodial blockchain network. Those assets then become freely tradable and payable. So what’s really happening is … No need for delegation; you keep your assets in your own account; those assets are then available to you to transact with anybody else in the network. And the netting and really the settlement is kind of … and even the atomic nature of the settlement, is really happening in two layers. There’s the trader, the trader level, so I buy Bitcoin from you with dollars, that trade is clearing and settling instantly against those provable assets that you are each holding in your own account at the custodian. So you have this atomic exchange of value that represents the clearing and settlement – in other words, the trade is the settlement, because you have a cryptographic proof of ownership of the change of assets at the custodial level. That is the first layer. The second layer of that, which you talked about earlier, with the cross-custodian net settlement capability and that working group of global custodians from Hong Kong to Calgary, that is essentially a secondary layer where the custodians are net settling on behalf of all of the clients on the network, periodically. So you have this real-time, as you trade, settlement that happens at the custodial level in terms of your ownership – protecting your ownership – of those assets, so you don’t have rest of your counterparties. And then those custodians, wherever assets do need to move physically between custodians, they have an atomic settlement capability. So that’s kind of the way I would describe it.
Dominic Hobson 11.53: Just to be clear on this point. So, you as the owner of a digital asset, it remains in your digital wallet, but in order to make it tradable, it is tokenised onto the Bosonic network. And then the custodians, the digital custodians, settle up, net, after the fact. Is that, right?
Rosario Ingargiola 12:14: That’s correct. So … and really, there’s a lot of nuance here, unfortunately. But the reality is these ledgers, the technology that we provide to the custodians, they actually run those ledgers. So they’re actually custodial blockchain ledgers, which are no different than any other ledger, except that they have maintained cryptographic provability of the assets and the transaction lineage, right? So it’s not much different than your assets sitting at your bank, in your checking account, in a regular database, except that this has cryptographic proof of those assets and maintains a cryptographic lineage. But the point being, that it’s not a Bosonic ledger. It’s a custodial blockchain ledger, where they’re merely re-ledgering or tokenising your assets in a way that they’re in a form that can be then utilised across all the services on the network.
13.00 How does the multi-lateral netting process work between the custodians?
Dominic Hobson 13:00: And when they come to these digital custodians, who are kind of operating the network, [when they] come to settle these net amounts, the good news about net settlement is it does cut … presumably cuts the transaction costs when they net settle on behalf of all their underlying clients. And that process, I further understand, relies on smart contracts. Can you – you mentioned a lot of nuance – can you explain in a bit more detail how exactly that process works, especially in terms of allocating those net amounts, whether you’re debit or credit, in terms of your underlying clients? How exactly does that process work?
Rosario Ingargiola 13.37: Yeah, so just to kind of, again, look at the two different layers. So on the intraday trading level, the real-time higher frequency trading side of things, when I buy Bitcoin for you for dollars, with the dollars I have in my custodial account, the Bitcoin you have in your custodial wallet, that transaction is happening in real-time in milliseconds on-chain at the custodian. So I now own the Bitcoin on-chain at the custodian, you now own my dollars on-chain at the custodian. That represents the first layer of, sort of, settlement. And the netting is that it’s affecting your collateral on-chain [with your] custodian in real-time at the time of the trade. So if you go long with one exchange, short with another, you are actually net flat immediately, instantly, in milliseconds, at the custodial level, right? So that’s the first level of kind of atomic swap and the capital efficiency that you gain from that, besides the elimination of the risk to your counterparty and that your counterparty may not settle. Because now you can go to the custodian and say, `Those assets that I own on your chain, I’d like to have those now, right?’ So you don’t have to rely on the counterparty to settle with you. You have a cryptographic proof of ownership at the custodial level. Now, if that is happening across many, many clients at multiple custodians, you then have a need to potentially bring assets back to your home custodian. So imagine you have bought a bunch of digital assets at an away custodian, you now own those digital assets on-chain at the custodian [and] that’s the away custodian, not your home custodian. How do you get those back? And I’ll talk specifically about how it works. But, conceptually, think about it like a certified cheque where you have certified good funds that are guaranteed by the issuer, but haven’t yet cleared your bank account, right? You can deposit it to your bank account, and it will clear and now it’s in your bank account and not in, you know, in your hand as a claim on the assets at the away custodian or away bank. The way that it works mechanically is, basically, we have the cross-custodian net settlement engine [that] does the net settlement calculations across all of the counterparties at the different custodians. [It] comes up with two main quantities. One is the netted quantities. So that’s the portion that does not physically have to move to get assets back to your home custodian. So imagine one custodian owes 1,000 ETH to another custodian and that custodian owes them on behalf of their clients 700 ETH. You can imagine that 700 nets off doesn’t physically have to move between custodian A and custodian B. That quantity can merely be burned and reallocated on the layer to a custodial blockchain ledger – and I’ll come back to the allocation side of it in a moment. But that’s basically a process that is an automated process that will do that burn and allocation on the custodial blockchain ledgers. The portion that needs to physically move, we refer to that as the residual quantities. So those residual quantities in that example, were the 700 ETH net off. 300 ETH physically has to go from custodian B back to custodian A for allocation to the clients that custodian A [has]. The way that we’ve implemented that is with direct Layer One blockchain protocol inter-operability. So we’ve now built the smart contract layer, starting with the Ethereum blockchain for Ethereum-based assets, and also to move Bitcoin on rootstock as a sidechain of Ethereum, where we can programmatically move Bitcoin in the same way that I’m going to describe. But basically those smart contracts are for the custodians to interact with. So they have their own smart contract addresses. They are able to deposit the 300 ETH in that example, into the Layer One smart contract into their own smart contract address. Those assets aren’t … They can pull them back out if they like, until they’re in a locked state. Once all of the custodians have loaded their residuals for different digital assets into the smart contracts, basically what our engine is doing is [being able] to detect that we’re interacting with a smart contract as well from a “read” perspective. We know when all the custodians have loaded the expected residuals. We can then close down one atomic transaction that does both the burn and reallocation of the netted quantities that don’t have to move on the Layer Two custodial blockchain ledger which allocates it to the clients. And we can then also trigger, as the executor of that smart contract, the movement of the assets between the custodians within the smart contract, based on them all having loaded their residuals. And what that does in effect, in plain English, is physically move the digital assets between the custodians themselves at the smart contract level, to cover the residual quantities that have to move, but doing it in an atomic way, where neither custodian has to go first. And on the fiat side of the equation, we have, you know, ultimately, you can use a Stablecoin of course, and we’re working on a programmatic workflow to buy in and sell out the Stablecoin and use Stablecoins as the transport on Layer One for the fiat residual quantities that have to move between custodian A and custodian B. But the other option that we have, available today, which we’ve just finished implementing, is actually an integration with JP Morgan’s JPM [Stable]coin. So as part of that one atomic transaction between the two custodians, what we’re able to do is not only do the transaction that does the burn and the allocation, the physical movement of the residuals within the smart contract, but also call out to the JPM Coin system to pay dollars from custodian A to custodian B, for example, using JPM Coin. So you have basically one big atomic transaction. And the whole point of all of this is that now, if you look at the traditional world, if you look at the FX markets, you have organisations like CLS that basically facilitate what are referred to as payment versus payment transactions (PvP) as opposed to delivery versus payment (DvP). Payment versus payment means there’s no risk that you’re not going to get the assets owed to you from the other party on the other side. What Bosonic has done is basically create the equivalent of that in the digital asset space where all the custodians are now able to settle with each other payment versus payment. And you don’t have the risk of a settlement failure and all of the problems that would come from that where … Because you can imagine how it would work today. If two custodians in the current world without Bosonic want to transact, even if they’re doing it on behalf of their clients, if they’re transacting with liquidity that’s away, at some point they have to net that all down and figure out who’s going to go first and make the payment to the other side. And that really is, you know, very problematic as you start to scale this to the levels that we believe it’s going to scale to.
20.32: Do exchanges, market makers and OTC brokers interact with the CCNS service directly?
Dominic Hobson 20:32: To be clear on those smart contracts. Those are supplied by Bosonic, and they’re used by these digital custodians to kind of load up the fact that, `well, my clients as a whole own 700 at that exchange and they owe 300 [at that exchange].’ And they load that information into the smart contract. And then the smart contract interacts with the smart contract at the cryptocurrency exchange so that the net amount is settled atomically?
Rosario Ingargiola 21:05: Well, you don’t … So what’s interesting about this is that the exchanges actually don’t have to directly interact with that. So on the cross-custodian net settlement … So keep in mind … Maybe we should take a step back and look at how exchanges interact with the network. So all of the parties, whether they’re takers or makers providing liquidity, have collateral on the network [and] a custodian [ with] their own account. So what we do with exchanges, is we have the exchanges hold client omnibus accounts at a custodian on the Bosonic network. So they’re literally holding client omnibus money at a custodian on the Bosonic network. At the trading level, when you transact with them, you have real-time clearing and settlement against those provable assets. So if I buy Bitcoin from an exchange that’s on the Bosonic network, I instantly own that Bitcoin on-chain at the custodian from the omnibus account that they’re holding, right? So they’re really like … As trading exchanges and market makers and buyside participants, they’re all trading entities from our perspective. So they’re benefiting from the trading level, real-time clearing and settlement, [and] no counterparty credit or settlement risk. So what that means is those assets are actually at the custodian. So, on a net basis, the custodians will do movements that cover all of the transactions, no matter whether the counterparty was an exchange, a market maker or a buy-side participant. So the exchanges don’t have to directly interact in cross-custodian net settlement. The settlement is literally just the custodians using this service on behalf of all of the clients that have been transacting.
Dominic Hobson 22:33: Although cryptocurrency exchanges are sometimes custodians as well, are they not? I guess, in your model, they’re just another custodian.
Rosario Ingargiola 22:40: Well, in our model, they’re not because none of our clients want to hold assets on the balance sheet of any exchanges. So if they’re … If an exchange has a segregated custody business, they’re eligible to participate in the Bosonic network as a custodian. But that has got to be the segregated entity, not the exchange entity that’s holding assets on the balance sheet. So if you look at some of the exchanges that do have a custody business, that’s a separate agreement, separate structure, separate legal entity, totally different liabilities, you know, in terms of placing assets there. They are certainly, you know, eligible to participate in the cross-custodian net settlement [service].
23.14: What is the timetable to which the CCND netting and settlement process works?
Dominic Hobson 23:14: And what’s the timescale on which this net settlement process – if I can call it that – takes place?
Rosario Ingargiola 23:20: Yeah, well, on the trading level … And it’s important to understand, there’s also an intra-custodian sort of net settlement process that’s running at the trading level. From a trading perspective, and a collateral perspective, you’re sort of getting real-time clearing and settlement, and you don’t have to think about the physical movements that the custodians do, whether it’s intra-custodian or cross-custodian. It’s all handled for you. So you don’t really think about that or have any direct involvement in that. And the cadence of it is basically whatever the custodians want the cadence to be. And it does have some practical, you know, realities, depending on how their wallet infrastructure is structured. So, for example, if the custodian runs an omnibus [account] for digital assets with our clients on top of their wallet technology, and the movement of assets between two parties inside that custodian is not hitting Layer One, therefore, there is no Layer One cost or slowness. They might have a cadence that’s every 15 minutes to do net settlement movements and we give them all the tools to automate that at the custodial level as well. We provide all of the net settlement, you know, engine and all of its capabilities. If they are hitting Layer One, they would obviously only want to do that maybe end of day or even on a lower frequency cadence. And then on the cross-custodian net settlement side of things, it’s designed to be super-flexible, where any custodian can initiate a cross-custodian net settlement movement with any other custodian, anytime they want. And the other party, if they’re willing to enter into that transaction with them, and cycle through that process with them, can do it literally anytime they want. And we’ve designed it to be flexible that way because obviously there might be needs to do this kind of ad hoc. And certainly, you know, until the group comes together, and decides on sort of how they’d like to see it, you know, operate as a group, you know, we wanted to make sure that we could support that. So it really is up to the custodians to decide the cadence on both levels.
Dominic Hobson 25:28: So I was bringing a wrong-headed, traditional idea to the process, saying, `This is not like batch processing.’ At the end of the day, the custodians decide bilaterally how they want to do the net settlement among themselves.
Rosario Ingargiola 25.42: Yeah, that’s correct. And I’m sure it will coalesce into sort of a preferred model that everybody, you know, is interested in. So that’s part of why we have the working group. It’s really allowing all of those organisations to sort of shape the rules of engagement and the legal framework and so forth, in conjunction with us and our counsel, both from a business side and from a legal and regulatory side.
26:07: Is Bosonic providing what the traditional securities markets would call a “prime brokerage” service?
Dominic Hobson 26:07: Can I inject here perhaps another rather old-fashioned or traditional piece of thinking about this, which is, `Is what’s going on here like or unlike prime brokerage?’ Prime brokerage began as a service in which buy-side firms could trade with all sorts of counterparties who would not normally trade with them because they were not creditworthy, but they could do it through an investment bank, a prime broker. And all those trades would come back to the prime broker and that prime broker would clear and ultimately settle those trades. Is what the Bosonic network is doing comparable with that, or is it completely different from that?
Rosario Ingargiola 26.51: Well, it’s a great question. And the reality is, it’s creating an alternative to that, but I want to explain that a bit, because it’s not at all eliminating prime brokerage. It’s really a way for prime brokers to leverage this and actually augment their own businesses. But you’re right, and I think you have to take a step back and ask, `What is prime brokerage?’ Prime brokerage – and there’s a lot of confusion around this, particularly in digital assets – at its core, the thing you are buying, is actually the credit intermediation, right? You’re literally paying somebody to use their trade name and credit out on the street instead of your name and credit, because it’s obviously easier to get to face other counterparties if you’re using a Tier One bank and their balance sheet. The problem with prime brokerage … And, by the way, all the lending and all those other things, the leverage, those are all sort of ancillary services. They’re important services, but it’s not the core thing you’re buying or thinking about. That is credit intermediation. Now, the problem with that model, or even any central counterparty model, whether it’s an exchange with a [central counterparty clearing house] CCP that’s novating the trades and is the buyer to every seller and the seller to every buyer, all of those approaches are reliant upon a single centralised balance sheet: the balance sheet of a Tier One bank as a prime broker, or as a CCP, the balance sheet of the clearing house and all the member-contributed capital. That’s basically the problem, right? So, in the Bosonic model, what’s interesting about it is that it eliminates the need for that single centralised balance sheet to eliminate the counterparty credit and settlement risks. It’s a pure technology alternative to that. And it’s obviously – you know, this is where it gets into a little bit of additional nuance – easy to understand conceptually what Bosonic does from a fully funded perspective. But what happens when you do bring in credit and leverage? Which is very important in a financial market, right? We’ve basically solved that problem through a particular approach around a repo market. And I’m sure we’ll talk more about that. But there is a way to bring in leverage and credit but do it in a way where it’s not originating solely from the CCP or solely from the exchange or solely from a Tier One bank as a prime broker. And so that raises some really important questions as you think about prime brokerage in digital assets, because we all know that none of the Tier One banks are doing prime brokerage and digital assets. In fact, most of them don’t want to do prime brokerage in plain vanilla assets. And we see them exiting the FX space and things like that kind of left, right and centre. You know, this is the problem with that … The reason for that is because they don’t want to have all the credit allocation. There’s capital charges around the credit allocation. There’s obviously risks. I mean, take a look at Credit Suisse and the debacle that they’re in now, possibly triggered by a US$6 billion plus loss from a single hedge fund client in their prime brokerage unit, right? So that gives you a sense of the scope of balance sheet that you need. And that leads to the big question in digital assets. If you’re going to look to a prime broker in digital assets, your first question as a traditional firm is going to be, `Who’s my counterparty?’ They’re obviously going to be your counterparty from a trading perspective. And then your next question is going to be, `How big is your balance sheet?’ And if you look at the firms that are offering those services, you know, even some of the larger ones, they don’t really have a big enough balance sheet to do that at scale, we would argue. And so the model that Bosonic has created, including this new way of injecting credit and leverage in through a repo market without changing the credit risk profile, or the settlement risk profile, of the downstream counterparties is a really, really important development in market infrastructure.
30:39: How reliant is a technology-based network model on atomic settlement?
Dominic Hobson 30:39: Now, as you pointed out, at the heart of this is credit intermediation. You’ve alluded to the fact that big investment banks are looking to exit the traditional prime brokerage business, if you like, because of the capital costs, among other things. And of course the risks are becoming progressively less attractive, partly because the buy-side firms are actually very alive to the investment bank itself as a liability of theirs. Now, you’ve been very clear that your core mission is to eliminate that counterparty credit, settlement risk, but you are – I hope it’s not unfair description – a technology firm. You’re not a bank with a big or indeed small balance sheet. How reliant is your model – and there are other factors at play here, which we’ll come to in a moment – how reliant is your model on the atomic settlement? By which I mean you to understand that any transaction where the other side doesn’t turn up with what it’s meant to deliver, the transaction simply dies. In other words, the counterparty credit risk is eliminated by the fact that it’s payment versus payment or delivery versus payment in real-time all the time. How reliant is your model on that?
Rosario Ingargiola 31.52:Yes, so, well .. The core thing that we’re doing to eliminate risks is making the transactions atomic both at the trading level and then, as we described it, even at the custodian net settlement level, right? So the automaticity of moving [assets]. And all that means, just in plain English, is a concurrent transaction that’s a total success or a total fail. So, in other words, nobody ever has to go first. Nobody has the risk of, `I went first and I didn’t get what I was receiving’ because a settlement fail upstream somewhere, we’re down or whatever. So that is really important. It is a very, very important part of what we’re doing. The key thing though, is that Bosonic is not a counterparty to the trades, right? So when you sign a Bosonic agreement, you’re signing a technology licence agreement, whether you’re a custodian or whether you’re a trading entity. We’re licensing those counterparties different technology components that communicate with each other. So we’re really operating the network in terms of a messaging layer. And we’re running the network as a business. But it’s really custodian licensed and operated ledgers connecting into that network. And it’s trading entities that are holding accounts at those custodians using various bits of trading kit that we deliver, or even their own trading kit if they like to participate and face other counterparties on the network. So that atomic swap concept at the custodial level is how we eliminate all the counterparty credit settlement risk. So we’re never a counterparty, and therefore we don’t need any balance sheet. There is no novation of trades going on here. This is literally, you know, happening on-chain at the custodian in real-time, independent of Bosonic and its balance sheet or it being, you know, a participant in any way in the trades.
33.33 Is Bosonic able to offer its users finance or access to finance?
Dominic Hobson 33:33: Now you mentioned credit by which I don’t mean credit intermediation. I mean actually credit. I understand Bosonic is planning to develop a repo market, presumably to ensure that participants who don’t actually have what they need to deliver into a real time atomic settlement can actually borrow what they need, presumably against collateral. Now is that collateral also tokenised onto the digital custodian network that you operate in the same way as everything else we’ve talked about?
Rosario Ingargiola 34:09: Yeah, that’s exactly right. And it needs to be …I’ll explain a little bit about how it works. So we’ve already built this – this whole repo market capability. We’re in the middle of putting the finishing touches around the auto-liquidation so that we can maintain basically all of the margin levels that people require. But we’re using really a lot of DeFi concepts here. So you have lenders that … Maybe it’s helpful to describe the current process. So today if you want to lend Bitcoin institutionally – you’re sitting on a Bitcoin position, you want to lend it out get a yield – you basically generally will go to some sort of an intermediary; you sign an agreement with them; you send them your coin; they then go find somebody to borrow it; they sign an agreement with the borrower; the borrower wires them the dollar collateral; they send them the coin. It’s a very disjointed process that’s heavily intermediated, and very clunky and not fit for purpose for a real-time transaction, obviously. And, obviously, you’re turning over your assets, both the collateral and the coin. In the Bosonic model of the repo market, we’re just leveraging all the same concepts. Lenders keep their assets in their own accounts at their own custodians; [and] they control the work they do to get those assets tokenised if they wish them to be utilised in this way. They can then post their interest into any of the repo books, right? So there’s different repo books that will allow … Where there’s competition around the yield or the price – the interest rate. But there are different initial margin terms on those different pools of liquidity or books. So imagine a 25 per cent initial margin pool, which obviously carries greater risk. You have the option to post your interest there. You might price it higher, because you have greater risk with only 25 per cent initial margin. Then maybe you have an over-collateralised pool, maybe even a 125 per cent collateral pool. Obviously, you’re going to price that lower, [because] you have a lot less risk around, you know, loss of those assets as a lender. Now, the key thing is, and it will become obvious why this is the right way to inject credit and leverage into an eco-system like this, because those people that are lending that have those assets on-chain at the custodian now, from a lender perspective, if you as a borrower have collateral on-chain at the custodian, and you do a repo transaction, that repo transaction changes the ownership of the underlying assets on-chain in real-time. So, to give you a concrete example, I’m in the 50 per cent initial, I’m looking at the 50 per cent initial margin, one day repo pool for Bitcoin. I can post US$100,000 of collateral, borrow US$200,000 of Bitcoin. That transaction is between me and a particular lender or a group of lenders. Let’s say it’s one lender. I now own that Bitcoin on-chain at the custodian that I borrowed on the repo. They now own my dollar collateral on-chain at the custodian. So it’s just another real-time, atomic swap of assets. Now, the benefit of that is that Bitcoin that I borrowed on the repo, when I sell that to another counterparty on the network, I’m actually fully funded against that counterparty even though I’m levered against my lender, right? So from a trading eco-system perspective, none of the trading counterparties need to care how levered any of the other counterparties are in the network, because all of the trading is actually still fully funded, because everybody is repo-ing in the assets that they don’t have that they’re transacting, right? So that keeps all the risk out on the edges between the borrowers and the lenders. And, basically, what it allows for is crowd-sourcing of almost unlimited balance sheet just against people … The only requirement is that you’re a client with an account at a custodian who wants to lend, as opposed to having to be a specialist firm or have special technology or what have you. And so we think that, you know, looking at some of the DeFi lending protocols, if you look at the amount of assets that they’ve been able to attract, even in anonymous pools, it’s quite impressive, right? It’s massive, you know, tens and tens of billions of dollars in some of these protocols. So we think that there is a very big market, and it’s been confirmed by, you know, all of the leading firms that are involved in lending, the ones that are still in business, including the ones that are still in business, that want to lend in this space, they’re very interested in lending in that way to get the yield. And the beauty of this is that you can have, for example, a market maker who might want to price 100 different assets and stream those prices out to different counterparties. They don’t have to hold all that inventory. They can have a relatively small amount of dollar collateral, or maybe a portfolio – ultimately in Phase 2 a portfolio of other assets. That becomes their collateral that’s appropriately haircutted. And they can basically repo the other assets that they’re selling to other counterparties, kind of just-in-time as those counterparties transact. And it gives them the ability to eliminate all that capital cost of holding all that inventory but does it in a way where it doesn’t endanger any of the downstream counterparties. They have no additional credit risk to each other, even though there is, or could be, a very large amount of leverage between the borrowers and the lenders. Hope that makes sense. I know it’s complicated, but it’s the right way to do this instead of a loan market, which is how some other markets are structured in terms of how they finance leverage. It’s obviously a much, much safer way for the broader eco-system to basically bring leverage into the equation.
Dominic Hobson 39:39: I read what you were saying as. `This is a proper repo market. The ownership, the title to these assets, actually moves. It’s not like anyone’s pledging this stuff and, well, if I fail to return your stuff, you might get this [collateral] and be able to sell that to recoup your loss.’ This is the real thing.
Rosario Ingargiola 39.55: That’s correct. And it has all sorts of implications. Because if you look at you know, the whole lot of the Three Arrows fall-out, a big part of it was just people thought they had pledged collateral [and] people thought they had perfected an interest in collateral that they hadn’t. There was re-hypothecation going on. There’s all these things going on. And if you relegate all that to custodial blockchain ledgers with cryptographic proofs, and everything’s programmatic, and programmatically enforced, you have a much, much safer world of repo transactions.
40:24: Can Bosonic provide the services that it does without becoming a regulated financial institution?
Dominic Hobson 40:24: You don’t re-visit Lehman Brothers International circa September 2008, and say, `Where’s my stuff?’ Now, one thing that occurs to me, if you are operating this repo service as well, the Bosonic network is becoming a very interesting entity. You’re kind of orchestrating trading by these firms between these digital custodians on multiple exchanges. You’re orchestrating the exchanges of assets between these custodians on whatever timescale they will actually agree between them. And you’re also now doing a kind of securities financing product, a kind of collateral management and financing, [a] generation of credit – again, working with the various custodians. So do you think Bosonic is kind of evolving away from being a technology vendor and is becoming almost a market infrastructure? I used the term market infrastructure when I was introducing you, but I was I premature? Or is that how you see yourselves now?,
Rosario Ingargiola41:24: No, I think … Look, I mean … I think we’re … I think there’s some open questions around some of this. We would like to, obviously, be a technology vendor and not be a regulated entity. But as a practical matter, there are things that are very, very valuable to the eco-system that require some kind of regulatory cover, right? And so, if you think about, for example, the repo market, those are basically securities transactions, so we’re not live with that today. But when we are live with that those that would require, or may require, in some jurisdictions, a broker-dealer licence or the equivalent. Bosonic actually has a whole regulatory roadmap. We’re actually very close to getting our broker-dealer and Automated Trading System (ATS) licence here in the United States. Part of that is so that we can offer services like that, you know, without having to necessarily partner with somebody that has the regulatory cover. But the truth is a lot of what we’re doing is with other [financial market infrastructures] FMIs, other large regulated organisations, and many of these are enterprise deals, where they might take our technology and essentially operate it under their own umbrella, where they do have the regulatory cover to deliver those kinds of services to the clients. So it’s a blend of things. There is a whole roadmap, you know, here. We do think that … We’ve done some analysis around sort of, you know, `Is this a clearing house?’ And, obviously, not having to require any balance sheet in play, you know, makes it very different from the typical clearing house. And we’ve looked at some of the clearing house agency exemptions that have been granted in the United States, for example, which is a very heavily regulated jurisdiction for that. And there’s clearly ways to get, you know, excluded from that necessity of registering as a clearing house agency. And so we’re looking at possibly pursuing that, you know, here in the United States. But the truth is, right now, today, there’s still a lot of grey area and a lot of lack of clarity on sort of what the requirements really will be. But we’re definitely thinking about that. And, obviously, you know, we cater to larger regulated institutions and fiduciaries, you know, many of those, you know, may ultimately become bigger partners in the eco-system, where perhaps part of what we’re running is actually a service offered in partnership with our technology and their regulatory cover. So that’s a fair point. But, obviously, we’re … The main thing is we’re not a counterparty, and we’re not using our own balance sheet. We’re really just providing this tech. So, to the extent that we technically fall into some regulatory, or regulated activity, we do have to find a solution to that, whether it’s directly being regulated or using a partner.
44:02: How do users of the CCNS network achieve capital savings?
Dominic Hobson 44:02: Okay, I’ll come back to your regulatory licences in a second. I’d also like to talk a bit about the members of the broader eco-system, your CCNS network, if you like. But just to clarify one thing, which is important in why people would start to use the Bosonic network. You mentioned once or twice that it leads to economies in capital. Now do those capital savings come purely from the elimination of the need to maintain liquidity, if you like, at multiple exchanges because you need to pre-fund in order to settle atomically? Or are there other sources of capital savings apart from that?
Rosario Ingargiola 44.45: Yeah. Well, there’s a number of things. I mean, obviously, funding different multiple exchanges is problematic. You can have one much smaller balance at a custodian. And if those exchanges are participating in the Bosonic network, you can still face them. So you do drastically reduce the amount of collateral that you’d have to have in order to interact with those exchanges. So that’s certainly one element. The other is really that real-time multilateral netting that’s happening, right? Because if you go long with an exchange and short with a market maker, and let’s say you’re happy to keep the position on the exchange, you’re going to look for an offset with your trading strategy, or whatever business you’re running. You’re going to have to settle that asset over on the market maker side, right? So you’ve now sold them maybe a digital asset, you have to deliver that at the end of the day. You would, if you don’t have that [asset] on hand, have to borrow that in. You have the cost of borrowing that in to deliver that asset to them. You are then still carrying that market risk on the underlying asset, in addition to the rate on the asset – or the rate on the borrow. So there’s a lot of capital efficiencies to be gained by having just this multilateral netting in real-time against collateral at custodians. And that’s obviously a huge component of this. And anybody that is, you know, wanting to get involved on the trading side of things … If you think about just not having counterparty [risk]… If you think about a regulated entity like, let’s say, an investment bank, or somebody like that, that wants to run a trading desk, or offer trading out to their clients – direct market access or otherwise – to be able to access the digital asset market, they might have all sorts of … If they’re taking balance sheet risks, they might have all sorts of additional collateral requirements or reserve requirements around those potential risks, right? So, to the extent that you can minimise those risks, and not have balance sheet risk, and not be doing this off your balance sheet, you may be able to also drastically reduce the amount of reserve capital and so forth, depending on the type of organisation that you are.
46.56: Which types of institutions belong to the CCNS network?
Dominic Hobson 46.56: Can I make sure I understand better this broader eco-system you’re talking about? We’ve touched upon this Cross-Custodian Net Settlement (CCNS) group you’re setting up. We’ve also talked about multiple exchanges. So is that broader eco-system … Does it … It obviously needs customers, obviously. But do you also need exchanges, as well as digital asset custodians, to be part of this, if you like, white-listed, Bosonic eco-system, network, group, whatever we want to call it? So how many different interests do you need represented on this group to really get that the scale and volume that you’re looking to build? What types of entities need to be there?
Rosario Ingargiola 47.41: Yeah, it’s a great question. And yes, it is an eco-system play. And we’ve spent the bulk of 2022 basically building out that eco-system. So I’ll come back to the custodians in a moment. That’s obviously – probably – the most important thing that we’ve done – organise this massive group of global custodians. And I’ll talk a bit more in detail about that. But yes, you do also need the liquidity sources, which includes both the market makers and the exchanges. And we now have a quorum of big-name exchanges and big-name market makers that are providing liquidity over the Bosonic network to clients that are on the Bosonic network. But it’s a huge benefit to them. Because if you look at the direction of travel, you know, from a regulatory perspective, and really, from every perspective, from a risk perspective, you know, in the wake of Three Arrows Capital, clients largely don’t want any assets on exchanges, they don’t want bi-lateral credit risks to the market makers, and they don’t want all these settlement risks. So what’s nice about joining the Bosonic Network is that you can face off against a very large set of clients that you may not be able to reach directly under your current business model, right? So we have clients, and I won’t name names, but fully regulated entities in the United States that are running a whole crypto business on Bosonic’s technology. They don’t have to do anything off their own balance sheet. They’re not taking any risks. They’re not touching digital assets directly. That’s really important. And they’re not able to put client assets on the exchanges. They’re not able to take risk to downstream counterparties and market makers with bi-lateral credit because of their view on, you know, their regulatory matters and they’re acting as a fiduciary for their clients. And so what that means is they’re just not going to be able to do business in the current model without something like Bosonic facilitating all that. And so if you’re coming on as a market maker or exchange, you’re basically able to face ultimately everybody that is in that category across all of the custodians that are on the Bosonic network with one account at a custodian of your choice on the Bosonic network. So just think about the leverage that that gives in terms of liquidity distribution. So that’s why we’re able to successfully build out this eco-system and get people to come on, because it’s really a win-win for everybody on every level. And even you can look at, conversely, just one kind of funny point. If you’re a market maker, and you’re giving out unsecured credit lines, and you’re not holding collateral of your end-client, probably there’s a lot of clients you’re not taking credit risk to, and you’re saying, `No.’ So you can use Bosonic to face all of those clients without any counterparty credit or settlement risk. You do have, though, to post a little collateral. They post collateral in their own accounts at a custodian. You can price them. And now you have, you know, the ability to face that client and provide them liquidity without any risk to that client from a credit perspective. So that’s super-, super-important. And kind of back to the custodial side of things, as you can imagine, the holy grail of the whole thing that we’re building – or have built, I should say – is this cross-custodian net settlement capability and bringing together what is now a group of well over 27, it’s probably even north of 30 now, participants in that cross-custodian net settlement group. And, again, I’m not going to name names, because it is a very special sort of arrangement that we have with all these folks where we are trying to foster openness and so we don’t want to say too much about the discussions or say too much about who the participants are, except the ones that have been happy to be named, which I think you probably have seen those in the press. But there are everything from the biggest global custodians that you can name, and some of the biggest banks that you can name, all the way through to the crypto natives. We do have a few platforms and venues that we think are interesting to have in the mix as part of how this all shapes up. So they have a voice and …. the working group is putting things together. But, ultimately, that very large global network of custodians tied together by a common protocol, common technology, that’s not a theory, not something we’re building that’s three years out, but something that we just demoed for the working group a few weeks ago. And we plan to do our first live, real money transaction, where we settle custodian to custodian using the cross-custodian net settlement functionality as early as mid-November this year. So next month is where we’re targeting. Fingers crossed, everything comes together the way we’d like. So this is a very real solution, with a very elegant and very good solution to the Who Goes First problem; to eliminate all those risks for those custodians; [to] make those settlement movements atomic. And everybody understands that, even if they don’t want to use Bosonic from a trading perspective to access the market, they all understand that at some point, even if they’re trading directly with other participants at away custodians, they’re going to have this problem of settlement custodian-to-custodian. And that’s why we’ve been able to assemble probably the biggest by far group of custodians that are co-operating in any initiative globally in digital assets.
53:04: When you use the term “market maker,” what types of institution are you referring to?
Dominic Hobson 53:04: I’ll come back to that custodian question in just a sec. But just to clarify one narrow point, when you refer to market makers, you’re referring to Uniswap, Sushiswap? Or are you referring to something else?
Rosario Ingargiola 53:13: No, I’m not referring to the DeFi Automated Market Makers (AMMs). So we haven’t really touched on DeFi. We have a whole roadmap for that. But that’s kind of a 2023 thing for us where we’re tying in some of those protocols, some of those decentralised exchanges, but that’s a different kettle of fish. When I talk about market makers, I’m talking about the usual non-bank market makers in all the other asset classes. So all the names you’re used to seeing by now in the news, you know, Citadel, Hudson River Trading, you know, Optiver flow traders – all of these guys plus the crypto native ones like Galaxy Digital, and others that are that are making markets in the space and providing liquidity. That’s more of kind of an OTC flavour of liquidity – that’s the way that the industry thinks about it. We’re definitely talking about TradFi centralised players that are very, very large, proprietary trading shops – Jump Crypto – all these guys that are making markets in the space.
54:12: Which custodians have joined the CCNS network?
Dominic Hobson 54:12: Okay. I’m glad I asked that question. So DeFi is project 2023 for you. You referred to 30 participants in your cross-custodian network now. You did put out a press release when you first set it up with four banks- First Digital in Hong Kong, Tetra Trust in Canada, Vast Bank in the US and of course Trustology, which has now been bought by Bitpanda back in February. So you’re not having difficulty adding to this group – by the sound of it, you’ve got another 25-plus that have actually become members of the group. And, as you said, that includes some of the big-name global custodians as well.
Rosario Ingargiola 54:53: That’s correct. And one of the key partners in the Cross-Custodian Net Settlement Working Group, and part of that solution, is actually J.P. Morgan with JPM Coin. So that is one of the fiat [currency] rail options. So when we do a cross-custodian net settlement movement, one of the ways that we move dollars between custodian A and custodian B for allocation to custodian B’s clients, is by calling JPM Coin systems and using JPM Coin, so those custodians that are J.P. Morgan clients, and want to use JPM Coin as the fiat rail, they can do that as part of this cross-custodian net settlement, so we are very obviously very proud of that partnership with them, and to have them involved in that. And they have a quite excellent product that lets us move dollars without all the normal banking cut-offs and constraints. We have 24/7/365 ability to move fiat currencies, starting with dollars, and ultimately with a number of other key currencies coming in early next year.
55:58: How broad is the range of digital assets being settled via the CCNS network?
Dominic Hobson 55:58: This might be a stupid question. But is the range of coins and tokens that can be settled through your network constrained in any way? Is it the top 20? The top 10? The top 50? Is it purely driven by the users of the network themselves, i.e., what the customers want to trade, what the custodians want to custody and what’s available on the exchanges? You’re kind of agnostic as to what people are actually doing? But if you step back and ask yourself, `Well, I don’t know, there’s 3,000 potential coins out there, tokens out there, which could be settled on our network, but actually 90 per cent of our business is in the top 15 or 20.’ Is that a stupid question?
Rosario Ingargiola 56.42: No, it’s not a stupid question at all. Yes, we’re totally agnostic. In fact, you can actually do … The whole thing that we’ve developed here, really, if you think about it, allows anybody to tokenise any asset at any custodian and do an atomic swap for any other asset of any other custodian. So that could be dollars and treasuries. It could be Bitcoin and treasuries. It could be, you know, dollars and Bitcoin. It doesn’t really matter what the underlying asset is. It’s whatever the custodians wish to tokenise and support from a custodial perspective on the network. And so it is dictated somewhat by, `What do the custodians want to custody? What do the liquidity providers want to want to provide? And then, ultimately, what do the trading entities want to trade?’ From our perspective, it doesn’t matter what we’re settling. So if you think about, for example, traditional securities, if you look at a lot of the enterprise deals that we’re working on, now, and some of them are in Proof of Concept phase, where, basically, we can even do something like tokenise the books and records of a [central securities depository] CSD. Just think about them like the custodian, or like the Layer One blockchain protocol. You’re able to take that tokenised representation, have atomic exchanges of value on the Bosonic network, and ultimately message that downstream for a change to the ultimate book record of the CSD – as an example. So it really doesn’t matter what the underlying asset is. And just to kind of expand your mind beyond crypto, if you think about traditional assets, think about CSDs – whether those are natively digital assets that are being issued in the future, or whether they’re non-native digital assets today that just exist in the current form. All of that can be tokenised and made tradeable using the Bosonic infrastructure.
58:28: Do users of the CCNS network contract with Bosonic or the custodians?
Dominic Hobson 58:28: Before we wrap up, I’d like to ask you some dull but very important questions about contracts, about regulation, regulatory licences, which you’ve touched on, and about the legal position here. So if you are – you’ve explained that Bosonic is never the counterpart – a user of this service, who are you contracting with? Do you contract with the exchanges and the custodians or are you contracting with Bosonic in any way?
Rosario Ingargiola 58:59: So what you’re really … It’s a great question, actually. So what you’re doing is you’re contracting with the custodian, because obviously you have to have a custodial account. We don’t intermediate that. That’s a direct custodial account set-up with the custodian. You’re contracting with us from a licence perspective – licensing some technology. Now, what’s ultimately happening is peer-to-peer transactions that self-clear and settle by virtue of how the atomic swap technology works, right? So that’s why we’re not a clearing house. We’re not an intermediary. We’re not a credit intermediary. We’re not a counterparty. It’s because you really are facing your downstream counterparties. Now, what’s important to understand, though, is that if you’re comfortable with how the model works, you understand why that doesn’t present a risk in terms of facing that counterparty, right? It’s all done using the Bosonic model. You have proof of ownership of everything at all times at the custodial level. You’re going to the custodian to request those assets if you want those assets from the custodian, not your downstream counterparty. And so you do have … The only risk really in the system is actually custodial risk. And there is a … It’s important to note that there’s a very big spectrum of custodians. All of the custodians in our working group and in our global custodial network are regulated entities that are approved in various ways. And I think it’s not hard to understand that, on the back of announcements like BNY Mellon, just yesterday or the day before, whenever that was, coming in and offering digital asset custody, there’s likely going to be a pretty significant shift for at least a certain segment of the traditional institutional market towards those larger banks. There are obviously going to still be specialist custodians that maybe have better access to staking and other reasons why you might have some assets at some of those other custodians, but we think there will be a big shift. So having a custodial risk as your main risk in the system is an acceptable risk if you’re looking at custodians like BNY Mellon, of course. So that’s really how to think about how it’s structured from a legal perspective. And it is a kind of a seemingly dull question, but it is really important. Similarly, the licences that we’re signing, again, with the custodians, to give them that technology to run their … perform their function on the Bosonic network, it really makes that kit, that technology, their technology for the purposes of servicing our mutual client.
Dominic Hobson 1:01:26: Just on that legal position – the term you used a minute ago, [that] your only risk in the system is custodial. In other words, if your assets go missing, for whatever reason, or your counterparty fails, it’s going to be the custodian who is expected to make you whole. And that will be you as the user who contracts with that custodian, which defines the terms of that. Regulation, I guess, might define it as well, particularly in Europe. But is that the position – is that the legal position?
Rosario Ingargiola 1:01:55: Wel, sort of, just. But you have to drill down on that a bit, right? So remember the custodians are really … The custodian role in the Bosonic eco-system is to safeguard client assets; tokenise those assets on the custodial blockchain ledger; burn those assets off the custodial blockchain ledger to remove them when it’s time to pull them off the ledger; and process net settlement movements on behalf of the parties. So as long as the real risks and liabilities lie around custodial … The usual things – custodial errors around what they’re putting on the ledger, maybe; maybe there are fidelity questions around it. Could somebody at the custodial level steal your assets? It’s very similar to the same risks you have with a checking account at any bank, right, in terms of the operational risk on the custodial side and then the risk that somebody inside the bank does something to your assets. So, those things are generally covered by the custodian in various ways. And that’s really the model here. If the custodians are doing their jobs correctly, you’re never going to have a situation where you don’t have an asset to settle, right? So that’s the kind of the whole point of the model. And ultimately, where we’re headed with this is, when there’s a government-sanctioned Stablecoin, for example, for the main currencies, or even for dollars, we could actually have the assets when they’re locked at the custodian before that tokenisation step – they could actually be deposited to a Layer One smart contract with our lot. And they’re provable, independently provable, and locked for use. And that would mitigate even some of the risks I just described and take it to a whole new level. It doesn’t make sense to do that just for the digital assets because it doesn’t make sense to force clients to hold a Stablecoin in the current environment, instead of dollars. But once there is something that’s government-sanctioned or sponsored, it would make a lot of sense to have those Layer One locks even at the custodial level, which gets rid of a lot of those risks that I just talked about.
1:04:07: What is the regulatory status of Bosonic and the CCNS service that Bosonic provides?
Dominic Hobson 1:04:07: Okay. Sorry, I promise this is my penultimate question. But could you share with me a little bit about your own regulatory status? You mentioned a little while back that this repo service you’re developing probably entails getting a broker-dealer licence. You mentioned an ATS licence. Obviously, in the United States, you have a rich variety of regulators you could be regulated by. But what exactly is the regulatory status of the service you’re providing? By which I mean, you know, who is regulating Bosonic and, I suppose, who can use the services that Bosonic supplies?
Rosario Ingargiola 1:04:42: Yeah, well, I’ll try to keep this to be a short answer. It’s a very big question that obviously differs in all the different jurisdictions. But at the end of the day, today what we’re focused on are digital assets that we believe to be commodities, you know, in consultation with our Regulatory Council. And we are doing this all for eligible contract participant business only, so institutional only. And so therefore, it kind of puts it into the Commodity Futures Trading Commission (CFTC) jurisdiction, and it kind of puts it in [inaudible]. And by the way, we’re only doing spot today in this way – spot currencies or spot crypto assets, and fiat currencies. And so it puts it into a very similar sort of regulatory framework as you would have for an anonymous FX electronic communication network (ECN). That’s all institutional and spot only, no derivatives or swaps. So what that basically means is you don’t have a registration requirement. And so we believe today that, even in the United States, amongst those parties doing that limited set of assets, we don’t have a registration requirement. As we move into wanting to do the whole long tail of digital assets, and obviously as we move into other jurisdictions, there’s going to be slight variations. In the United States, what we’ve done is we’ve already many, many, many months ago filed for broker-dealer, and alternative trading system or ATS licence. And what that will do when granted is give us the regulatory cover to be able to offer all of these services in the United States, even all the way down to the retail level, even for things that might be digital asset securities, if we ever get to a point where we actually have a clear definition, and we can actually say with some confidence what those assets are, whether they’re securities or not, we’ll be able to transact them. And also our repo market transactions will fit, you know, well within that framework. And so we’re very, very close to getting those licences granted, we think, with full digital asset permissions, and they will let us reach all the way down to the retail level as well, so we’ll be able to go beyond institutional. And, you know, we’re pretty excited about what that means. And we think that anybody that is transacting anything that might be a digital asset security in the United States is going to have to get that regulatory cover here in the US, right? So we think that there’s probably a long queue of folks trying to get those licences right now.
1:07:06: What is your competitive differentiator?
Dominic Hobson 1:07:06: My last question is this: you were very early into this into this game, if you like. You’ve made a lot of progress. But you are not alone in trying to create this type of network. There are other entities who are developing so called prime brokerage services. We’re seeing crypto exchanges, inter-dealer brokers, some of the market makers, getting into this game as well, even digital asset custodians looking to develop prime services, if you like. What do you feel – apart from your early entry – are your competitive differentiators as you look at other providers that are getting into this marketplace now? What makes you different?
Rosario Ingargiola 1:07:47: Yeah, it’s a great question. And it really comes down to a very big technology moat that we’ve built over the last nearly seven years now. And it really is the fact that it’s all technology-driven, blockchain. There’s a whole bunch of really important factors. If you were to ask me, `What do you have to be to be sort of the category leader in the space?’ I would say that you have to be custodian-agnostic, multi-custodian and cross-custodian capable; it needs to be blockchain-based so you have cryptographic proofs even at the trading level, which is non-trivial to do. You can’t do it on a Layer One. You need to do it on a Layer Two, because of the transaction scalability and throughput problem. It needs to have direct Layer One inter-operability and facilitate settlement movements between custodians where they are payment versus payment and you don’t have any risks. You have to do it in a way that is not constrained by a single centralised balance sheet in order to really scale this across the whole market. If you look at any other point solution – I’ll just think about them as point solutions – even the biggest names, none of them have a big enough balance sheet to actually prime broke the whole space at scale. Right now, not even the biggest name you can think of that’s supposedly offering prime brokerage services. They also have other problems in that they’re generally not custodian-agnostic, meaning you have to custody with them, or their designated custodian, and they’re generally not liquidity-agnostic, meaning you have to take their liquidity if it’s offered by an exchange and maybe two or three other liquidity sources. You don’t get that true direct market access and freedom of choice around the custodian. If you’re institutional, what you really want to be able to do at the end of the day – thinking slightly down the road – is you want to pick your biggest, best digital asset custodian; you want a direct market access solution to access whatever exchange, whatever market makers you want; and you don’t want to have to think about at all the back-end – you know, netting and settlement and all that stuff. That is very, very, very hard to achieve. And I would argue you don’t achieve it with legal agreements and promises to pay and delivery versus payment and people going first and all of these parties becoming credit intermediaries that have very small balance sheets. I just don’t think those are scalable solutions.
Dominic Hobson 1:10:06: Rosario Ingargiola, thanks very much for taking the time to talk to us.
Rosario Ingargiola 1:10:09: Thanks, Dominic.