What Institutions Need to Know about Digital Assets in a Post FTX World?

[MUSIC PLAYING]

PETER HAYNES: Welcome to Episode 49 of TD Securities podcast series, Bid Out a market structure perspective from north of 49. I'm your host, Peter Haynes, and today we are going down the rabbit hold of digital assets to look at what institutions need to know in a post-FTX world. To be clear, we are not going to rehash what has already been well-told, and that is the story of FTX and its high profile former leader, Sam Bankman-Fried. That said, we will get some high-level thoughts from our guests today on some of the lessons learned in this recent and sad episode in financial markets. Today I'm pleased to be joined by two experts in the field of digital assets. Jeremy Lai, my colleague, is a portfolio trader on the TD Securities Equity Trading Desk in Toronto, but also has a side gig as Chair of the Digital Assets Advisory Committee in TD Securities. Mark Garabedian joins us today from Wellington Capital Management where he recently transitioned from the position of Director of Trading Data and Analytics to become Wellington's current Director of Digital Asset Strategy and Tokenization. Gentlemen, Thanks for joining me today.

MARK GARABEDIAN: Thanks for having us.

JEREMY LAI: Thanks for having us.

PETER HAYNES: Before we get started, I want to remind listeners that this TD Securities podcast is for informational purposes. The views described in today's podcast are of the individuals and may or may not represent the view of TD Bank or its subsidiaries, and these views should not be relied upon as investment, tax, or other advice. Well, Mark, I'm going to start with you. Given some of the recent headlines, any regrets on taking the new job?

MARK GARABEDIAN: Absolutely not. I am as excited about the technology and the innovation as ever.

PETER HAYNES: You know, I'm glad to hear that, and I do want to be serious here. And you are a very senior and experienced veteran of capital markets and we spend a lot of time together talking about market structure issues, and it says a lot about Wellington's commitment to digital assets to put you in this position. What is your mandate and where do you see the opportunities in the short to medium-term?

MARK GARABEDIAN: Yeah. My mandate, we'll start with that. So my mandate really is to experiment with digital assets, and in doing so, using our own internal resources to be able to purchase them, make sure that we can custody them, report on them, pass them through our systems, and to explore the technology itself with different parts of our business operations in order to give the firm optionality in the future. Should this technology have broad adoption and proliferation, we want to be in the position to be able to take advantage of it rather than trying to build at that point in time. And so my mandate very much is-- let's call it a big internal proof of concept. In terms of the short and medium-term opportunities, I guess what I'll say is, tokenization is really the fascinating concept here. I think of tokenization, people are always looking for the killer app with this technology, and I think of tokenization is as that killer app. You'll hear people say Bitcoin has been around since 2012, 2013, and it hasn't really been incorporated into the broader financial structure. What's taking so long? Why haven't we seen anything? Smart contract platforms, the most prominent smart contract platform, today being Ethereum, have only been around since about 2016. Applications didn't start really being broadly developed until a couple of years later. And so we're still very much in the early stages of tokenization. Tokenization allows for both existing real-world assets to have a digital representation on a blockchain, as well as for new types of assets to be created as a token out of the gate. And so I think the opportunity short-term-- not talk in my own book here, but the opportunities short-term are to learn and to get your hands dirty with this technology and to really understand it. Medium-term, what could potentially happen? So again, assuming broad adoption takes place and assuming proliferation of the technology, there could be high-value investment opportunities that come to market only in a tokenized form, and it's only going to be those firms who have the capabilities to be able to, as I said before, execute custody, have relationships built up, have systems that are prepared for this that will be able to engage those high-value investments and we owe it to our clients to make sure that we're there.

PETER HAYNES: It is unbelievable how quickly this space is evolving. We're going to dig in on some of the technical aspects of digital assets. We're going to try not to go too far down the rabbit hold on the technical side of things, but understanding some of the key terms such as smart contract, we'll dig in on that a little bit later on in this discussion. I'll pivot over to Jeremy here. You're chairing this Digital Asset Advisory Committee at TD Securities for the Global Markets Group. What exactly is the mandate of this committee?

JEREMY LAI: Yeah, so the Digital Asset Advisory Committee's main mandate, core mandate, is to advise senior executives on the evolution and the direction of the digital asset industry and how it can be applied towards elements that our clients care about, be that from infrastructure or investment. And understanding why our clients are caring about certain things, the direction the industry is going, and how we can stay aligned as a dealer and be competitive and understand the competitive landscape as it continues to evolve.

PETER HAYNES: It feels a lot like the mandate of this committee is somewhat similar to what Mark is doing, and generally speaking, what I hear about Wall Street in general, and that is everybody's keeping a close eye on this space, and making sure that they don't end up being the one firm that's-- to be on the buy side or sell side that's not able to act when the time comes, as Mark talked about earlier, whether it's an investment mandate or change in the way the business works. And I do think it's a good initiative for TD. And good luck with that, Jeremy. I want to ask a little bit here about some technical aspects from-- really, I'll call it traditional finance perspective, which is where my background is. There's a lot of questions about what went wrong at FTX and whether market structure was in part or completely to blame for the ultimate demise of this crypto marketplace. In many ways, this debate comes down to Centralized Finance, or CeFi, versus decentralized finance. And I'm a TradFi, or Traditional Finance person, and a lot of our listeners will be, too. Jeremy, can you please explain the difference between CeFi and DeFi, and the pros and cons of each structure?

JEREMY LAI: Sure. So in a centralized exchange-- and so that would include FTX, Binance, and basically all exchanges that traditional finance understands. A centralized exchange requires two parties, buyer and seller, to meet, but then have a third party essentially be the intermediary where that third entity is controlling the overall experience and the platform. And so when you as a buyer try to meet with a seller and you both bring your assets to then eventually exchange, be that you buy stock and then the seller delivers you the stock and you deliver the cash for it, ultimately, in a centralized exchange model such as FTX, control of those assets at some point has to reside with that central party. And so the idea of the FTX follow kind of supporting in many people's-- and it's the purist ideology of needing decentralized finance, is that in the FTX scenario, because there was a central party that ultimately had control of those assets as buyers and sellers were trying to connect to the exchange and trade, that central party ultimately was able to commit the fraud that it did. And so pivoting to a decentralized model, in a decentralized model, essentially there is a platform, an application that is essentially just a block of code that is algorithmically matching the buyers and sellers in a way that the application has designed the exchange of whatever you're looking to exchange to be done. And so how it works is if you had a sort of decentralized exchange like Uniswap, buyers and Sellers are connecting to Uniswap with their wallets, and there are intermediaries, there are participants who are balancing the asset pools on Uniswap. And so when someone tries to buy-- do a swap of an asset on Uniswap once they've connected their wallet, delivering, say, Ethereum and then swapping it out for USDC, that whole process is conducted and dictated by the block of code that has been implemented to have this transaction take place. And so at no point in this process are you then sending your assets to become to be in control of a third party at any time. So as long as you have trust in the application and the code that represents what the application will do-- or so the smart contract, you basically-- you do not have to trust a third party who is the intermediary to engage in this transaction.

PETER HAYNES: And we'll just be careful here. We'll say-- I know you use the word fraud, Jeremy. We don't know if it's a fraud, but certainly everybody can have their opinion with respect to what happened on FTX and we'll watch that one play out. And it's certainly playing out in the court of public opinion right now and on Sam Bankman-Fried's apology tour. But Mark, just carrying on that whole question of CeFi, or the FTX model versus DeFi, a recent report penned by Oliver Wyman entitled, quote, "Institutional DeFi, The Next Generation of Finance," lists several benefits of DeFi solutions for capital markets. And these benefits include mutual ledgers allowing for streamlined back-office processing, and atomic or real-time settlement, amongst other things. But it also lists some of the safeguards that are the key to the future of institutional DeFi, and these include KYC, AML, data privacy, and cyber protections, and I do understand that some cyber hacking issues are one of the problems that the DeFi world has dealt with. Where are we at in the life cycle of DeFi in terms of solving for these safeguards?

MARK GARABEDIAN: Let me start off with an analogy that touches on your last question. I've been working on my analogy game because I find that when you're talking to people who have not experienced this technology, it's very difficult for them to understand exactly what it that is the proposition here. And so trying to bring it to life. I think of a centralized exchange as-- like a store that sells a product. When you go into the store, there's a human behind the counter. You can ask them, hey, can I have the candy bar that's on the shelf there? They say yes, you have to pay the store owner. The candy bar is from the company, the store owner is the intermediary here that you're interacting with in order to exchange your money and receive the candy bar. Decentralized finance is a vending machine. So with a vending machine, there's no humans or anything. The machine is coded up and you're going to stick your dollar in and you're going to tell the vending machine that you want the candy bar. It's going to take your dollar and it's going to give you the candy bar out and the exchange is sort of done, it's all automated. When it comes to the internet, you can think of that centralized exchange model now is a little bit more like, say, Amazon where you're going out to Amazon's platform, Amazon is that intermediary for you. And importantly, you don't really own that asset until Amazon delivers the asset to your house, but you're probably giving up your money in advance to exchange for that asset. The internet version of the vending machine is a smart contract, and that's these decentralized exchanges, how they've implemented smart contracts. And so that, as soon as I give up my capital for whatever it is that I'm looking for, I receive the asset in exchange-- I immediately have ownership of that asset and the exchange takes place. You mentioned it at the outset. My background, I've been in trading for the past two decades. I've focused exclusively on electronic trading and market microstructure. I know the equity markets and the FX markets and the futures markets really inside and out. It was with that lens that drew me to this space. And really, in interacting with a decentralized exchange myself, was where my head sort of exploded. And so I'll just say, if you're in the world of electronic trading or even, really, the world of finance, I highly encourage you to go use this technology-- you don't need to make gigantic investments, but there's a lot of people who say, whoa, I'm in digital assets, I'm in crypto, I own some Ethereum on Coinbase. That's not the same thing. Taking the asset off of Coinbase-- and Coinbase would be the first people to tell you this. Taking the asset off of Coinbase, putting it into a digital wallet, and going out and doing things on chain, that's when you'll get the real sense of what this technology is. Where are we at in the lifecycle? I'm going to give you another analogy. It's 1997. You opened up your mailbox and you pulled out a CD, and the CD sent America Online on it. You said, what is this? You stuck the CD in your computer, you booted it up, you loaded the application. All of a sudden, the modem screen comes up and you hear all of these weird sounds. God forbid anybody pick up the telephone. If they pick up the telephone, you were you're immediately kicked offline. But if you did connect, the first thing you were greeted with was, you've got mail! You opened up your email inbox, it's the first time that you've ever seen email. Oh my gosh, what amazing technology here! Now you can use this platform in order to be able to send messages across the globe. The only problem was, at that point in time, you didn't know anybody else that had an email address, so there was nobody that you could send an email to. That's where we are with this technology. That point in time is where we are with this technology. So it is still very early days. All of the questions that we have with respect to how our traditional financial markets are run and how regulatory functions within our financial markets have been run, those questions are all coming to the forefront now. FTX definitely sped up the process. There are important regulations that we've had longstanding within traditional capital markets that need to be ported over on to this new technology before you're really going to be able to have institutional engagement. Things like taking identity and bringing identity on chain, making sure that AML and KYC policies are in place with interactions that occur on chain. All of that development is happening as we speak. There are some very new products that are in the market right now that do focus on digital identity, that do have permissioned pools where only KYC institutions can go in. That is all definitely happening. But again, it is still very early days.

JEREMY LAI: I'll make two comments, actually, on those analogies, because I thought they were really good, and I think dovetails into a couple additional elements of centralized finance versus decentralized. Is that-- sort of in that storefront analogy that you were talking about is implicit that-- implicit in that transaction is trust, that when you go to that storefront and you hand the guy your cash, that guy is then going to hand you the product and not run away or kick you out of the store. So there's a trust in that intermediary to allow that transaction to fully occur. In the vending machine model, and the decentralized model, using that vending machine analogy, the idea is that as long as you understand the mechanics of how the vending machine works, you put your coins in, the machine will just give you your product. And so the core trustless element of the decentralized finance model is that as long as the mechanics of how the vending machine works is transparent to you, you don't have to trust an entity, an individual, to then deliver you product when you put your coin in. That said, I think one interesting part about the decentralized finance that people are still trying to wrap their heads around or work around is that even in this, quote-unquote, "trustless model," there's still an element of risk or reliance that you have on even the developers of the decentralized application itself. Because let's say, again, using that vending machine model, you're selling product, you put your product in the vending machine so people can buy it. People then put their coins and then the coins are now in the vending machine that you are then supposed to take delivery on. Well, what if the developer-- or the creator of the vending machine just opens the box for the coins and runs away with them? And in a similar way, that's an issue in the early days of decentralized finance, that people were creating these decentralized applications, allowing this amazing utopia scenario where two parties are meeting and maybe on an exchange that they're transacting and wouldn't require the intermediary. But then the developers pull the project and run away with all the money. And so in that sense, there's still this process that we're working through to really understand what decentralized finance means and protect against the pitfalls of it all.

PETER HAYNES: I certainly was wondering as I was listening to both Mark and Jeremy speak there, what if the vending machine steals your money? And certainly, I think that's one of the issues that the industry is working on as you said, Mark, early stages. I don't think anyone's really happy to see what happened at FTX and some of the other celebrated missteps in the cryptocurrency space. But in some ways it might have been a good-- it might have been good for the future adoption of digital asset infrastructure as really, it's separated the discussion around crypto into components. First of all, the actual cryptocurrency itself, and then secondly, the blockchain or distributed ledger technology. Mark, what should institutional investors know about the use cases being worked on with blockchain technology?

MARK GARABEDIAN: Important distinction here. You will hear folks say there's a difference between blockchain technology and cryptocurrency and that you can have blockchain technology without having cryptocurrency. That needs to be further clarified. Yes, you can have a blockchain application without cryptocurrency. And the way to think about that is an intranet, which is like a website application that can be hosted within a company's walls and everybody within that company that's within the company's network is able to access that website, but nobody from the outside can access that website. There are use cases for intranet, and we know today that those use cases are smaller than the use cases for the internet. So a permissionless blockchain, that's akin to internet, and that cannot be released today as far as the innovations have taken us. That cannot be separated from the reward coins that are generated by those permissionless blockchains. And so, by that, I specifically mean Bitcoin for the Bitcoin chain, Ethereum for Ether for the Ethereum chain, even Dogecoin for the Doge chain. And so there are applications for a permissioned chain that's more like an intranet that does not have a necessary cryptocurrency component, that's anywhere where you need a shared ledger system and you have a limited number of people that want to validate transactions and you have complete trust in them to validate transactions. So if there's an individual vendor-supplier relationship and they wanted to share books between each other and it was just the two of them that were going to be validating transactions and they didn't really have a need for others to be validating transactions, a permissioned blockchain could be a great resource there. And I think you're starting to see some development in that space for a lot of companies. Having the shared ledger allows them to manage inventory levels a lot easier and more efficiently. When it comes to applications where, if you think of the financial markets, ideally you have an open system that everybody can have access to and everybody can transact on. That's probably going to be more of a permissionless system. You can here, think of the blockchain as sort of like being the public accounting system. The reward coins that are generated by the blockchain is what you're paying for people to validate all of the transactions within that system. And so it's important to separate those two. I think the permissionless blockchains, the internet version, I think that's where we're going to see some of the real broad-based applications for public consumption.

PETER HAYNES: So Jeremy, one of the first capital markets initiatives involving the integration of blockchain technology into markets was the Australian Stock Exchange's decision in 2017 to replace its equity clearing and settlement system known as CHESS with a blockchain solution. This project was originally scheduled to be done at the end of 2020, and last week, the ASX took a $250 million writedown on the project and mothballed it. One of the reasons cited for this project failure was the decision by the ASX to literally replace its entire clearing system with a proprietary blockchain solution instead of starting small and slowly integrating the new technology application by application. But to me, the bigger story is, it shows the difficulty of integrating DeFi into traditional financial infrastructure. And critics might suggest the benefits of moving to the blockchain do not outweigh the cost to the industry in terms of how difficult it is to complete one of these big transactions. Where do you stand on that, Jeremy?

JEREMY LAI: So I'm unsure if the challenges the ASX faced were necessarily specific to blockchain as a new upgrade to how they do clearing and whatnot and post-trade tasks. Or was it specific to ASX an organization that may have had challenges in being able to innovate and find the capacity to push this across the line? Because I think, trying to do a big bang approach isn't in itself necessarily incorrect because you could definitely see elements of how you do settlements not able to just transition small parts at a time. And so, for example, DTC has been running parallel projects that they-- I'm not sure of the current status of it, but I think it was like in August that they had announced the project, that they were looking to build this-- in parallel, build a new backend system that they can utilize blockchain technology to be the future of how they do settlement. But doing that in parallel means it is also that big bang approach, because one day they'll do that transition over if it becomes successful. It's hard to imagine why in itself the big bang approach is the reason why that they had failed. That said, there are smaller-vendor solutions that I can imagine maybe it would be more palatable if it really is a big change to take on a completely new technology. If-- for example, people who are trying to do collateral management applications, are looking to utilize blockchain specifically for that, there are new services, new vendors, that offer those kind of services for, say, like hard-to-borrows, that they're tokenizing the securities when they're difficult to move. Maybe if you were like piecing together small parts and then assembling that way, it would have been more palatable. But just from reading the story, I couldn't see that it was specific to blockchain that they couldn't get it across the line.

PETER HAYNES: It's interesting that the rhetoric in Australia coming out of that failed project-- and you're right, Jeremy, I don't think at all it was anything to do with the blockchain. I think it was just a scope creep issue, the project just got too big, it just ultimately-- there was a miscommunication or a lack of communication between their third-party assistant as well as the ASX staff. And at the end of the day, it's very interesting to read the critics now in Australia who are really jumping on top of the decision for the exchange to try to really do something that was about their shareholders more so than their stakeholders, and that's certainly a warning shot for some other exchanges that might want to do stretch assignments like this.

MARK GARABEDIAN: Let me say, we've seen this before, too. In the first wave of the internet, it wasn't the US Postal Service that created email, and it wasn't Encyclopedia Britannica or libraries that created Google. In this new wave of the internet, it wasn't banks that created stablecoins. It's folks like Circle that are out there creating stablecoins. And so I think it's-- we're seeing it's difficult for legacy institutions to come in and just adopt this new technology. It's much easier for smaller startups to come in and be very nimble and start with embracing this technology and implementing it. And that's part of the friction that we're seeing, because that's easy to do with email and with search engines, but this is finance, and those legacy institutions know the regulatory framework and the startups don't necessarily the regulatory framework, and so that's part of the frictions that we're experiencing.

JEREMY LAI: Yeah, and I think that's actually a really good point because it also-- it became prevalent when you saw centralized crypto exchanges like FTX trying to push their view on tokenization and whatnot. Because they had built from scratch, they were able to do certain functions like instantaneous liquidations and whatnot that were ingrained in their infrastructure that were the supporting elements of the value of tokenization. And being that they had built from scratch, you can imagine that they were able to offer those services and those exchange functions more easily than incumbents that were traditional finance trying to replicate and build those functions out.

PETER HAYNES: I want to be fair to the Australian Exchange, too, because there's a post-trade modernization project here in Canada for CDS, our clearing and settlement system. It has absolutely nothing to do with the blockchain, and I believe it's two years behind schedule. So it's not-- it just, I think, is more reflective of just the spiderweb of complexity built in to legacy infrastructure. And so that's going to be a really interesting aspect, Mark, as we run issues in parallel, as certain products, as you mentioned at the start, come on-chain and how that integrates well with the traditional finance markets. I see that in Switzerland, the Swiss crypto exchange, UBS launched a bond that's eligible to be traded both on their digital exchange as well as on the traditional exchange either through traditional finance means or through decentralized finance. And how these products run in parallel, how a market integrates between the decentralized blockchain market versus the traditional market through this transition will be really interesting. And just-- Mark, that leads into-- you mentioned the term stablecoin. There's a few terms that I think we all in traditional finance need to understand. so maybe, Mark, if you-- and maybe you have analogy, but can you explain what exactly is a stablecoin? And you also mentioned a term earlier, smart contract, which I believe has to do with just what's on the blockchain, the information related to the product you're trading. But can you just explain more technically what those two terms mean?

MARK GARABEDIAN: Sure. And they relate to each other, so this will be pretty easy. A smart contract is a computer application that can be written on a blockchain platform. Importantly, when Bitcoin was first developed, it was a permissionless blockchain system, which meant that validation could be done by anybody who went through the mining process in order to do validations and they could receive Bitcoin as a reward. And then the Bitcoin itself could be stored and transferred over the internet. And so you have the ability to store value and transfer value over the internet. And today, Bitcoin is very largely the same as it was when it was originally conceived. Some would even point to that being a strength of Bitcoin. It wasn't until Ethereum was developed and put into place and the smart contract language solidity was developed at the same time and put into place that we actually had the additional layer of being able to write programs and applications onto that permissionless blockchain technology. And again, that was Ethereum and its conception. So a smart contract is a computer program that can be written on a blockchain. That computer program can automate functions, it can house all sorts of code, and importantly, it can create tokens which are digital representations on a blockchain that can allow you to store and transfer value. Stablecoins is the first broad application of tokenization where we are taking real-world assets-- and in the case of, say, USDC, Circle's USDC, we are taking real physical dollars, we are locking them up into a custodial framework, traditional finance, locking them up into a custodial framework, and then we are issuing out tokenized versions of those dollars on chain. And so in the case of Circle, those are one-for-one fungible. I've got one USDC token, I can convert it into one fiat US dollar that the custodian is holding, and I can have that dollar given to me physically, or euphemistically, you can have it transferred to your bank. Or I can take that digital version of the US dollar and I can move it across the Ethereum protocol and have interactions with all sorts of applications on the Ethereum protocol. I say it's the first application. Really, what-- I think the prize here is if we can get more real-world assets housed within custodians, digital versions of them issued, distributed, and then be able to be stored and transferred over the internet using blockchain technology.

PETER HAYNES: So Mark was talking about stablecoins and one-for-one, and I know that some of the issues that have happened involving stablecoins has been whether they actually do have the backing asset one-for-one. And Jeremy, we've heard through this entire FTX staga-- saga, I should say. We kept hearing this term proof of reserves. What exactly does that mean?

JEREMY LAI: So proof of reserves is a process whereby an exchange is demonstrating that it has, on a one-to-one basis, all of the assets that customers have deposited onto the exchange. And just for clarification, because in crypto exchange structure as it is today, an exchange is not just like the TSX where it's the venue for people to buy and sell on exchanges. It also acts as the broker and the custodian in many cases or custody entity. And so when you deposit your USDC or any other kind of crypto onto the exchange to be-- assets that you want to then exchange for whatever reason, you want to make sure that the exchange is not then taking those assets and sending them somewhere else for, say, lending or just betting on the value of those assets by trading them on their own book. And so there's supposed to be that idea of segregated assets for clients. So the proof of reserve process, what it essentially does is it takes all of the customer data and turns it-- and the premier way to do this is to turn it into something called a Merkle tree. Not to get into the technical aspects of what that really means, but essentially it is a way to structure the data so that it's easy to see whether a particular customer's assets are represented in this overall data structure. So using this Merkle tree structure, when you take every customer's assets-- so say you've deposited 10 USDC with the exchange, I've deposited 1 Ethereum, and you combine all of that data into this tree, when you as an individual want to verify that your assets are on the exchange, you can take this tree and you input the value representative of your assets into this tree. And-- without getting into all the technical elements of how hashing and whatnot works, it should result in definitive results based on how the tree works to show you that your asset is on the exchange. There's complexity to how the Merkle tree actually works, but that's basically the premise. So there is definitive way when they show you this data structure to verify that your assets are there. The other elements of proof of reserves is that the exchange itself has to demonstrate that they actually have control over the wallets that these assets are sitting in. And so a third party auditor, as part of this proof of reserve process, will give instructions, like send a certain amount of crypto at this certain time to this address, certain tasks that you would have to do to demonstrate you have control of the actual assets that are being proven to exist in these wallets. So part of that complete process is-- the proof of reserve process, that exchanges now are being asked to do to demonstrate they haven't sent client assets elsewhere. Now one of the issues that exist with this process, even though it's better to have it rather than not, is that one of the risks is that the tree that you're trying to verify, which represents the current state of all of the deposits that customers have with the exchange, is a snapshot in time. And so what some exchanges, in theory, that are trying to be nefarious could do is they are actually, say, doing something fraudulent like lending out your assets elsewhere even though they're not supposed to, but then when they take the snapshot for the audit, they transfer all the assets back just to do the snapshot for the tree, and then once the auditing is done, they start doing their nefarious actions again by sending them back to whatever they were doing it with. That is one weakness. And the other weakness being that if there are liabilities for the exchange that are not on the chain-- so say they just owe money to-- a client has, say, a fiat deposit, something that is not a liability that can be verified on chain, then that would also not be a verifiable proof of reserve.

PETER HAYNES: I guess we just have to hope the auditor isn't in the metaverse and will be all right here if they're doing their job.

MARK GARABEDIAN: This is an opportunity to just really quickly touch on what's unique about the way that these exchanges have come onto the scene and how they operate. We don't live in a world where if I go buy Microsoft stock, that NYSE hangs on to my Microsoft shares. That trade ends up going back to wherever my brokerage account is and wherever it's going to be custodied. That, in the crypto sphere, is something that people have to do themselves. Self-custody is available. There aren't really retail custody options. The exchanges allow you to custody your assets on the exchange. It sits in an omnibus structure. Many people-- I don't know if they know this or not, but the trading that occurs on a cryptocurrency exchange like Coinbase, or Gemini, or Binance, or FTX, the trading that occurs on those exchanges never touches the blockchain. Has nothing to do with the blockchain, there's no blockchain transactions. This idea of proof of reserves is inherent because if you're leaving your assets on the exchange, you're in their omnibus wallet. You're basically-- you have assets that are in their control, that are in their ownership, and they have an accounting record that says for the benefit of you, a certain piece of it. And then if they have a bankruptcy or a situation what happened to FTX, who really knows where your assets are. And so taking your assets off of the exchange and putting them into segregated custody-- for retail, that can look like self-custody, and for institutions, that can look sending it to a qualified custodian, now your assets are on chain, and now you can verify those assets on chain, and they are definitely separate from anything that can happen to the exchange. That's one of those things that's prevalent in traditional financial markets, and we've had that separation of duties for a very long time. The crypto ecosystem has not gotten there yet.

PETER HAYNES: And Mark, yeah, let's be fair. Let's think back to the financial crisis. I don't think anyone really understood that CSA commissions could be co-mingled and ultimately end up in the bankruptcy and not end up getting to the people that deserve them. That was a problem around Lehman, so these things do-- unfortunately, through crisis we learned, and hopefully we learn and never let a good crisis go by. But Mark, I want to ask you here, if you were to guess which country or marketplace will be the first to trade tokenized equities, and do you think it'll be on an exchange operator that we know of, like the NYSE or TMX?

MARK GARABEDIAN: So I can say that I'm impressed with the Swiss regulators and the Swiss government. They have created some great frameworks for digital assets to be issued and traded, and there are some tokenized securities that have come out of that framework that are live out today. I know that the European Union is working on the MICA, which is the Markets in Crypto Assets akin to MiFID. I know that the Monetary Authority of Singapore has done some great things in their region as well in terms of creating their closed systems, but they are pretty broad-based and use blockchain technology and tokenized securities. It's hard to say at this point which regions will come first, or really even who the operators will be that put this-- that put these assets broadly into the ecosystem in terms of specifically tokenized equities. I would say that there has been the underpinnings of some proliferation in tokenized-- the tokenized fund space. And I think one of the bigger headlines was KKR, tokenized a portion of their private equity fund, they did that in conjunction with Securitize, and then they issued it out on the Avalanche blockchain. Private equity funds putting those in tokenized forms-- that can be compelling because it creates a lot of efficiencies, both on the accounting and recon, but then also on the asset-gathering and distribution side of things, and that's a tokenized security. And so there are situations where tokenized securities are out there today, but yeah, we're at the whim of really these global regulators. To some degree, there will need to be some global coordination on this as well so that we're not maintaining this period of regulatory arbitrage and differences across borders.

PETER HAYNES: What about the issue around velocity and the ability to process transactions and the slowness of current chains? Is that being-- is that getting better?

MARK GARABEDIAN: So when you put that CD in your computer and boot it up on the modem and you had all those weird screens coming up, and God forbid if somebody picked up the phone, your question to me is like, well, when are we going to watch TV on the internet? So it's a little early. Yes, those-- scalability is the big outlying question. There are some people who think that this technology cannot scale any further beyond where it currently is, but Layer 2 technologies for Ethereum, which offer a tremendous degree of scale scalability, started launching last year-- middle of last year if you're talking about Arbitrum which I think is the largest roll-up in terms of Layer 2 technologies. We have zero-knowledge technologies that are coming online next year, and there should be some zkEVMs that are issued out next year. Those are scaling technologies that should, again, further increase the output. So I think-- again, in 1997, if someone was to say to you, hey, we're going to be watching TV on the internet, you would have been like, oh, you're crazy. Today, if somebody says, oh, we'll be trading Treasury bills over the internet on DeFi platforms or in tokenized form, it feels similarly crazy, but it's not always been a good bet to bet against the ability of technology to scale.

PETER HAYNES: Well, I think we can't complete this discussion without addressing the elephant in the room. And despite the advice from every lawyer in the world, Sam Bankman-Fried continues on his personal apology tour and he's sticking by the notion that FTX is a big mistake and that he did not intentionally co-mingle client money with Alameda money. Jeremy, I'm going to start with you. Based on what you've heard, do you believe anything he is saying? And what do you think are the key lessons that are learned in this ugly chapter in financial markets?

JEREMY LAI: So he frames his responses in a way that allows him to say things that are kind of true, but he's-- and this is just my opinion, masking an underlying intention that may have been the-- and the intent of so some of the structures that he had in place. So co-mingling of assets. So the key thing that we understood once he had his interview was that it was Alameda-- Alameda as an entity is essentially acting like a client of the exchange FTX. And where the core losses for FTX as an exchange occurred was because of margin leveraged positions that Alameda held on the exchange, essentially. Alameda was-- I think he said something like 40-plus percent of the volume that was being traded was Alameda itself, probably in many capacities as a market maker. But Alameda itself taking margin positions on the exchange, which is no different than-- in theory, it's no different than any client of FTX taking large margin positions. Sometimes-- let's say you're a massive market-maker who is running their a market-making program on the exchange, you might have to potentially take on leverage, or you just an individual and you're just taking wild bets. So in that element, it's not really different. But there was clearly a sense that preferential treatment was given to Alameda in its relationship to FTX. And so when the margin became too risky or would have otherwise passed limits that would have caused liquidations for normal clients, those didn't occur in a way that needed to occur so that Alameda would have been liquidated and the assets for the clients would have been protected. So in this structure, when FTX as an entity, which obviously has, in some way, relations, either officially or not, with Alameda, allowed Alameda to take these outsized bets using margin that was offered by the exchange. Is that commingling of client accounts, of client assets? In the strictest sense, maybe not, and which is why he could make that statement. But that doesn't override the idea of fraud that could have been perpetrated given that the Alameda entity was given these loans-- or the access to margin and the ability to prevent liquidations that most people would not. And so if you step back from here and think of it as, you're a prime broker, you have a top-tier client, the client is not just top tier, but is a relative of yours, and that top-tier client is facing issues with margin. They would have faced liquidations that any other client would have normally faced. But you give preferential treatment to that client because of your relationship to that client, and you trust what they say, that they're basically claiming that, just give me some time, I'm going to find the money, I'm going to raise some money and then we're going to fill the hole, and then they don't. And then the prime broker goes under and the bank goes under as well. Everyone else loses money as well. Is that fraud? Probably. Is that commingling of assets? Maybe he could get away with saying it wasn't.

PETER HAYNES: So Mark, maybe we'll l carry it from there but lessons learned. Do you have anything to add to what Jeremy said there or a different perspective? And do you have some key lessons that you're making sure management of your organization is aware of through this mess?

MARK GARABEDIAN: I'll say, our focus at Wellington is to do our research and do due diligence so that we can make the best investment decisions for our clients. Through that approach and with that lens, we determined that FTX did not meet our investment standards. We had concerns about their governance structure and specifically the fact that they did not, and I believe still do not, have a board that were overseeing their business activities. And so that, for us, was a nonstarter. So I think that's one lesson, process works. The other thing that I'll say is, again, we've experienced this in traditional markets. The idea of conflicts of interest within shared business models. You have an exchange that operates as a broker, that operates as a custodian, that has all of these different businesses. We look forward to the regulators coming in and clarifying the distinction and creating some smoother rules of the road. Again, this is a really unfortunate circumstance that occurred. It set back adoption, which is probably the most important piece. The building is still taking place. Everything on chain is still being worked on, but it slowed the pace of adoption for broad consumers because trust has been broken. And I think our best shot at regaining that trust is regulators coming in and implementing some of the tried and tested policies that have worked for generations.

PETER HAYNES: I think the last exchange that owned a broker-dealer was the NYSE when they bought Arca and turned it into an exchange. They inherited Wave Securities and ultimately had to sell it off-- I think they sold it to Merrill Lynch back after that deal happened. And yeah, there are reasons why there are pillars up here and they don't allow all of these activities to occur in one organization. Mark, you mentioned that you think we've been set back a little bit here. Interestingly enough, a couple of the institutional investors I've spoken to recently actually have said that the FTX saga has woken them up to the need to understand the applications of distributed ledger technology. What's your advice on where investors and traders can go to learn more without being too buried in the minutia?

MARK GARABEDIAN: I advise people to use the technology. This is-- could you imagine trying to explain the possibilities of the internet and email to people who never touched internet and email? It's just so difficult to do. Your best way to learn is really just going to be through trial. It doesn't-- don't have to move your entire life savings. You can onboard onto Coinbase with $10. You can take that $10 of fiat, you can buy some Ethereum with it, you can move the Ethereum from Coinbase to the Coinbase wallet, or a MetaMask wallet. And then you can go to Uniswap, and you can do your first decentralized exchange, and you can swap that Ethereum for USDC. I think just doing what I said right there will be enough to open people's eyes up to exactly what it is that we're talking about and will probably whet the appetite to do more.

PETER HAYNES: Well, we learned a lot, we covered a lot in this conversation here today. On behalf of TD Securities, I want to thank you, Mark, for joining us; Jeremy, for helping us understand some of these technical issues around not just the FTX saga, but more specifically, some of the things that we've learned around that and what we need to know about this technology, is we really don't know how quickly this technology is going to intersect with our daily lives and capital markets. But certainly Wellington is of the belief, as you said, Mark, that we need to be paying close attention to this. And I'm just curious, Mark, how many of your peers would you say are also-- like are you noticing a lot of people in the asset management space that are-- or a lot of organizations that are essentially putting their equivalent of a Mark Garabedian in the same chair as you're in?

MARK GARABEDIAN: Yeah. It's early days. There are some firms that are ahead of us, there are some firms that are alongside us, and there are some firms that, thus far, have decided that not to come along for the ride, or just watching it from afar. I would say everybody is coming at it from a little bit of a different lens, but we're all ending up in somewhat similar places thus far in terms of what we're focusing on and what the initiatives are. But it's a lot of household names that are around the same spot in the journey that we are.

PETER HAYNES: Yep. I think a simple Google search of asset managers and tokenization and you can see a pretty lengthy list of parties that have been getting involved in this space in various aspects of it. So thank you very much, gentlemen, for joining us today, and we'll look forward to perhaps having you join us again down the road as we see how this business has evolved. Take care.

MARK GARABEDIAN: Thank you.

JEREMY LAI: Thank you.

[MUSIC PLAYING]

This podcast should not be copied, distributed, published or reproduced, in whole or in part. The information contained in this recording was obtained from publicly available sources, has not been independently verified by TD Securities, may not be current, and TD Securities has no obligation to provide any updates or changes. All price references and market forecasts are as of the date of recording. The views and opinions expressed in this podcast are not necessarily those of TD Securities and may differ from the views and opinions of other departments or divisions of TD Securities and its affiliates. TD Securities is not providing any financial, economic, legal, accounting, or tax advice or recommendations in this podcast. The information contained in this podcast does not constitute investment advice or an offer to buy or sell securities or any other product and should not be relied upon to evaluate any potential transaction. Neither TD Securities nor any of its affiliates makes any representation or warranty, express or implied, as to the accuracy or completeness of the statements or any information contained in this podcast and any liability therefore (including in respect of direct, indirect or consequential loss or damage) is expressly disclaimed.