Episode 78 of the Public Key podcast is here and we are happy that you love the refreshed look. Institutional adoption has always been deemed as crypto’s much-needed catalyst. In this episode, we speak with Kyle Downey (CEO & Co-founder, Cloudwall) to discuss the risks we are not mitigating and keeping institutional traders on the sidelines of DeFi and crypto.
You can listen or subscribe now on Spotify, Apple, or Audible. Keep reading for a full preview of episode 78.
Institutional crypto investment has long been the key to mass adoption in the cryptocurrency industry but with the chaos of 2022 and the demise of several large industry players, we have to start assessing what the real risks are in this space.
In this episode, Ian Andrews (CMO, Chainalysis) gets to speak with Kyle Downey (CEO & Co-founder, Cloudwall), whose team is building what they hope is the world’s most sophisticated digital asset risk platform named Serenity.
Kyle emphasizes the need for better infrastructure and risk management tools in the crypto space, particularly for institutional investorsWe also discuss the challenges and opportunities in the DeFi space and the importance of finding a balance between privacy and transparency.
Ian and Kyle double down on the industry’s need to prioritize counterparty risk and also touch on what real-world asset tokenization will look like for portfolio managers.
Quote of the episode
“The [risk] tools for institutional traders are not suitable yet for the people who are going to be coming in 2024 and 2025. I do believe we’re behind the eight ball in terms of infrastructure investment in the space and that’s across the board,” – Kyle Downey (CEO & Co-founder, Cloudwall)
Minute-by-minute episode breakdown
- (2:02) – Kyle Downey’s early foray into Bitcoin, Mt. Gox and crypto trading
- (8:08) – Need for more investment in institutional-grade tools and systems for crypto
- (13:20) – OTC markets, DeFi platforms and capital efficiency challenges with transparency on the blockchain
- (17:06) – Momentum and interest in DLT and crypto within financial institutions
- (22:25) – Potential impact of risk management platforms and limitations of risk systems in detecting fraud
- (26:02) – Importance of addressing counterparty risk in the crypto industry
- (31:45) – The role of risk management tools in attracting risk-averse investors
- (36:13) – Open questions and challenges in regulating DeFi
- (42:27) – Serenity’s upcoming release and portfolio analytics suite
Related resources
Check out more resources provided by Chainalysis that perfectly complement this episode of the Public Key.
- Newsletter: The Seven Deadly Sins of Finance: Building a better financial architecture in institutional crypto
- Announcement: Copper partners with Cloudwall, Bringing Stronger Operational Risk Controls to its Institutional Investors
- Blog: Institutions in Central, Northern, and Western Europe Broaden Horizons with DeFi and Web3 Experimentation
- YouTube: Chainalysis YouTube page
- Telegram: Chainalysis on Telegram
Speakers on today’s episode
- Ian Andrews Host (Chief Marketing Officer, Chainalysis)
- Kyle Downey (CEO & Co-founder, Cloudwall)
This website may contain links to third-party sites that are not under the control of Chainalysis, Inc. or its affiliates (collectively “Chainalysis”). Access to such information does not imply association with, endorsement of, approval of, or recommendation by Chainalysis of the site or its operators, and Chainalysis is not responsible for the products, services, or other content hosted therein.
Our podcasts are for informational purposes only, and are not intended to provide legal, tax, financial, or investment advice. Listeners should consult their own advisors before making these types of decisions. Chainalysis has no responsibility or liability for any decision made or any other acts or omissions in connection with your use of this material.
Chainalysis does not guarantee or warrant the accuracy, completeness, timeliness, suitability or validity of the information in any particular podcast and will not be responsible for any claim attributable to errors, omissions, or other inaccuracies of any part of such material.
Unless stated otherwise, reference to any specific product or entity does not constitute an endorsement or recommendation by Chainalysis. The views expressed by guests are their own and their appearance on the program does not imply an endorsement of them or any entity they represent. Views and opinions expressed by Chainalysis employees are those of the employees and do not necessarily reflect the views of the company.
Transcript
Ian:
Hey, everyone. Welcome to another episode of Public Key. This is your host, Ian Andrews. This week I have the founder and CEO of Cloudwall, Kyle Downey. Kyle, welcome to the show.
Kyle:
Thank you. Thanks for having me on the show, Ian. I’m really excited about this conversation.
Ian:
I am super excited to dive into the technology that your company’s building. I think it’s an area we really haven’t gotten into on the show, related to how institutional investors think about risk management across their crypto portfolio. But before we go there, I think I’m always interested in how people got into crypto. You worked at Morgan Stanley for nearly two decades. In your last role, you were managing director. Talk a little bit about your career there and what led you into the world of digital assets.
Kyle:
So it goes back a decade on the crypto side. So in the middle of the 17 years that I spent at Morgan Stanley, I spent seven years out in Asia, three years in Shanghai, and four years in Hong Kong. And that actually ends up dovetailing to the story of Cloudwall a little bit later on. But at the time in Hong Kong, I remember stumbling across a news article, South China Morning Post, I think, that was referencing a Bitcoin mine in Kowloon, so relatively near our home. And I remember thinking, what is Bitcoin and why are they mining it? I had not heard of this at all.
So this was a couple of years in already, in 2013. So I googled it and I started reading the white paper and I was utterly fascinated by it, by this idea. And then I started looking at what was going on in the trading side. And the thing that just came to me relatively early on was it seemed like people were using these internet technologies to rebuild a lot of things that were very familiar to me in my day job. And it was day traders and retail and all this, but they were building exchanges and they had free market data and it rhymed a lot with the institutional side.
And I started to wonder, is the day going to come where the things that I was doing at Morgan Stanley and what was going on in crypto were going to converge. So over that long period of time, that was always in the back of my head, when is that moment that that is really going to take off and what triggers it? Because at the time, this was not obvious. In 2013, this is the time of Mt. Gox. People may not know this, but Mt. Gox stood for Magic the Gathering Online Exchange. So people were trading Bitcoin on a trading card website. It really was a joke. Bitcoin was a joke before Dogecoin was a joke.
And then it really took off and suddenly it wasn’t a joke any longer, but it wasn’t really obvious that institutional traders would touch this thing at all. And you had a lot of those scandals around used on the dark web and Silk Road and all of that that were really dominating. But still from a technologist point of view, so at the time I was working on equities and equity derivatives and e-trading technology, which that was the latter part of my career and that kind of final role that I was in when I left MS.
So we were building market making and hedging and liquidity provision systems for MS traders, mostly equity derivatives, mostly options, actually, at the time. So when I was in Hong Kong facing off with an equity derivatives desk, and it just struck me despite all the weirdness of it, how similar it was to what was going on one floor below me from where I was sitting in IT. And so that was for a long while, just okay, this is cool, this is interesting. I was trading myself. I built some automated trading tools along the way. I was writing about it at one point.
Ian:
I’m curious what the reaction of your Morgan Stanley colleagues, like did you come into work and say, “Hey guys, have you heard about this Bitcoin thing?” Did anybody react?
Kyle:
Absolutely, but you get a real reaction when you resign to do it. But people knew because they saw my writing, they were aware I talked about it. I had a very senior quant sit me down and basically say, “This is utterly without value. What are you doing?”
And I had conversations with compliance along the way, because obviously when you’re working in a big bank, there are rules around outside personal trading. And so trying to do the right thing. Hey, I’m doing this, is this okay? And they were like, “Yeah, don’t worry about it. This has nothing to do with what we do.” And it was to the point where I asked several times, “Are you sure? I’m [inaudible 00:05:13] registered. Is this okay?” And the really interesting thing, of course, is that when it gets out that you’re interested in this sort of thing, people will come up to you like, “Hey, I’m reading this, too.”
Ian:
That was totally my experience as well, was when I took the job here at Chainalysis, all of a sudden people that I thought I knew well and had known for long periods of time came up to me and go, “Ian, you’re getting to work in crypto?” I’ve been trading Bitcoin since 2014.” Or, “I got into Ether in the ICO.” Or, “I’ve had mining rigs in my basement for the last five years.” And I had absolutely no idea. It was a very kind of underground movement within a circle of technology people that I knew who consciously chose to not mention it casually at work, which surprised me.
Kyle:
Yeah, exactly. Out on the trading floor, both quants and traders, there were multiple people who were doing it nights and weekends for fun, who got into it. I heard people at other hedge funds that were sort of running crypto accounts. And on the tech side, after I left, I found another managing director who had a whole fan cooled set of Dogecoin rigs in his basement at one point. And so he’s sort of like a mini Doge father.
And I was doing it, too. My daughter and I built an Ethereum GPU rig at one point in the early days of Ethereum that was running for a while until all the cards burned out. So people were doing it as just this hobbyist thing. And then around the, I would say acutely during the pandemic, but probably around 2019 or so, just things started popping up. This is the first announcements around GS setting up a trading desk, which they stopped and then they started again, and it was starting to come into the light and people were starting to take it a lot more seriously as a product for institutional trading.
And of course then the space races ahead again, and you have the summer of DeFi and that’s well beyond what most people on the institutional side had any tolerance for. The safe space was still Bitcoin and Ethereum and things like that. But I remember the survey that GS did of some of their family office clients and just being startled by the numbers, not just on the number who were actually already through mostly indirect vehicles invested in crypto, but the ones who had it on their roadmap for the next year, that they were already talking about it.
There was another survey, it was like of the analyst class, I think at one of the big banks, and again, showing the dynamic around the demographics, it was overall interest to the clients, but then how many of your just out of school analysts coming into the bank own crypto? And it was almost all of them. And so you have that too, that there’s this generational shift where there’s a lot more interest with the millennials and Gen Z that was starting to come into Wall Street because they were starting to take jobs.
But still, leaving in 2021, I got a lot of why are you doing this? My co-founder got it probably even more. You’re ruining your career. You’ve been here for 17 years now. You’re doing this crazy small thing. I had one very senior MD tell me flat out, “Yeah, I would’ve never guessed that you would do this sort of thing.” So I think for those of us who did that transition, it was easier than it was a couple years before, but there was still a lot of, are you sure you want to be associated with this? Is there reputational risk around it?
And a lot of that was outdated. And I mean, Chainalysis, you guys have done reports on this. A lot of the headlines don’t take into account the growth of the overall ecosystem, and they’re still quoting numbers of hacks and activity in the dark web and on chain and not recognizing that the percentage keeps going down over time. So in many ways, some of those people, those views are a bit outdated. They’re pointing to things that were real but not necessarily dominant anymore.
Yeah, and thus what ended up happening is I was planning a move probably around 2019 and just kind of late stages looking at it. And then the pandemic hit. And made a personal decision to not leave at that time. And so I just went heads down, because there was an amazing thing within four weeks of getting a 150 person department out of offices and into their homes and really just spent a lot of time and energy on that. And I stopped looking at crypto for the whole summer of DeFi and all the way until the restrictions on the pandemic in the summer of 2021 lifted.
And at that point I was on a family holiday in New Hampshire, where I grew up, and I started looking at it again. It’s like, I was really enjoying this. I’m kind of bored after hours. I’ll start looking at it. And it was, what the hell happened in the last 18 months? And I just felt this great sense of urgency that like, oh my gosh, it’s this thing that I’d kind of been watching for years, suddenly it’s moving really, really quickly. And I resigned in less than six weeks after that realization. I had a three month notice, and the company started in the fall of 2021.
Ian:
Amazing. You made a comment earlier about how when you first got into the space, you looked around and you saw that people were recreating a lot of the systems that already existed inside of institutions like Morgan Stanley. I’m curious, what’s your perspective now in the fall of 2023? Where are we on that progression of rebuilding the infrastructure that supports traditional finance, if you will, inside of the crypto ecosystem?
Kyle:
And there’s a degree of self-interest in this statement in that I’m-
Ian:
You’re absolutely building part of it.
Kyle:
So of course I’m going to come out and say we’re not spending enough on it, but we’re not spending enough on it. The tools for institutional traders are not suitable yet for the people who are going to be coming in 2024 and 2025. I do believe we’re behind the eight ball in terms of infrastructure investment in the space, and that’s across the board. And so that’s trading and custody and data.
I’ve had portfolio managers from large traditional hedge funds say to me flat out, not a single vendor and they work with series B and above large vendors, is institutional grade, not one. And so for those of us who came out of this space and have that standard, you see things happening and say like, this would be utterly unacceptable at a large institution. And I’m not talking about fraud or anything like that. I’m talking about just how you operate and the resilience and reliability of the systems and all that. We’re still really investing a fraction of what Wall Street is investing overall. So there’s that problem first.
The second thing is that I feel like in some cases we have reproduced things that we should not have reproduced in terms of creating centralized infrastructure, in terms of repeating mistakes, in terms of large opaque OTC markets with not very good automation around them. So we’ve created a lot of systemic risk in the space by not looking to some of the strengths of the technologies that this is built on and reimagining it.
So I think, I hope, that one of the values of people coming into this space who have that background but are also passionate about the opportunities and actually using this as a chance to reinvent, will do more than just say, “Let’s take NASDAQ, but put Bitcoin on it.” We can do better than that. And I think even the traditional exchanges, they know that too. They’re looking at ways to reimagine their businesses by incorporating this technology, and that I think we haven’t done enough of.
And a PM that I respect in the space, head of one of the early digital asset funds, one of the things she said is part of the problem has been the start and stop. The boom/bust cycle within crypto is people get into it, they start building, they start to build infrastructure, and then they get wiped out. And then we do it again. There needs to be a little bit more of a long-term commitment to actually invest in this. And I hope that the financial services industry as a whole sees it as an opportunity to redo systems from the seventies and earlier in a much, much better way.
Because ultimately, crypto is not the only application of this. I firmly believe there’s going to be widespread tokenization of traditional assets as well, and this is going to have huge implications for asset managers. So we have to solve this for everything, not just for crypto anyway. And if that is the case, we are massively underinvested right now in the space.
Ian:
When you bring up the large and opaque OTC markets, what’s the alternative to that? What’s the right way to do it? Is it all on some sort of DeFi type platform, where it’s all in the open?
Kyle:
So I think the ultimate destination is via DeFi protocols. I think the more general principle is that a lot of this has to be manifested on chain. The two big problems with that, the first is capital efficiency. So if people are doing large OTC options deals, they’re not going to over-collateralize that. There’s no point. The DeFi option protocols, they don’t have the capacity to absorb those really big deals and they just don’t make economic sense for someone, which is why the OTC market exists.
But the second issue is that that whole mechanism of having bilateral trades, it doesn’t work with a fully transparent blockchain. And so that’s the other issue right now. As you may know from some of my writing online, one of my side interests is privacy preserving technologies on blockchain, particularly ones that allow for selective disclosure, because there are certain financial use cases where full transparency doesn’t work and full obscurity doesn’t work. You need something in the middle.
And the OTC market is a great example of that. Like, you and your counterparty both need to see the details of the deal. You don’t want your competitors to see the details of the deal, but to assess overall systemic risk, you may wish to have high level summaries of what’s going on in the market visible to other parties, to auditors, to perhaps regulators. There are wonderful examples of this from early stages of my career. I worked on credit derivatives, back office technology with pre-financial crisis, what was being done to try to get a handle on the fixed income derivatives market, particularly CDS, where they wanted to have a central view of the net exposure across the street. Because it was going through fax machines. It was just not visible.
And so that’s a really good example, I think, of something that could be done better on blockchain, but maybe the technology is not entirely there yet. And so we’re in this kind of transitional phase where quite frankly, I’m like, have I been teleported back to 2004? You’re seriously negotiating this deal on telegram and then emailing a PDF summary of the deal terms, and then you have a bunch of operations people trying to make sure nobody made any mistakes. That we’re doing that in crypto in 2023 is horrifying to me. There are better ways to do that, but it’s kind of a case of [inaudible 00:18:35], there’s no good alternative to it right now that meets those requirements of confidentiality and capital efficiency.
Ian:
Yeah. Do you think firms like your previous employer, do they have the appetite to actually try and go rebuild some of the infrastructure around or on top of blockchain technology? And I know they have incredible technology budgets, like the spend at Morgan Stanley or JP Morgan is measured in billions of dollars annually. So there’s capacity, but it’s always seemed to me like we still run most payment systems on mainframes at the end of the day. So like you said, 1970s technology predominates when you dig deep enough under the covers.
Kyle:
So I mean obviously I can’t, particularly several years out, I can’t speak for where they are now. What I can say is for projects that JP and others have spoken about publicly, there is a lot of engagement and there’s a lot of interest both on the DLT side and the crypto side. The DLT side is easier to move ahead because it doesn’t have the regulatory constraints. And so to the extent that it doesn’t maybe move as quickly, it’s more like just typical budget prioritization, whatever.
So I’ve seen a lot of companies starting to move out of what I would call the innovation lab phase, where they were kind of stuck for a couple of years. It’s like blockchain was kind of alongside quantum computing. Like you would say, “Hey, we’re doing all this cool stuff. We’re doing blockchain, we’re doing quantum, we’re doing all this big data stuff. Please come work for us rather than going to work for Google.” That was the play for a while and blockchain was trapped in that.
But I think if you look at some of the original sins of financial IT systems, one of the big ones is reconciliation. It’s having multiple masters, multiple copies. I cannot tell you how many systems had these issues that I’ve seen over the years, where they just had copies of copies of copies of data, and then of course someone comes in and they’re an internal audit function, a regular whoever, and they’re like, “This number in this financial report, where did it come from?” It’s like, well, kind of awkward. It went through 15 different Sybase databases and a person touched it manually over here. And we’ve got all these scripts rubbed together that we’re fairly sure the number’s right.
This is architecturally not great, and people recognize this. They know this pushes toward central databases. Taking Goldman as an example, one of the great powers of [inaudible 00:21:31], their sort of master risk database was having it all in one place, that they could run those reports, they could run those analyses from a single, central reliable system. And so this is well before talk of DLT. A lot of the banks looked at that and built equivalent systems. BAML has one, JP has one, MS has, they all looked at that and said, “We want that. That is an improvement.”
And of course the opportunity with distributed ledgers is, well, why make copies at all? What if the copying mechanism is built into the protocol and then everyone has reference to a single source of truth? That fundamental concept of having that, and you tie it to settlement, that is huge for finance. This is a legitimately very big deal, and there’s lots of people who understand that. It’s not just the crazy people who left to go do crypto who get that. People internally get that, and they are building, and it does seem like there’s some momentum around it.
And so I almost see it going on two tracks. There’s what people are doing to renovate systems and leveraging the technology. And that may be private blockchains, that may be public, it may be hybrids, and then you have our clients are interested in trading crypto and thus we need to renovate systems to be able to handle these new assets as well. So we generally look more at the latter in Cloudwallet. It’s like how do you apply and get those same functions that an institutional asset manager or trader needs on this novel set of assets?
Ian:
Well, that’s incredibly encouraging to hear, that that momentum exists inside those organizations because it’s not an overnight change to any of the complex technology that runs big financial institutions. It’ll take years, if not decades, to fully plow into the architecture. But I think it’s a great segue. Let’s talk a little bit about what you’re building at Cloudwall. Maybe sketch the big picture for folks and then we can zoom into the details.
Kyle:
Sure. So my co-founder is Jia Yng Wee, and so we started in the fall of 2021 and the first stage of the startup was really kind of looking at the opportunity space. And so we looked at our background. She’s from capital markets operations. She was head of FX Asia operations and based in Shanghai when my family was there. So that was the personal connection. And when we were doing that look across the space, it’s like data, regulatory risk, KYC AML type problems, custody, execution. So my background was e-trading technology, and we looked at, it’s like actually there’s a lot of people doing pretty good things for algorithmic trading and smart order routers and things like that.
And just kind of running through the list of the institutional stack. And then it’s like, okay, and so who’s doing risk? Why is nobody doing risk? It is like this is fundamental to finance. Why is there no borrow or risk metrics or Moody’s Analytics? Where are all of these risk vendors from traditional finance in crypto?
And people asked me when we were fundraising, because of course when you come up to somebody and say, “There’s this complete area of white space, we’re going to build a company,” the first question they’ll ask you is, “Is it because you’re the only one stupid enough to go after it, that this white space exists, or have you actually identified a legitimate opportunity?”
And what we tell people is you have to look at the difference between a day trader and an institutional trader. Day traders think in terms of price. Institutional traders think in terms of risk. Who were making the running in the early days of Bitcoin and crypto overall? It was the day traders. This is just not the model that’s used for reasoning about investments.
So our thesis really was, well, the mix is changing in terms of people coming into crypto. It’s going to be people who think about risk first, not about price, who are coming into this space, and this gap will be filled sooner or later. So we pitched to build a risk management platform for digital assets, and that became the Serenity platform, which we spent over a year and a half building, currently out in production. The focus of the current platform is listed instruments, so CeFi, not DeFi, spot and derivatives. So spot futures, perpetuals, and options, with a focus on market risk.
But our view is that actually a portfolio manager in the space ultimately is going to have to have an integrated view of multiple dimensions of risk, that they’re going to have to be looking at their market risk, their liquidity risk, their operational/smart contract risk, their credit risk. And so there’s actually a lot that needs to be built to understand the risk of a crypto portfolio.
We also took the view very early on that tokenization was over a decade timeframe going to be a massive trend in asset management. So we always framed Serenity as a digital asset risk management system. It’s not a crypto risk management system. We have decided to enter the market with crypto hedge funds and crypto asset managers, but the system was designed and all of its data models do not assume that the underlier is crypto. The underlier might be a bond or a money market fund or an equity or a piece of real estate. That was very important from the early days, to recognize that for this to really work, you actually need to handle anything tokenized, not just crypto, but crypto is the big problem right now for the people who actually have risk on their books. And so that’s where we began.
Ian:
This is amazing because I think the big theme of 2022, as we saw multiple firms collapse, starting with the Terra Luna ecosystem and running all the way through until FTX and Alameda shut down at the end of the year, was nobody’s looking at counterparty risk. There was no visibility of the kind of intertwined nature of loans and leverage across the ecosystem. And I think it was then confounded a little bit by the fact that you had a lot of naive retail investors who didn’t actually understand where the return that they were getting was coming from, like what the nature of the underlying investments that some of the firms were making.
But I’m curious, had your firm maybe started back in 2019 pre-pandemic when you had the idea, could this have potentially averted some of those crises or at least made better informed decisions for the people that were honest operators? Because there was a layer of fraud happening here as well that I think maybe is out of scope.
Kyle:
We talked about this a lot, and I at one point made the point, one of the nice things about our risk tools is you can run them back in time. So I showed people what an FTX linked portfolio looked like a week before the CZ tweet and a week after the CZ tweet to make the point that risk systems do not catch fraud. There was zero. Nothing showed in there.
There were a couple of people who found interesting things on chain around the time of the CoinDesk report. That was probably the only way to really see that coming, unless you had insights, operational insights that you’d done proper due diligence on FTX, which some had done. And unfortunately Taylor Swift did better due diligence than some other people.
Ian:
Apparently so, yeah.
Kyle:
But some people did ask questions and get answers that they were not comfortable with, but a lot of this just, it wasn’t visible. And to the point earlier about the OTC market, a lot of those loans were bilateral. There was a lot of exposure with OTC derivatives as well, on the option side, that people could not see. And there was a true climate of fear post Terra Luna, Celsius, et cetera, about counterparty risk.
And we actually did a survey with acuity in the spring of ’23 and asked people to rank risks, and counterparty risk came up number one. That was the one where people were putting a lot of risk investment right after FTX, was trying to get a handle on it. And the issue with counterparty risk, going back to that earlier point and why I have that side interest in selective disclosure of information, is it’s a disclosure problem rather than a modeling problem. I know people got super excited about, oh, don’t worry, you can put all of your crypto on our exchange because we have proof of reserves.
And it’s like, okay, great, yes, you know that it’s there. There’s this general idea, there’s a difference between you have assets and you’re solvent. And it’s sort of like, hey, I will show you half of my balance sheet. How do you feel about your counterparty risk? It’s like, it’s not good enough. You actually need to represent financials on chain. You need to have secure ways of auditing those, putting attestations on chain. There’s a lot that just hasn’t been built to do that. And so as a risk vendor, I don’t want to overpromise. If we have the data and we have the models, we have the people who can build systems that will solve that. But the disclosure wasn’t there in 2019, 2020, 2021.
We could have been just as hammered out of all of that, having built Serenity earlier. Would you have gotten a sense through things like value at risk and stress testing and all the other tools we’ve built of what is the worst that could happen? Yeah, that would’ve helped. Maybe people would’ve positioned themselves a little bit better?
And just one last thing, I feel I have to restore the honor of those retail investors because I’m sorry, the guys at 3AC were not like these sort of, oh, we’ve never heard of finance before guys, and they were doing spectacularly stupid things with leverage. The institutional players were making some really poor decisions, particularly around leverage and counterparty risk there. But to be fair, for a lot of the players, a lot of the people we’re talking to right now, they just didn’t have the information. They didn’t know what was going on under the hood. But the rounds of amplification of leverage on staked ETH and things like that that those guys were doing with the Genesis trading desk, this was almost due to explode in their faces.
And that I would say is more about risk culture than risk tools. I think for people who’ve been in those spaces and have been through a couple of cycles, those scars usually lead them to be a little bit more careful. And in fairness to the 3AC guys, if you look at their professional histories, they started on Wall Street right after the great financial crisis. It always went up into the right for those guys. They never actually saw bad times until it was spectacularly bad. And that’s the importance of having history and experience in the space.
Ian:
I think that’s such a widespread problem. I mean, the industry has attracted a younger generation that just hasn’t seen that broadly. Obviously I’m sure we could find exceptions. But I’m curious. So okay, let’s take counterparty risk out of it. What are the things that if you were describing Serenity to a potential client, what does it do incredibly well? What am I getting by adopting the platform?
Kyle:
So there’s a couple of things. So our focus so far has been on market risk and the strength is definitely quantitative risk models that are sort of backward looking. So I would say we have the best implementation of those sorts of models out there right now, and we offer the ability to run multiple models and get different views on it. Some of the classic tools that like an equities trader would just expect to have in terms of being able to run a factor risk analysis on a Delta One portfolio, the ability to price and run risk slides on options and option strategies. Probably most importantly for crypto, the ability to stress test. One of the things that Serenity can do is you can replay the 2020 COVID crash on your portfolio today and actually get an attribution of the shock P&L to basically see where did the losses come from in the portfolio if that were to happen again tomorrow.
This is a space that is rife with hundred year events and black swans and all that. So a lot of people who do think very seriously about risk management in the space, they think in terms of scenario analysis, stress testing, which post-financial crisis, there’s a lot more focus on the traditional side on those methods as well. So we look a lot at what happens under the extremes, whether it’s with conditional VAR analysis or stress testing.
What we probably do less well is looking ahead into the future, more predictive risk. This is something that we’re looking at. This is where I think on chain really comes into its own. You really need that information to be more predictive in this space, and that’s something that in the future we’ll definitely be incorporating into the models, but we wanted to get solid traditional historical risk models in place and build all the infrastructure around it before we went out to the things that are truly novel in terms of risk models.
Ian:
One of the things that you and I talked about when we first met was my perception that most assets and markets in crypto have some amount of, I would say, manipulation that don’t exist maybe in the traditional equities market sense, just because there’s not the rules and the regulations and the oversight and the compliance. Trade surveillance is just really not a thing. I mean, there’s a couple of companies trying to do it obviously, but it’s not widely adopted. There’s nobody who’s looking or expecting trade surveillance, so companies who are attempting to try and use that technology or doing it of their own volition.
So does that quantitative modeling and back testing and stress testing, do those tactics still work in crypto? When I think about something like Dogecoin that is so sentiment driven, like an Elon Musk tweet is the biggest market moving activity, does it still apply? And you had a fantastic answer when we talked about this before. I’d love for you to dive into this here.
Kyle:
So specifically on Dogecoin, there’s been some academic research into risk factors already in the last couple of years, and one of them was doing analysis across crypto market that incorporated Doge, and they ended up saying, well, one of the factors in the model, we call it the Doge factor, because the only thing that explains Doge is Doge. Sorry. It is unique and it has elements of sentiment and crowd behavior and all of that that just don’t apply to the rest of the universe. And models often have to correct for that. If you have these truly idiosyncratic instruments, you want to make sure they’re not steering the model in the wrong direction.
So I do believe that eventually we will need, in the same way that we’ll need submodels for stable coins, which have some unique risk attributes, we probably should have risk models for [inaudible 00:38:33] coins that incorporate on chain data and sentiment data and social media and all of that. I actually don’t think it’s a joke. I actually think that there is a different financial behavior there, which whether you agree with it or not can potentially be modeled and understood. And so if you’re playing in that space, you probably do want to have a submodel that handles that use case.
The first point I’d like to challenge just a little bit more, because this is a common answer that I get when I talk to people. I say, “Hey, we’ve got a risk system.” And sometimes the answer is, “Well, everything in crypto is event driven. There’s no conclusions that you can draw from history. It is all, it springs from the forehead of CZ and a crash happens.” And this is not true. It is not true at all.
And there is at this point enough data, particularly if you’re looking at high resolution data, that you can actually use some of these models. There are some types of tokens they won’t work as well for. There are some that they’ll work better for. But just saying, well, there’s no point in doing this because there’s no information, I think at this point there’s enough evidence from academic research, from things that we’ve done, that there are things that you can do to better understand what could happen at the extremes.
It probably still will not perfectly capture it. And yes, there are events, there are manipulation in the space. I mean, the most recent Bitcoin flash crash, we were doing a deep dive over the weekend, actually, combination on chain and looking at some other things and just conversations with people within the industry, trying to piece apart what happened, what was the triggering event. People I think had a good explanation for what happened after the trigger, but a lot of doubts and speculation about what actually set it off. But you are going to have that.
But the reality is it’s not traditional finance is entirely without manipulation. I mean, LIBOR was not that long ago. And so the monitoring tools and the surveillance are maybe not as robust, and you have offshore exchanges that are kind of out of reach that are potentially sending price signals whether you’re on them or not.
So the trigger of the collapse appears to have been on OKX last week, but when you get these cascading liquidations, people just start seeing that happen on one exchange, they start doing it on other exchanges. So even if you have a well-behaved exchange, if you’ve got a big enough one that isn’t putting those protections and surveillance in place, it could still send price signals elsewhere in the market that people respond to. And so I think that’s going to be a problem for a while. But this is a really good example of there are tools that have been built, there’s history around what is needed. There’s history around good regulatory frameworks for it.
And it is important, and I think people often undervalue this one, the degree to which having trustworthy benchmark prices is important for big financial products. The reason we don’t have a spot ETF is fundamentally about trade surveillance. It’s not about Gary Gensler hating crypto. And so if that gets cracked and BlackRock has an interesting proposal for cracking that particular issue, that might open things up.
But regulators really care about people being able to trigger a [inaudible 00:42:23] contract by moving things a little bit. And so they’re a little uncomfortable with the fact that you can bump an oracle hard enough and you can suddenly get all of these wonderful consequences that are profitable. And so we definitely need more work there, but we at least know what needs to be built. It’s not something coming out of nowhere.
Ian:
Yeah, I’m really curious about, you’re almost two years in since launch, I think, and it’s obviously been a turbulent kind of market experience during that period, but I’m curious, what’s been the response from the customer side as they start to see your technology? Is it actually sparking more institutional adoption? I would have to say people probably look at Serenity and they’re like, oh, thank goodness this exists.
Kyle:
I think it’s a little bit early for that still. The fairly universal reaction has been this is really needed and I haven’t seen anything else like this, so that’s a good sign. But the asset managers are recovering as well. And you have the larger ones who are maybe more resourced who would go into it, but they’re looking at what’s been happening in the US on the regulatory side, particularly in the spring of ’23, and saying, well, maybe not this year. So they get it, but they’re not there yet.
The smaller guys were very badly damaged last year, and so whatever the interest may be a little bit slow coming into it, but we’ve seen a remarkable number of emerging managers come into the space post Terra Luna, post FTX. So the fund formation rate is actually quite healthy. People are coming in and they’re setting up, and the profiles that I’m seeing are people with pretty amazing traditional hedge fund backgrounds who’ve decided, hey, I’m going to leave Greenwich and I’m going to start a crypto fund. It’s a very different profile from the early days of crypto asset management.
So the challenge that they’re facing where we think we can help perhaps a bit over the longer term is getting assets under management. And that means going to more risk averse investors who are doing robust due diligence of the fund. And one of the questions they’re asking is, what is your risk management strategy? What tools are you using? What independent verification do you have of your strategies? And so really what we’re putting in front of potential clients is this is your answer. This is the thing that you can go to a more traditional family office that’s more risk averse and say like, look, we are coming with all this experience from, I don’t know, commodities or equities or whatever, TMT investment in Greenwich, and we’re bringing in institutional grade tools, and so we can reassure you that we have a handle on the risk here and thus we can take that larger check to build up AUMs.
So we are really hoping to help here. We want that segment. The industry is massively undersized at the moment relative to the interest. And I do think part of why it’s undersized is the people who could be allocating, who might have the capacity to do it, are reluctant due to the risk concerns, and we hope that we can address that and really bring more patient capital into the space so it can grow.
Ian:
I think there’s a lot of people cheering for that outcome, more capital into the space. You mentioned, we talked a little bit about DeFi earlier, and you talk about that is one of the big open questions from a regulatory perspective, even in the EU where MiCA is kind of driving the forefront of I think reasonable crypto regulation. They punted on DeFi as a topic until the next iteration of that framework. What’s your perspective? Where should DeFi regulation go?
Kyle:
It’s a really timely question with Tornado Cash, right?
Ian:
Yeah.
Kyle:
But there were, including from some multilateral organizations, some thinking from the regulators about how to handle this issue, which really boils down to when is coding a crime. And there’s a lot of people, including me, who are really uncomfortable with the idea that having written something and putting it on GitHub, that this brings in a lot of regulatory accountability.
But once you get to a point where you’re operating a platform and supporting it and there’s a real organization around it and you’re profiting from it, which is essentially the case in Tornado Cash that they’re trying to make, I do believe it’s a spectrum. And there comes a point where, yeah, it’s not as decentralized as you think. There truly are accountable people who you could ask to control these things. But I think the problem and the element of fear in the industry is it’s really unclear where that line lies.
And I’m a huge fan of some of the frameworks that have been put forward, like from Hester Pierce, where there’s this idea of almost a gradation where at a certain point in the early days of the project, you have a particular sandbox set up. Singapore has done this really well as well, the UK too. It’s a nice model of basically saying if something is small enough that it cannot do systemic damage and the relative impact within the market is not that big, we should allow a space for experiments to be conducted. But we should have a transition where when it gets to a certain size, a heavier weight set of rules apply.
And one of the struggles we’re facing, I would say more generally, not just around DeFi in crypto right now, is the one size fits all problem. Trying to apply securities law to something that has some securities like elements that definitely could prompt attention on the regulatory side, but has lots of other elements that are not securities-like and that doesn’t necessarily have a structure that’s conducive to doing the necessary disclosures and filings and trading on the regulated exchanges, et cetera. There has to be some middle ground.
And so I like that approach of dialing up the oversight based on the level of risk. I think there’s a good principle for risk management and risk mitigation, that it really should be proportional to the damage that could be done, rather than one size fits all. Now the problem with DeFi is there people are playing around on the KYC AML solutions with whitelisting, et cetera, I think there’s some really interesting and potentially more scalable ideas that could help solve that.
But talking to people in the market, that’s the one that stops them. It’s like, I’ve got a commercial bank account at Chase and they’re going to shut it down if they think I’m [inaudible 00:49:55] funds with North Korea, and if I cannot give a solid answer about who else is in that pool, they’re just going to assume that I’m doing something wrong. And they do have a responsibility. Like if those funds are transferred through the fund admin and they end up in a pool, someone has to be held to account if actually that is used in violation of some of these laws.
I’m not a big fan of the ways people have been doing it so far. I also don’t like the fact that a lot of these solutions tend to be one-offs. But I do think we will come to a point where there will be a broadly recognized token that sits in people’s wallets that contracts can recognize and say, oh, this person has been KYCed and they were signed off by this person. Thus, they’re allowed in. I don’t think whitelisting scales, I don’t think issuing your own ticket that you passed KYC AML scales. You really need the equivalent of what Vitalik calls a soulbound token for an organization that recognizes that.
And there’s a lot of interesting technical work going on right now that helps make that not totally break privacy concerns. I’m a big fan of healthy balances, not all the way to just totally ban it, but the free for all probably is not workable at scale either. So what can we do in the middle? And I think when you see that, you will see a lot more institutional interest in DeFi.
My personal take is that 10 years from now, it is going to be the dominant mode. You’re going to see a lot of peer-to-peer across the buy side through DeFi that never transits through a prime broker or a sell side institution. That is the true Napster moment for finance, once that starts happening. And that is potentially very appealing to a lot of people in the market for that to happen if it allows for greater efficiency, greater transparency, et cetera. So the tech isn’t quite there yet. The standards aren’t there yet. But it will happen.
Ian:
I think you’re hitting on the key point that comes up frequently on this podcast, which is this balance between anonymity and privacy. Because blockchain as first conceived with Bitcoin was a hundred percent transparent, we said, oh, well, privacy is like a derivative of the fact that it’s anonymous. And obviously through technology like Chainalysis and others, we’ve shown that the anonymity is not really there, but nobody actually wants anonymity. I mean, not legitimate actors. Criminals clearly do, but institutional participants, they want privacy. And I think retail also wants privacy.
But the anonymity I think is where we run into trouble here. And so we’re now having to rebuild that layer on top of the fully open and transparent protocol that gives us a reasonable amount of privacy that allows both my personal information to be protected, but an institutional trading strategy to be protected as well.
It’s great to say, oh, I’m legitimate and I’m trustworthy and I’m not North Korean hackers. Here’s my soulbound token proving it, even if you don’t know my actual identity. But now that token’s there and I can see your wallet, I’ve now unearthed potentially your entire trading strategy as well, which is probably undesirable. So I think there’s two steps there which are necessary to advance the pace.
Well, I want to wrap with, last question, what should guests who are interested in the tech and curious about what you’re building, what’s coming next? Where do you see this going for the company? I think you’ve got a big release coming up here in a couple of weeks.
Kyle:
So we aim to be super agile. This release is going to be, I believe, release number 27 since our initial launch, July 1st, 2022. So it’s constantly changing. There’s always new stuff coming for Serenity. So right after Labor Day, we’re adding Arbitrum and Optimism support. And the biggest thing that we’re adding, and both of these are driven by client requests, we try to be very, very responsive to that, is a whole portfolio analytics suite. So the ability to take any crypto portfolio and get your standard performance statistics, what they sometimes call a tear sheet, summary of how that portfolio has performed relative to benchmarks, as well as doing some more advanced performance attribution within the portfolio. This is something that we’re going to be adding to substantially in the fall.
But yeah, and lots of other small improvements on the performance side, the usability side, we have a number of people who spent many years working with institutional traders. We’re quite passionate about usability, so sometimes the improvements, they’re not really big new features, they’re just things that make it a little bit easier, a little bit faster to do that job. And so we have some things coming along those lines as well.
And then a lot more ahead. I mean, the vision is for really a multi-year project to get to where we think it needs to be, but trying to get there an iteration at a time, just constantly improving it, because it’s needed now. So we’re trying to get the best available with current technology out there as quickly as we can.
Ian:
Exciting. I’m certainly going to be watching along. Kyle, this conversation was terrific. Thanks so much for joining us.
Kyle:
All right, thank you for having me on the show.