10 October 2024
Ian Simpson | 36 min read
He used to be a rocket scientist in South Africa, but now James Kilroe and his co-founder Geoffrey van Ryneveld run Tendex, a market neutral crypto hedge fund with a focus on medium- and high-frequency trading strategies.
Spot On host Ian Simpson and James discussed the background of Tendex’s activities in niche crypto markets and its transition to more sophisticated strategies. Along the way, James explained what he thinks of the “financialization” of crypto, the weaknesses with current market infrastructure, market making strategies from TradFi and crypto and how they overlap.
Episode show notes:
He used to be a rocket scientist in South Africa, but now James Kilroe and his co-founder Geoffrey van Ryneveld run Tendex, a market neutral crypto hedge fund with a focus on medium- and high-frequency trading strategies. Spot On host Ian Simpson and James discussed the background of Tendex’s activities in niche crypto markets and its transition to more sophisticated strategies. Along the way, James explained what he thinks of the “financialization” of crypto, the weaknesses with current market infrastructure, market making strategies from TradFi and crypto and how they overlap.
(1:05) Intro to a rocket scientist crypto hedge fund manager
(3:32) The “financialization” of the crypto market
(6:39) The elements coming together for market maturity
(7:18) The backstory of Tendex
(8:12) Early arbitrage opportunities
(8:55) The simple beauty of the crypto market in the early days
(10:38) The missing puzzle pieces at the beginning
(11:50) From Python script to advance trading venues
(12:16) Home-built tech is still the best
(14:07) Getting around the fragmentation of unstandardized tech
(15:35) Challenges and opportunities of being a market neutral fund in volatile crypto markets
(17:22) Overcoming the “flat and the crab” with unique market-making trading strategies
(18:19) How Tendex turns tens of millions 10x per day
(19:05) Plugging in “slow” strategies to the cutting-edge tech stack
(19:40) The decision to go after the biggest, most liquid markets
(20:22) The leveling of the market and “time-based” strategies
(21:18) Where is the “basis trade” going?
(22:18) Why faster venues like RULEMATCH will make a difference
(22:28) Why a price discovery venue is needed in Europe
(23:18) Cloud-based tech vs bare-metal infra
(24:15) The curious thing about latency in crypto trading
(25:26) The need for deterministic venues
(26:00) What happens when “flows” flow off AWS
(26:55) How risk management gets forgotten
(27:45) Two lessons from FTX
(30:17) How FTX highlighted exchanges playing “bank”
(32:02) Handling data and the tech
(35:32) The crossover between TradFi and crypto market making approaches
(37:44) Trying to determine true price in crypto
(38:22) A TradFi strategy that isn’t “here yet”
(40:31) Post-trade settlement enhancing the role of a prime broker
(41:55) How the PB market is changing
(42:15) The tricky thing about custodians
(43:36) The diverse array of custodial service providers for institutions
(46:59) Prime brokers face off against exchanges
(49:24) How Tendex is set up with crypto service providers
(52:11) Sitting at the heart of Europe
(52:25) Understanding crypto investor sentiment and attitudes
(55:35) Looking out for stablecoins and digital identity
Listen and watch:
Follow on social media:
Episode transcript:
Ian:
And so you’re very focused with what you do. Also, if I can just ask you straight out, not a huge fund, right, as far as AUM?
But trading very, very actively.
Can you put some numbers on that?
James:
Yeah, sure. So we trade in the hundreds of millions a day. The funds is in the tens of millions.
So it has a very active role. It’s a very busy fund, shall we say.
It’s obviously completely automated. You can’t put that number through with sort of…over a UI that wouldn’t really be feasible.
That gives us a lot of advantages that we think that some of those more high-chains strategies can level out the P&L, increase the sharp ratio if you want to get into some more technical stuff.
And it’s forced our technology stack to be cutting edge.
Ian:
Hello and welcome to RULEMATCH Spot On, the only podcast focused exclusively on the institutional crypto and digital assets industry.
I’m your host, Ian Simpson. And if you want to know what banks, hedge funds, asset managers, trading firms and others are doing in crypto and how they’re doing it, this is the place for you.
My guest today is James Kilroe, director and co-founder at Tendex, an algorithmic trading company based here in Zug with a market neutral crypto fund.
Fun fact, James got his degree from the University of Cape Town in mechanical engineering, where he also studied “space studies” with a master’s that focused on a device to deorbit satellites.
James, welcome to RULEMATCH Spot On.
James:
Thanks for having me.
Ian:
Is it fair to say – just to start out – with that we could call you a rocket scientist?
James:
I think my team might disagree.
No, yeah, it was a very interesting master’s I did in the University of Cape Town, looking at, I guess, satellite technology, designing a part for a small satellite to deorbit it.
It was a very interesting time, quite cutting edge, and always fascinating.
I think I’ve got a huge fascination in that area. And then I did another master’s at the University of Cambridge at the business school in technology policy looking at regulating new technologies.
Uber was very topical, but so was blockchain, which was funny. That’s sort where I fell into with my career and have been involved in this space since 2016.
It’s been a very, very interesting journey so far and lots of up and downs from token economics, ICOs. In the early days, I worked at a venture capital firm for a couple of the guys who were
in the Bitcoin Foundation and one of them was known as the Bitcoin Oracle at the time.
Maybe that is still his name, so his nickname at least on Twitter.
So it was a very interesting time and that was when there was only one or two or handful of digital assets.
Now there are obviously tens of thousands. And so it’s been a very, very interesting journey in the space and I guess some of that history gives me bit of knowledge and helps me in my day to day with Tendex.
Ian:
Super cool.
Thanks for coming on the podcast.
James:
Thank you. Thanks for having me. Excited for the conversation.
Ian:
You know, we’re going to be getting to many serious topics, diving down into the weeds of what you do with Tendex – with the trading, the setup of the fund and all of those things that you do around that with the tech.
But just kind of high-level..
Crypto funds now have become more and more of “a thing,” developing in different ways with different strategies.
At the same time, we have in the US ETFs now a big, big milestone for crypto – ETPs in Europe – for quite some time.
Just from your perspective with the fund and also being in the space, where do you think we are with the financialization of crypto?
In other words, crypto assets as underlyings in financial products?
James:
Yep. I think it’s actually quite interesting because I’ve been involved in the space since 2017 where there was, well, actually almost 2016 and sort of one of the big topics then was Ethereum had just launched.
And so it was Bitcoin, Litecoin, I think maybe Ripple and Ethereum.
So it was those four and there wasn’t this imagination of the thousands of products we have now.
The ERC20 standard that was launched on top of Ethereum was really the beginning of the “opening up” the ability to, you know, the ICO boom and all of these mechanisms which investors sort of grappled with and understood, and a lot of money was thrown at it.
A lot of money was made, a lot of money was lost.
But I think that happened in isolation where the rest of the world wasn’t paying attention, or maybe it sort of hit the front page of maybe the back page of the FT, I would say.
You know, not nearly front page stuff at that stage.
And that was maybe a bit ridiculed and laughed at by more sort of serious investors.
If we look, you know, seven years later, it’s completely changed.
And I think that the traditional ecosystem, or maybe TradFi, is entering in and undergoing a very similar process, maybe on an accelerated speed.
For example, it sort of makes complete sense to me that the first asset that had a spot ETF in the US has been Bitcoin.
It’s the one that’s been around the longest one that people can understand. The next one was Ethereum, right?
That kind of mirrors what happened 10 years ago. And I think that that will keep on happening at least for the top 10 or maybe 20 assets.
When you start to get into much longer tail of assets, liquidity problems, structures, etc it becomes of an issue.
But I definitely think that this financialization has only really just begun in terms of… If you just consider how few assets are open to a regular US retail investor through their retirement fund.
Coinbase has obviously offered them more earlier, but I think now is the time when the institutions are only beginning to look at the space in more earnest.
Obviously there’s maturity in the market, right?
Liquidity and all of those things for those top assets. Exactly. And I think that was required, right?
And I think regulation is a huge one as well because without the SEC and a lot of maybe the crypto ecosystem.
Bitcoiners are very anti the SEC chairman and maybe have some strong views around what the SEC’s done in the space.
I think it’s…without the SEC approval, no one who’s more serious in the US or institutional in the US can have a look at it even really if they wanted to, right?
It’s a huge endeavor for them to get access to this space. That’s changed.
Ian:
You’ve come along with Tendex and developed now with the fund and with other things.
Just kind of walk us through the process, how you got to where you are today.
James:
Sure. Yeah.
So in 2016, I started an internship at a venture capital firm that was in the blockchain space technology policy background, which was sort of looking at more regulations.
I thought it was very interesting and got going.
That was the beginning of a bit of a journey. I did some ICOs for some portfolio companies.
Some of the partners are still very involved in the space. It was a growing venture capital firm.
It was very exciting, that whole 2017, let’s say, cycle. In that cycle, I realized that these tokens were trading incredibly inefficiently, that there was a huge opportunities.
I frankly… I was at the time based in South Africa, and I am South African, I was based in South Africa.
There was arbitrages that appeared inter-country, there were arbitrages that appeared across venues.
There was really just the beginning of the tokenization in terms of more assets were being launched.
That 2017 was really that period where we saw more and more assets.
And so then myself and my partner Jeffrey, who is an old school friend of mine (university-friend as well, we’ve known each other for a long time).
He was trading commodities at Glencore. I was maybe more in the weeds in the blockchain stuff and we got chatting about how could we do this?
And we got going in sort of late 2018 with just some development of some trading strategies, very, very simple code.
That was really quite the beauty with the space because these exchanges were API enabled, so they allowed maybe much smaller players to get involved initially.
That had some good success.
Then we went and we raised capital from investors through a fund structure that was started, probably started about June 2019.
We could show them some of the results and grew and then in Jan 2020 we launched the fund that was Tendex with the aim of being market neutral.
I think it was our first fund.
We were very open about that with investors and I think what was important is that they could understand some of these strategies.
At the time, if I’ll be frank, a lot of them were very skeptical of the asset class.
They maybe understood it was an asset class, but they kind of maybe some of them believed and maybe some of them still do believe that Bitcoin was going to zero.
I think the COVID shock and maybe the money printing associated with COVID after that changed some of those views, but that was very much the dominant view.
And that allowed us to get going and drive a bit more quantitative strategies and try and be the smart, more sort of maybe cool-headed money in the room as you go through these cycles.
And yeah, that was over four and a half years ago.
Lots happened since. We’ve grown the team. We’ve refined our strategies. We’ve really nailed down what we want to do.
And it’s been a big journey and lots of learnings, but a lot of fun.
Ian:
And when we talk about building it, we’ve had some others on the podcast talk about “in the basement.”
You said you and Jeffrey got started just kind of building some things yourself, so basic to start with and then grow in from there.
What maybe was missing, or what were the things that you had to put together in order to get started and launch with the strategies that you started with and then developed?
James:
Yeah.
Really, if you wanted to get going, you needed to code everything from the API level up, understanding what the asset classes were, how different …
Because at that stage, it’s all different venues, and they still do have very different technical interfaces, how all of that works, how to safely send assets between exchanges, what that meant, what a custodian would look like, how that fits in.
All of those pieces were missing down to access to bank accounts as an account.
There was a long journey of being able to put all of those together.
I think we were lucky in the sense that we could keep it simple and simple strategies were sort of sophisticated enough to make money at the time. Those strategies wouldn’t work now.
That in itself has raised the barrier to entry because if you wanted to launch now, you would need more capital and a couple of years of development.
I would think to launch at the cutting edge. Of course, there are more venues coming in that require latency then, could have been sort of a second….
A Python script could have almost done some of the trading that you needed to do now. We talk about micros. Not even a milli down to a micro.
It’s changed drastically, and that puts huge burden on the teams that keeps teams driving forward in the ability to need a certain speed to trade.
Ian:
We’ll talk about some of the strategies maybe a little bit later and some of the tech and so on.
Yeah, speed has become quite a thing. You built all this tech and now you also sell it to others or you make it available to others?
James:
So we work with a few very exclusive partners and we assist them with some of their needs.
It’s not a retail offering in the sense that there are very big companies in Switzerland, a few come to mind, that do this professionally.
That requires a whole different suite. We’ve been lucky enough that the technology suite we can refine to our trading needs. We’re on version three of it now.
And being in version three, we have both built a generalize enough that we can adjust as new strategies come into the system, but also have made sure that it works incredibly well.
We’ve put a significant number of orders through it every day, every minute.
And that’s been really, really interesting as well. So we’re very happy with it.
And we did toy around an idea, should we sort of outsource this, but there are other well-capitalized players in the market that provide this.
To be honest, the people that want that sort of system are OTC desks or other providers that you would require quite a few other pieces to do it.
We think it gives us bit of an edge on tech, and so we’re holding onto it.
Ian:
…for yourself.
James:
Exactly.
Ian:
You talked about at the beginning with the different exchanges, shall we call them. We have maybe a different view on that word, but anyway.
All of those different standards and the different things that you have to build kind of custom here and there.
I imagine that was quite a challenge at the beginning to connect and do all of these different strategies when nothing is standardized anywhere.
Nothing is really built at the beginning for institutional kind of strategies, right?
James:
Exactly.
And I think a lot of the learning in the firm, if you call it sort of the institutional knowledge or the firm’s knowledge, is understanding the patterns that are required so that we can spin up “venues” or whatever word we want to use, into our system, being able to take the product that the venue offers, whether it’s spot or a future of whatever form or an option.
And then break that up into well-recognized, understood risk categories, how we ingest it into the system.
All of that needs to be standardized. that was a very time-consuming thing to design.
The nice part now, at least the way we believe it, that it’s generalized enough that when …
The one thing about crypto is you can always be sure that another thing is going to come along.
Even the biggest instrument, which is the perpetual future, is not something that is TradFi. It’s a novel instrument relative to TradFi.
We’ve built it in a way that it’s adaptable and changeable as new interesting products come along.
Ian:
Let’s talk just a little bit about the fund you mentioned, the market neutral fund.
With volatility in crypto, this is quite an interesting thing, as you alluded to before. There’s maybe challenges with being a market neutral player or opportunities.
Just kind of walk us through how you see that in crypto versus maybe TradFi as well.
James:
That term can encompass a lot of different funds. And I think it’s quite difficult for investors to work out exactly what type of fund it is and then speak to the manager to understand.
I think something like 2022 showed us where yields began to dry up and some funds, understandably, but maybe increased their risk profile, started to take risk that maybe wasn’t necessarily in the spirit of what investors understood.
The way we understand it is it’s a fund for all markets.
We limit our market exposure, our “deltas” quite significantly, while within the parameters that the fund and we share with investors.
And that allows us to, on any particular asset, that allows us to generate returns in all environments.
Of course, typically volatility and a bull market brings a few things.
So, you know, the sort of old analogy of up on stairs and down on a ladder.
There’s a lot more volatility and a lot more sort of movement in a bull market.
There’s obviously more trading volume as well. Maybe they’re also less discerning fundamental investors that are willing to buy at any price.
You know, as a market neutral, you can help facilitate that with them as providing liquidity.
In a bear market typically, it will move quickly in those days, can be very profitable, but then it can be flat for a while afterwards.
The flat and the crab can be a little bit more challenging.
We’ve spent a lot of time understanding the market: our strategies, of micro market strategy, the broader market strategy.
We think we can offer investors quite a unique proposition because I think more…well, there are many different ways of doing market neutral.
We’ve lent more into the mid-frequency, some of the more high-frequency, almost market-making stuff.
It’s actually quite difficult to get access to those strategies in traditional finance.
Traditional finance, market makers, and increasingly this space dominated by other big investors and other big shops, you’ve had some on [the podcast].
And you can’t necessarily access those as an investor.
It’s nice to be able to offer that sort of strategy to interested parties.
Ian:
And so you’re very focused with what you do. Also, if I can just ask you straight out, not a huge fund, right, as far as AUM?
But trading very, very actively.
Can you put some numbers on that?
James:
Yeah, sure. So we trade in the hundreds of millions a day. The funds is in the tens of millions.
So it has a very active role. It’s a very busy fund, shall we say.
It’s obviously completely automated. You can’t put that number through with sort of…over a UI that wouldn’t really be feasible.
That gives us a lot of advantages that we think that some of those more high-chains strategies can level out the P&L, increase the sharp ratio if you want to get into some more technical stuff.
And it’s forced our technology stack to be cutting edge.
If we want to and we can and we sometimes do bring in more slower mid-frequency or sort low-frequency strategies.
Those can plug into that same strategy set very, very easily.
We took a view in 2020 and 2021, we were training much more niche markets.
We had some great success, but in sort of April 22, there was a big market reset.
Volumes went way down. And in about June 22, kind of, Jeff and I had a discussion.
Do we take the money? We return the money we’ve made for investors, return the fund. We’ve done very well.
Or do we take all the profits that we’ve earned from that time and build out a team and really go after the biggest markets in the most liquid instruments in the space?
And we chose to do the latter.
So it was a bit of a building period as we desperately and as quickly and efficiently as possible try to retool our systems.
We did that in ‘22 up until the end of last year.
And we launched that in December ‘23 right into the beginning…I’m hesitant to say…I don’t necessarily think this is a bull market like everyone else’s.
This strikes me a lot more like 2019, but that’s been very successful.
That cutting edge version three of the system, we think it’s got a significant durability just because of the flexibility, the flexible nature of it, and the speed and reliability that it trades at.
Ian:
So always adapting…. That’s very, very interesting.
Because as you mentioned in the early days, there was quite some differences probably in speed across venues.
There was differences in liquidity, price, so on and so forth that you could take advantage of. Now things have kind of leveled out. We’ll talk about the fragmentation in the market, which is also an opportunity. I think some of the strategies you employ are time-based, in other words, futures versus spots and those kinds of things.
Derivatives have been very popular in fund strategies up until now.
Is there something that would shift things towards spot or what do you think it would take for there to be kind of a change in some of that?
James:
I think that the strategy that most, a lot of market neutral funds, right, it’s not secret, is the basis trade. That worked very well at the beginning of the year and it has its moments.
It worked very well in 2021.
I seem to think now that with sort of Wall Street level of capital and being able to access the CME and the spot, I’m not necessarily convinced it’s coming back to the same degree.
Just in the sense that the second it becomes attractive again, any fund on Wall Street can put the trade on and there’s enough liquidity there to crush that.
And so we’ve worked quite hard to make sure that that isn’t the only fund that we have in our in our system.
Obviously, if it’s available, we’ll take it. Why would we leave it out?
But we’ve worked hard to make sure that we can perform around that.
That was the spot, some of the spot versus the basis.
I think in traditional finance, if you look at maybe FX, FX spot versus FX forwards or derivatives on stocks, there is always an order of magnitude more trading on the derivative side because your margin requirement is much lighter.
What I do think is interesting though is that more venues like RULEMATCH are coming out that are very fast and will drive, I think will drive price discovery and I think that’s quite useful because I don’t think that there’s a big venue in Europe that’s currently driving that.
So if you look where the flows are, at the moment it’s in Tokyo, it’s actually in AWS region of Tokyo if you want to get specific, is where the biggest venue in the space is.
And then on the US side it’s sort of in North Virginia, which is again the AWS region of the US’s biggest provider, but there’s nothing in Europe.
I think it’s quite important and useful as you look at the three different trading sessions and time that there is a strong price discovery that goes through Europe.
Ian:
Interesting you mentioned AWS. We like them very much. We use them for some things, but not for everything. Maybe in the technical weeds of things that has been one criticism or maybe among professional traders, a criticism of venues and the tech that it’s cloud-based, it’s slow, there’s different things.
Is that something that you see and watch and try to avoid or take advantage of?
James:
Yeah. I’m always a little, know, speed, if we get down to one micro, you know, let’s just say, I don’t know if it can go much faster if we get down to one mic or maybe eight to 10 micro.
Wall Street operates in about a 20-mic range, maybe even half of that now.
You will just have the players that can operate on that range will remain in the market and everyone else will fall away.
So in a strange way, it being a bit slower isn’t an issue because everyone just recalibrates that speed.
It does introduce probably more competition because smaller shops can still compete at that range. Different strategies will work.
However, what is very different in the cloud ecosystem that’s more frustrating is the indeterministic nature of it.
So under load, might not be millis, it could be hundreds or even seconds if something gets overwhelmed or that kind of thing.
That’s a problem.
And that’s a problem for anyone because your trade, doesn’t matter how fast you are there, your trade might not place when you wanted it to place.
So if everyone knew it was 100 milliseconds or one micro, the game would be played, who could play it might change.
But it’s the indeterministic nature that causes some issues.
Having said that, some of those venues have, they obviously are looking to deal with that.
And the technical side, we spent, and that’s maybe another bit of firm knowledge we’ve picked up over the five years, which is how to optimize some of that.
We don’t actually consider our edge speed. We just know we need to be fast enough because otherwise anyone could pick it up.
So we need to be on a level playing field.
But we don’t…we don’t think of ourselves as sort of speed being the thing that that’s going to give us an edge.
What is exciting is when (funny enough) you know venues become more deterministic, which is a big element that for us is actually much more interesting. Because bigger shops are probably better at working out sort of non-deterministic/first deterministic speed and how to sort of modify their orders associatedly.
That’s just another element that we wouldn’t have to worry about.
So “deterministic speed” for us is probably more important, kind of “at what frequencies” – a little bit less so.
And yeah, I mean, I think what will be interesting to see how flow off AWS into more sort of bare metal servers, you know, such as yourselves and RULEMATCH as the space develops and maturity grows.
Ian:
So obviously the end of ‘22 was the time of FTX – when that all blew up and it seemed like you at Tendex weathered that pretty well.
But due diligence and “know your venue” was pretty important. That continues to be important for you today?
James:
Yeah, that’s my primary job, I think.
You have to make sure. I think we are getting to a stage now.
It’s been very interesting because the space, maybe I’ve got a bit more of a track record, a bit more of a memory than others having been more involved in it, but in 2016, 2017, there were a lot of fly-by-night shops.
That’s where trades were happening, and people were very concerned about risk management, and probably because a lot of the hangover of the industry was Mt. Gox.
That was forgotten in 2019, 2020.
And I can distinctly remember in 2022 thinking the yield-
Ian:
Don’t forget about that.
James:
Yeah, exactly. 100%. Exactly.
And the yields that were being offered, you could lend money to certain counterparties that have turned out to blow up at yields that we weren’t finding in the market.
The risk was getting higher and higher. And so we were gradually de-risking throughout the year.
And that was just to make sure that we cut our exposure to as many venues as we could.
And I think with FTX, the lessons there are twofold.
One is that the market, I think, was a bit naive.
I think it’s become much more skeptical.
But also the power of the blockchain maybe wasn’t identified.
People could see the assets on the balance sheet in the sense of the deposits available on FTX.
Those wallets have been mapped by open-source investigators, if you want to call them that, or open source researchers.
So everyone knew what FTX had.
What maybe people didn’t fully understand was obviously the other side of the balance sheet.
So you could see the assets, but you couldn’t see the liabilities.
But when the assets started to dwindle, you know, the space was relatively slow to understand, and maybe more importantly, not just the value of the assets, the composition of the assets.
I think that that’s become quite interesting.
I know of one venue that, and these venues have all now fought to keep, there’s been a huge drive by fund managers to say, well, off-chain custody.
There are obviously a bunch of solutions that are trying to come in. I don’t think all of them have yet quite cracked that – not across the whole industry.
But I think that that’s something that’s coming.
But a bunch of the venues have also started to open up. I know there’s one that you can log in and you can see your balance, and you can kind of cross-check and reference that onto their deposit stack.
And the idea being that if any sort of assets being double-counted, people would work it out.
So it’s a bit of a crude method, but there they’re trying to give you not just access to the liabilities, sorry, the assets, which you can see, but also being able to link some of the liabilities.
And just to try and make sure that there isn’t a significant mismatch.
Yeah, it’s very interesting. I think that the space has got a little bit too… I think that after FTX, naturally, a lot of fund administrators, a lot of fund managers promised investors that they had good custody, they had good solutions, and maybe they fell a bit short.
But then on the other hand…
And so as a result, the market’s gone a lot more, almost too over the top in terms of trying to understand risk.
That’s maybe left quite a lot of off on the table.
And I think the risk at the moment is in the space is very low.
On the other hand, it’s driven a lot of good innovation and much more sort sophisticated players and venues coming along.
And that, I think, will already be a positive in the longer term.
And it’s made it easier for the regulator.
Ian:
So one thing that the FTX scandal probably highlighted, and of course people have talked about it and people are talking about it more and more, is the “separation of duties” when it comes to an exchange, a market maker, a custodian, somebody like FTX basically playing bank.
How do you see that?
James:
I think it’s important. I think we need to play …
Well, it seems like the space is going through financial history on a sped up time frame.
A couple hundred years ago, exactly the same set up would have been the stock exchanges, then things blow up, the regulator steps in.
It seems that the space is trying to adopt some of the best practices.
It does make sense to me. I can understand why Binance, for example, or other venues want to keep their assets because for them, if they just become a matching engine some of their … They may have to compete in just being a matching engine.
To be honest, that technology stack isn’t as good as … More of a commodity. Yeah. I think that they would struggle there relative to others.
They want to keep the full stack.
I think that the other thing is that a lot of the space is more retail at the moment.
It’s very hard for other venues to onboard thousands …
Well, not other venues rather, but other parts of the market to onboard thousands and thousands of retail investors. They had a different decision set that they chose.
For us who are more institutional, we’re desperate for those sort of solutions.
And as soon as they sort of become practical, of course we’ll adopt them because all that does is lower our risk, which sort of increases our risk adjusted returns by function of the fact that the risk is lowered.
Ian:
So probably two things looking just a little bit more under the hood here that are important for you at Tendex, data and technology.
And I imagine as a very actively trading firm, you use a lot of data, you process a lot of data.
How do you deal with that? How do you process it? How are you set up even technically to deal with that amount of data?
James:
So that’s where think cloud infrastructure is really good, right? Because you can scale databases, can scale instances or computers to process if you need to.
You can run time-based jobs that periodically scrape or fetch data from whatever source.
That’s another whole part of our technical stack, which is the data ingestion side.
It needs to take it and it needs to be reliable. That’s a given.
Then we need to both store it, process it, and then understand, push it onto a trader after verifying its integrity.
That whole stack runs in the cloud. Most of our stack runs in the cloud.
In fact, all of our stack runs in the cloud. And it’s been a very interesting part.
I think now the space is getting better, because a lot of that also comes down unstandardized.
Small examples, but candles. that’s open high, low, close, over whatever period, a minute, an hour.
Some exchanges have a different understanding of whether that candle’s of the open or the closed minute, right?
So even just is that candle, that open time referring to minute one or minute two?
You know, that’s just a small example that would throw off a lot of your backtesting.
So you have to go and check all of that.
What we are increasingly seeing now is, again as you would imagine, more sophisticated data providers.
There’s a whole swath of on-chain data that we have a very limited exposure to. That’s coming in.
Of course, there’s been the sentiment sort of X or Twitter or any other sort of maybe Reddit, any other sort of sentiment area.
That’s another whole section of data that we don’t really touch.
A lot of those, and there are professional shops that are doing those.
So we try not to build our tech if we can avoid it, but sometimes it’s required.
Ian:
And so you’re not putting any AI on this data to help you analyze and…
James:
Not at this point.
I think I found with investors, it’s quite hard to explain AI.
I mean, AI is a big box. I think AI is one of the few things that’s maybe at the moment very in vogue, and so people are willing to maybe take a risk.
But think there is a lot of more simple strategies still out there and not really required.
In time, that might change, so there are some sort of basic …
It is useful for some things, but at the moment there hasn’t been a huge need for us to execute a bunch of sort black box thinking.
Also it just becomes much harder to explain to investors where … That’s something that’s very important to us.
Ian:
It can go one way or the other or a million ways all at once.
James:
Exactly. And for us to ensure that it behaves the way we expect it to behave. It’s maybe a small part of the system, but it’s a part that could … Hey, why …you know, a small script, very different to how small scripts could.
Ian:
So obviously market making is a big part of what you do and part of the bread and butter, shall we say, of what you do at Tendex.
Of course, market making in crypto is a little bit different than market making in TradFi.
Some strategies overlap and can be used, but there are some particularities or peculiarities in crypto. Just tell us how you think about that and what you’re doing and how you’re approaching market making.
James:
I think a lot of those players have been around for quite a while.
I think some of them Jump Trading, made a lot of noise, you know.
They had a sort of crypto division that seemed to maybe backfire on when some of the sort of, you know, maybe the LunaTerra and they were sort of investigated…were a firm that on Wall Street made a lot of money by not making a lot of noise – tried a different strategy in crypto and maybe they’ve rolled that back.
I’m pretty sure that all the big Authorized Participants of the ETFs, Jane Street, I know you’ve had Flow Traders on here.
Those are all very big firms that are very involved in the space.
I think even more, some of them are even involved in DeFi, much deeper in the weeds than people would realize.
Yeah, I think for us, we can be a little bit more nimble, maybe a little bit smaller.
We can provide some strategies that are maybe more crypto specific.
We thought of this and we have the ability to build this in crypto sort of from the ground up.
And we think we can provide liquidity in a slightly different way to them and where their strengths are maybe being able to move significantly large checks at once. Maybe we can focus in on the other side, which is quoting as tightly as possible on sort of slightly different tokens.
So we’ve managed to find a decent edge that we think is durable.
It will only get more competitive over time. I think it is already.
If you look at the best data that you can find in some of the sort of spreads on the traditional markets, and then you look at the tops of the crypto markets, they’re not far off, to be honest.
I think that there is a whole difference in crypto, maybe closer to FX, is what is the true price?
Because there are a bunch of venues across the world that trade USD, Euro, much like they are the trade BTC, or BTC USD like RULEMATCH does.
And then the question would be, okay, well, what is the true price?
And that’s slightly different to a stock, right, which typically would trade in one, maybe in two places. so, you know, understanding that whole ecosystem requires more “FX way” of thinking about it.
Ian:
You mentioned some of the players in the space and Flow Traders being one of those that have come from traditional finance.
Maybe just to throw a question at you, is there some strategy, market making strategy in TradFi that you think definitely would not work when it comes to crypto?
James:
So some of the ones that are, I guess, interesting is that in the traditional markets, because of the role of clearing house and PB, you can leverage up significantly more.
Because that’s in the traditional space. In the crypto space, it’s 24-7 settlements.
If you take out of a large leverage position on a future and the market moves against you, even if you are hedged somewhere else, you need to margin manage very closely.
Of course, you could over-collateralize both sides, but then effectively you’re losing returns.
What you need to do then is be very cautious around some of the leverage.
I think a lot of the strategies that can be run in traditional finance, because you have that PB or the role to see both sides, that allows you to…
They effectively take the risk, and there’s a T+2 settlement.
They take the risk with a lot of your collateral management, and they can of see.
Obviously, we’ve had the credit space issue where PBs blew up and the damage involved with that. There is genuine risk for them there.
And they need to manage it well, but in the crypto side, the exchanges or the venues have this 24-hour settlement.
If you get into a liquidity zone and you get liquidated, that can happen to you.
And so I think that what will probably happen, and as other parties come along, RULEMATCH being an example, and because of your guys’ setup with qualified security firms, banks. Exactly…
I’m not quite sure what the regulatory term would be….
They can take the risk and so you can have more leverage.
I’m sure eventually, as you guys get more sophisticated spot, eventually maybe other products are offered where leverage becomes a thing. Those other providers can provide a key role in enabling that.
Ian:
Also, with post-trade settlement, that adds another layer of…
James:
It’s all pre-funded. It’s marked live. As a result, actually, a lot of these strategies that you run, you’re very, very conservative.
If you’re not, over time, you’ll be found out.
Moving that capital around, being capital efficient, effective margin management is another huge part of the game that maybe…
It doesn’t exist as much when you’re trading as a small shop. Like us, we would have a single PB that we would work quite hard to get a relationship with because we would be a small client for them.
But in this space, we do enough volume to have a relationship with multiple exchanges and have very good relationships with them.
But the one place they’re not going to give us any flexibility on is margin. That is something we just have to manage ourselves.
Ian:
Right, so you talked about prime brokers and obviously for a fund like Tendex there are many different pieces of the puzzle.
If we talk about service providers, all of those things that come together to help you do what you do.
You have banks, have prime brokers, you have liquidity providers, you have many other players, custodians as well.
Just tell us how that constellation looks for you and what you’re doing and maybe how it has developed maybe all.
James:
It’s ever-changing, which I think is a little frustrating because I spend a lot of time thinking about where the risks are and that constellation, what’s changing.
Obviously, a lot of these firms are still finalizing their product offerings.
Some of them, they offered one product for a couple of years.
They decided that that isn’t the way they wanted to go.
They pivot, so that affects us. I think it’s going to look more and more …
If the big money… The space will bend into what the big money wants.
I think the big money wants what it can see in TradFi and I think hopefully there’ll still be fast settlements, that kind of thing.
There’ll be some of the benefits that this technology stack enables with some of the more security that maybe more traditional finance has…a clearing house, a proper custodian.
One of the big challenges at the moment with the custodian is you can have a custodian and you can store your assets there.
But if you wanted to trade with those assets, it’s sometimes a bit harder, right?
So a venue like RULEMATCH is perfect.
You have your assets in the bank. The bank handles that part.
The trading venue trusts the bank. That’s no problem.
There are beginning to be other solutions in the space as the demand is ticked up. I guess maybe these venues are being leaned on. But it’s still not to the level.
And Switzerland has been incredibly in the forefront of this. And so I think that’s incredibly promising for the Swiss ecosystem within the space of being based here.
But I think that we’re still a way away from where we want to get to.
Ian:
So maybe just on the point of custodians, actually, it’s very interesting how things have developed in the crypto space. have, of course, exchanges who kind of play custodian – play bank in some ways, at least for retail clients.
Then you have technology providers like Fireblocks or a Metaco or others who develop the technology, then you have banks who are starting to come into the space and offer custody.
How do you think that’s going to develop and will that change?
Will it all move towards the banks or will we have some kind of maybe partnerships and crossover between one kind of service provider and the other?
James:
Look, I think you can split them into two.
I think you can split them into maybe using crypto parlance, both a hot wallet and a cold wallet custodian.
If you bought Bitcoin early, you sold two pizzas for 10,000 Bitcoin or whatever happened.
Ian:
Hopefully not.
James:
And if you sold them, if you got the 10,000 Bitcoin, you the trade of the century if you kept it.
So assuming you kept all of that and you’re now very wealthy and you’ve got your 10,000 Bitcoin, then there are a few banks that offer this.
And you want to be very sure that that’s secure.
So you need to put that away somewhere.
There is another element which the space is slowly beginning to question.
And I know a lot of people working this, which is how do you earn yield on that?
That is now an enormous sum of money.
The custodian will obviously charge fees. So is there a way of earning yield on that?
But parking that aside, that custodian is someone who you want to have a big balance sheet, a bank you want it to be safe – as cheap as possible secure, you know
Think Swiss Alpine vault, right? Sort of ex-military vault. I think there are a few providers that offer that.
The flip side is what we would want, which is sort of custodian, but also access to being able to trade because, you know, we’re on the other side of that.
So maybe more sort of hot wallet. There it’s going to require an integrating into every venue. You know, maybe the venues consolidate, but consolidate to three to five or 10, you would still need to integrate into those.
You’re have to be able to deal with each of those venues.
What does that mean? Where are the risks?
How do you ensure that the capital isn’t being double spent? All of that kind of stuff. What is the agreement with that venue?
Who has the right to the capital? When?
That’s not a legal question. Then that custodian needs to be big enough that we have comfort in it. It doesn’t need to be a very good technology stack because of all of the unlocking and locking of capital quickly.
It doesn’t need to be a bank with the Bitcoin that we’re looking to not touch in a thousand years.
And so I think that there will be role for two of them. It think that we got technologists that went to solve the custody problem.
So you’ve got firms that are very good technology suites that enable an OTC but aren’t qualified custodians.
And then you’ve got now firms that backed by big banks and that are more on the custodian side.
So it’s really about just trying to find your balance to work out exactly what you need as a firm. And because there are multiple needs in the market, I think there’ll be multiple service providers. Multiple angles.
Ian:
So you mentioned prime brokers, and obviously that is something that you use quite a bit and maybe even need as an active trading firm.
How can we expect those to develop? Do you think there’ll be more people coming into the space, broadening, deepening their offering?
Adding more services, will others also come into the PB space and try to take market share from others?
James:
Yeah, I think so.
I think some of the venues have been more hostile towards prime brokerages, which is interesting.
Ian:
For any particular reason?
James:
I think prime brokerages make fees competitive because if you just think of it in, I don’t know, strategy, company strategy – you suddenly have a thousand counterparties that are all quite small, or you have one big one, that changes the leverage power.
I think that there’s been a bit of a sort tussles around who deserves the fees, what that looks like.
And so I think that these prime brokerages need to really refine their offerings.
And it’s been interesting working with them as they go through some of these problems and challenges.
They’re very, very switched on teams, well-funded and smart, and obviously looking to do that.
It’s not just been a proliferation of prime brokerages, because I think the exchanges are, for lack of a better word, fighting back or saying, well, “Hang on, we can provide, also can provide fantastic access to these…”
Ian:
And so who’s going to win that battle?
James:
We’ll have to see. But I think that competition’s healthy, right?
That provides a better end user experience if we are the end user in that case.
The one thing that’s missing there is off-chain custody. So if one of the PBs could, and I think that they’re beginning to, but partner up and provide off exchange settlements, et cetera, then that suddenly becomes interesting and people are willing to pay more for it.
And I think the exchange can sort of understand some of the offering.
Ian:
So you mentioned obviously being in Switzerland and that’s a subject that we like to talk about, of course, quite proudly since we’re glad to be here and feel that Switzerland is a good place for crypto companies.
Not only because of legal and regulatory clarity, but because of the maturity of the market and the different service providers that you can find here.
Talk just a little bit about how you are set up, I mean, obviously as a fund, you have banks, partners, have operations maybe in different places.
Tell us how that constellation actually looks for you and why Switzerland is an important piece of that puzzle.
James:
Yeah, so we’re a regulated investment manager. We regulate under FINMA as a portfolio.
When we started, we were under exemption, and now we are sort of full portfolio manager.
That was back in 2019, so that was a long time ago.
And FINMA, I think, the regulator in general here has been very pro the space, and that’s part of the decision.
When we looked to launch this, Jeff was based here. I’ve been a bit of a bias to stay with he is a Swiss citizen.
Well, we could have gone inside back to South Africa, we could have gone to the UK, but the clarity that the Swiss… One, obviously the reliability of the brand. being in a risky asset class is nice to have the of the Swiss brand, which the countries work very hard on, of reliability and stability, et cetera.
But I think that the regulator is enforced that.
And I think that a lot of ecosystem players, it’s not just the regulator. There’s been a huge amount of work done with the ecosystem to educate the regulator.
Critically, there’s been a desire for the regulator to listen. that’s been, you that’s the death knell in regulation is when you don’t allow innovation, you put regulation ahead of innovation.
I think Switzerland’s really struck the balance well, at least in my understanding, and you know, you can see the crypto value, et cetera, of saying “what risk does this do to the more broader population? Can we tick off the most important things, so AML, KYC, all of those kind of anti-money laundering, all of those protections.”
So those are the non-negotiables.
Fine. So that’s done.
The next thing to say, okay, well, does this put the broader financial ecosystem at risk?
Is this putting very sophisticated investors at risk? Do they understand the risk? Well, they’re very sophisticated investors. They probably do.
So then we sit in a situation where FINMA has been able to say “Okay, well, let’s start to understand these digital assets.”
And then from there, it went to ETPs, et cetera. I think that’s been a huge credit.
So we sit with a Cayman Island-based fund and a Swiss investment manager.
The fund is all offshore. the investment manager’s here. It’s nice to be here because you get good access to lot of the exchanges and brokerages and venues that are sort of reliable in the world that have a presence here.
And you can access and have discussions with custodians and a bunch of things because this is a global financial center and so this is a critical mass.
Ian:
You’re sitting in the heart of Europe.
James:
Exactly. It’s a critical mass and it’s pretty easy to get anywhere else if you need to go sort of see a partner or potential investor or whatever.
Ian:
And so your investors are here in Switzerland or in Europe or all over? you say all over?
James:
All over, yeah.
As per regulation, we can only take qualified investors.
Ian:
So you wouldn’t be able to say something to the level of understanding of Swiss investors versus investors in other places versus more globally.
James:
So I think it’s been interesting.
I think it’s maybe broadly done a bit of a wave. It was initially a skepticism, but understanding, I think, the initial cohort of investors you could sum up with mostly senior professionals, typically maybe senior financial professionals that were skeptical of the asset class but understood markets, understood inefficiency that is in markets and that people can provide liquidity.
That’s been done for thousands of years and will continue to be done in a new market and do well.
Then there was a bit more “hopium”, sort of the hope around the sort 2020.
And then, obviously, I think FTX set a lot of the ecosystem back quite a few years.
And if you just look at when last the space was on the front page of the FT, well, was a lot of the 2021 stuff.
And then it was just the FTX trial.
Then that finally disappeared. And then it was the Bitcoin spot.
So investors are only now, I think, beginning to really wake up again to the asset class. I don’t think there’s been a huge amount of new capital raised more broadly in the ecosystem this year, other than maybe just the Bitcoin spot ETF, which is set into Bitcoin.
If you look at the outperformance, Bitcoin has outperformed other than maybe three or four other assets.
And normally it lags, right?
Because if you see it as the less risky of all of these digital assets, normally it’s the one that lags and it sort of outperforms in a bear market.
That hasn’t been the case. And that’s kind of partly why I’m a little bit skeptical that this is, you know a fully blown bull market just because we haven’t seen huge amounts of risk on exuberance within the digital asset space.
And I think investors are now beginning to pick that up.
Investors are sophisticated. Everyone has heard of this space and I think we don’t try and convince investors about the space.
Look, we’re lucky enough that our strategy is a little bit space agnostic. I think it’s tougher if you now, with the Bitcoin ETF to be honest…
You have some competition there. you’re a long-only trader and the question would be, can you outperform a very, very cheap ETF, let’s say effectively free, relative to your “2 and 20 fees”, I’m sure some funds believe that they can and maybe have shown that they can.
Investors can also think, well, I can park and get 80 % of the exposure based on this. So maybe investors offering volatility dampening….
Ian:
So the last question that we always ask our guests and I’m always interested to get the answers of those who come on the program.
Is there something out in the market in crypto or maybe even at Tendex itself that is coming up in the next six to nine months that we should watch out for that actually nobody is talking about yet?
James:
I think that there are two things.
And it’s, the one is maybe more spoken about than the others. I’ll start with that one first.
Stablecoins are obviously a huge topic, that’s, you know, anyone who’s listening to this will understand.
I think what people don’t necessarily understand is how much they’ve penetrated commerce within sort of Africa or in more emerging markets.
I think we sit here in Switzerland and, we pay for our lunch using Twint and, you know, money around is relatively quick and we struggle to imagine that.
You could be in a country where access to dollars or being able to send dollars is very hard.
Just an anecdote, last week a friend of mine was booking a fairly sophisticated or fancy safari, had the option to pay in USDT or Tether for this safari.
Ian:
In South Africa?
James:
Yeah, I think that was into Africa. But he’s based in South Africa. that gives you an option where this is a sophisticated, a couple of thousand dollar transaction that a provider would rather use than necessarily USD because of some of the challenges and that’s something that will only increase to grow.
I think that there Tether sits with a huge opportunity set because it is incredibly profitable, I think more profitable than BlackRock in 2023.
I do think it has, of course, a huge amount of regulatory challenges but they have a huge bankroll to solve a lot of those issues.
I think that it has deep market penetration so it’s going to be very interesting to see can the regulatory incumbents or “up and comers” sort of beat the less regulated incumbent or kind of reform in time.
So that’s the challenge.
And then the second one I think is maybe a little bit more on the crypto side, there will be I think an increasing, I think the first wave of crypto and AI was simply just, I guess, renting out GPUs and renting out computer software for AI.
I think the next wave might be look a little bit more around digital watermarking and sort of the other side, which is how can we be sure that a tweet was genuine?
Can we use some of the blockchain technology, is sort of digital scarcity, to marry with the digital abundance that is AI, to find the things that are true when they’re true?
And that might be through digital identity, so we all have some form of identity.
And I work with a few companies that do that.
So it’s quite interesting looking at that challenge.
And I think that that next generation of sort of blockchain and AI will be quite interesting as people are grappling with the problem of digital abundance that is AI and fake news, et cetera.
Maybe there’s a solution there for the blockchain. I don’t think it’s hugely, I don’t think it’s a massive profitable thing. I think it’s more just using the core technology in a way that it can be used.
Ian:
Very interesting. So with that, we’ll wrap up. Thank you very much, James, for joining RULEMATCH Spot On.
It’s been very, very interesting to talk to you.
I’m sure at Tendex you’re going to be staying right at the forefront of everything that is cutting edge in the space.
So maybe there will be an opportunity to come back and talk about things more in the future.
Thank you once again and look forward to future conversations.
James:
Thanks, Ian. Appreciate it.