The Future of Capital Markets
a CoinShares work-from-home series
In this series, our team at CoinShares will speak with three technology entrepreneurs building new types of market infrastructure, and our CoinShares Capital Markets desk. We explore the fundamental technology innovations driving new applications in cryptocurrency markets and how they might apply to capital markets more broadly.
At CoinShares, our mission is to expand access to the digital asset ecosystem while services as a trusted partner for our clients. We achieve this by building regulated investment products and financial services, including exchange traded products, managed investment strategies, and capital markets services.
CoinShares invests in innovative companies building the digital asset ecosystem, focusing on making early stage equity investments in financial technology companies. Note that CoinShares, associated group entities, or the directors of the firm may be investors in one or more of the companies or assets discussed in this series. We have made an attempt to note instances where a relationship exists between a company and the Group.
In this episode, Meltem Demirors leads a discussion on the evolution of the crypto market structure (past, present, and future), and how software is set to disrupt the entire trade lifecycle from price discovery to post-trade reconciliation. You’ll hear from an all-star panel that includes:
- Anand Gomes from Paradigm
- Darshan Vaidya from X-Margin*
- Sharon Goldberg from Arwen*
- Jean-Marie Mognetti from CoinShares Capital Markets
This panel was recorded live in front of a virtual audience on April 1, 2020. This episode is the final installment of a week-long series where our team at CoinShares explored the fundamental technology innovations driving new applications in cryptocurrency markets and how they might apply to capital markets more broadly.
Note that CoinShares is an investor in some of the companies mentioned in this post, which are denoted with a * above.
The Most Exciting Conversation About Trading Software that You Have Ever Seen
Welcome everyone to our live panel on the Future of Capital Markets! I’m excited about this conversation because I get really excited about trading software and hopefully by the end of this 45 minutes you’ll be just as excited as I am. In my email to everyone who’s watching live, I promised the most exciting conversation about trading software you’ve ever been a part of. So time to fulfill that promise. Now I’ll let all of our panelists introduce themselves and their companies.
Anand Gomes, CEO of Paradigm: Paradigm is an institutional-grade messaging platform with embedded workflow automation. In other words, it’s an RFQ-style trading protocol embedded within a chat/communication workflow. This allows you to trade any crypto-asset with any counterparts and clear/settle anywhere you want.
Darshan (Darsh) Vaidya, CEO of X-Margin: X-Margin is a Zero-Knowledge cross-margin solution that allows you to trade across multiple venues with one pot of funds in whatever custody solution you choose.
Sharon Goldberg, CEO of Arwen: Arwen streamlines and secures the settlement of institutional digital asset trades. Arwen handles the settlement in a non-custodial way, meaning traders maintain full control of their funds during the settlement process (Arwen does not take custody of the assets), effectively eliminating counterparty risk.
Jean-Marie Mognetti, CEO of CoinShares, Head of CoinShares Capital Markets: Jean-Marie has run a crypto trading desk for the last seven years and has watched the market evolve from a voice-driven, brokerage-driven market where traders were primarily bilateral or on exchanges that no longer exist, into a market that’s mostly electronic. CoinShares Capital Markets has historically been focused on servicing its own internal products, but is now in the process of opening its business to external flow and onboarding external clients.
Are Markets Having their AWS Moment?
So let’s begin. Software is eating the world. Look at what Amazon Web Services (AWS) has done for software-as-a-service and the ability to consume software on-demand and in the cloud. In my opinion, finance is beginning to have its “AWS moment”. A lot of consumer finance and SME finance are going through this democratization process, where the back-end bank or financial institution is the platform, and new products and services can plug into the back-end via API which allows you to do new and exciting things. I think it’s only a matter of time before software eats capital markets, and the crypto market is one of the first places where this can be done today due to some of the unique features of these natively digital assets (settlement finality, 24/7 liquidity, etc.) Before we discuss this, let’s take a few steps back and talk about the problems that each of you faced that led you to build your company.
Anand: The idea for Paradigm was conceived of in the energy markets, but the problems are similar across all OTC markets. Liquidity is fragmented across tons of venues (OTC, dealers, P2P, desks, etc.), and there’s no single-standard for workflow automation, etc. These problems are endemic to OTC execution across all asset classes, not just crypto.
Meltem: That’s an interesting point. Let’s talk about the idea of “non-standard” for a moment. Jean-Marie (JM), you started your career trading in the commodities market. Tell us how standardization differs in the crypto market vs. the commodities market?
Jean-Marie: In the commodities market there are some things that are not standard, but in the crypto market, nothing is standard. You go from one broker to another and you get all different sets of rules and procedures. This was a clear problem from the beginning.
Meltem: What’s really cool about Paradigm is that it doesn’t force everyone to standardize, but the RFQ system is built so that market participants can do non-standard things through a standardized screen and with a standardized clearing and settlement solution. Participants can plug-and-play different components.
Anand: Well put. What Paradigm provides is not a set of standard instruments, but a standard framework which can lend itself to non-standard things. Non-standard things are really just aggregations of smaller standard pieces, so we give people the choice to combine and plug these components in and out. Paradigm provides on-demand liquidity, in size, for any crypto-asset, instantly, privately, and with automated settlement, all within a single point of access. This is not rocket science, it’s not new, it just hasn’t been done properly.
Meltem: What is new, though, is instead of having to onboard with you and having to custody all of my assets with Paradigm, I can connect all of the places where my assets are directly to Paradigm and use a layer of software as the single execution venue as opposed to executing across all of these different places independently. I can even do OTC trades through this same single screen.
Anand: Right. And this is only possible because we are able to unbundle the execution layer from the settlement layer. At the end of the day, customers want choice. Nobody wants to be tied into one bank or one wallet provider or one broker, humans love choice. Building software that is an extension of this idea is why all of us are a part of this space.
Meltem: Thanks Anand. Moving to the next component of the trade lifecycle -- clearing. Darsh, you came to this market as a trader and realized how capital inefficient it was to trade in this market. Talk to us about that problem.
Darsh: It’s strange because crypto is known to have crazy amounts of leverage. I came to this market originally as part of a derivatives desk and we tried to apply our traditional markets technology to the crypto market to take advantage of inefficiencies and leverage. The problem that we ran into was that in order to capitalize on this, we would have to trade across several different exchanges, deposit collateral with each of them (and take on their credit risk). Your ROI is then massively hampered by the fact that you don’t get cross-collateral benefits. If you have off-setting positions across different exchanges, which most market makers would, you don’t get the collateral offset. You have to post double collateral at both places which significantly reduces your ROI. The solution for this in traditional markets is either a) central clearing or b) a huge amount of trust, but neither of these solutions work in crypto.
Meltem: Let’s talk about why this doesn’t work in crypto because this is where Arwen comes into the conversation. When you deposit collateral on exchanges, you don’t just post collateral in an escrow account. Any time you send cryptocurrencies or stablecoins to someone, you are settling with finality. So the stack becomes inverted, you actually end up settling before you’ve gone through the clearing process, and then continuously re-settling until you’ve closed out the position. This creates an interesting risk that you’re taking on because you’ve already settled with these exchanges with finality. If the exchanges goes under overnight, your collateral disappears.
Darsh: I want to be careful not to use the wrong analogy, but it’s essentially like getting casino chips. It looks like your money is there, but it’s not really your money. It’s the casino’s money.
Meltem: I think there’s actually a phrase for this -- not your keys, not your coins?
Darsh: Exactly. So we had the problems of both counterparty credit risk and collateral efficiency. Central clearing did not solve this because you’d then have to trust some other central counterparty, which, in crypto, is quite hard to find because you need someone with a lot of collateral to backstop these trades. There are a lot of different products within crypto, so it’s very hard to get a lot of standardization. If you want a central clearinghouse they need to have x number of methodologies for x number of products and instruments that have been invented in crypto.
Meltem: There is no ISDA manual for crypto.
Darsh: Right. I think this standardization is coming into play, which will help, but crypto still has an inherently bespoke element which makes central clearing difficult. So what we decided to do at X-Margin was to replace or disintermediate this central clearinghouse with a sort of black box software which was powered by zero-knowledge technology. You can funnel your trades through this zero-knowledge calculator and then get the portfolio effects that a clearinghouse would give you. In the future, we should then be able to collaborate with solutions like Arwen to then eventually be able to settle that trade across the different venues.
Meltem: This is so beautiful! Clearing now becomes a totally different function where I can have any number of bespoke contracts, I can feed them into this black box margin calculator -- and I think margin calculation is a really interesting challenge in a lot of ways. We’ve certainly seen challenges with reference data that is used in pricing the contract, and in finding ways to clear things. This, by the way, is not just a crypto problem. Only 30% of the derivatives market is actually centrally cleared. JM - are you having PTSD flashbacks to what your life has been like for the last 7 years?
JM: No, I love it. This is the future. We’re moving in the right direction. I’m not sure how much of it will be achievable tomorrow -- some elements work today/tomorrow, some elements may be a bit longer-term. I’m not fully sold, for example, on the idea that central clearing will be replaced by software, but it will definitely be helped by software. I’m still convinced that we’ll need some extent of central clearing, but it will be done in a much more efficient way. Euroclear, for example, was created in 1960 by JPMorgan. The technology underlying almost all central clearing in Europe or the US is ~50-year-old technology that has barely been upgraded over time. There are very few people in these organizations that understand the full value chain, which creates an opportunity for companies like X-Margin to come in and improve.
Meltem: Thanks JM. Now let’s move from clearing to the next part of the trade lifecycle, settlement. One problem with bilateral trading is that I don’t know if I should send you money first or if you should send me bitcoin first. How do you figure this out?
Sharon: This is the dilemma in OTC bilateral trading. The way this has been solved historically is whoever is less powerful has to send their assets first. This means that at some point in time, the more powerful party holds both assets, and the less powerful party experiences counterparty risk. Today in OTC settlement there is a lot of this counterparty risk. There are also a lot of manual processes that large traders undergo to ensure there are no issues when moving significant values of bitcoin (including test transactions, triple-checking addresses for accuracy and validity, etc.).
Meltem: JM - you’ve been doing this for years. Tell us about your first OTC bitcoin trade.
JM: My first trade was in 2013 with a gentleman who now runs the OTC desk at Bitfinex. I bought $200,000 worth of bitcoin when I was brand new to the market. And Sharon’s right, it was a dance to determine who sends first, etc. The first time we had to send bitcoin we had a bunch of crazy processes - printing wallet addresses on pieces of paper, crossing off each letter one-by-one to make sure it was accurate, etc.Meltem: So Sharon -- how does Arwen help?
Sharon: There are two pieces to what we’re building. First, there’s a security piece that reduces counterparty risk -- instead of having one party send first, we have built a blockchain protocol that allows both parties to send simultaneously. The second piece is removing the manual processes involved in these trades. So for test transactions today, you have to wait for your counterparty to confirm that it’s received. We streamline all of this to provide simple, one-click settlement. But, the cool thing is that we do not take control of your assets. You can still keep your assets in custody wherever you’d prefer.
Meltem: So to put this all together -- in this beautiful future world that we’re describing, I can go on Paradigm, I can choose if I want to access an existing venue (currently Paradigm supports CME & Deribit), or I can do RFQ and request a quote from all of the dealers on the platform. I can write the contract however I want, and I can do multi-leg execution. Then, instead of using a central clearinghouse, I’ll choose to use X-Margin as my calculator. Then, finally, I’ll verifiably provide proof-of-escrow through Arwen, then settle through Arwen’s API, which is integrated to whoever my back-end custodian is (regulated custodian or self-custody). So basically what we’ve done is taken the entire trade lifecycle from being vertically integrated and custodial to being beautiful, unbundled, non-custodial software.
JM: So when do we start?
The Timing of This Shift
Meltem: That’s a great point -- this is all amazing, but it’s not all possible today. As we said before, some things can be done today, some are more long-term. It’s April 2020, which is hard to believe because March felt like a year, but we have just witnessed one of the most cataclysmic events in global capital markets history. What’s been really interesting is that, yes, there was a lot of volatility in crypto markets, there was a rapid drop in price, but all of this infrastructure that’s been built has allowed people to do things like manage their risk, manage their margin and collateral, liquidate positions almost seamlessly. This is due largely in part to the rise of stablecoins. All of a sudden we’re able to represent fiat dollars (and potentially other types of assets in the future) on the same network where we’re settling the trade. So we now have several types of assets other than cryptocurrencies represented, traded, and settled on these networks. Along with stablecoins, the second component that enables this is the 24/7 nature of crypto markets. Due to COVID-19, people have been talking about shutting down certain market venues. What I’m hearing in the market is that people who are using custodians are seeing a lengthening of their service level agreements (SLAs) from 12-24 hours to 48-72 hours because when you require people on-site you still require a lot of manual processing. In my view, the last two weeks have made everyone in legacy markets really interested in what’s happening in crypto markets. Do you all agree or disagree?
Anand: I totally agree. We’ve had conversations with a few pit brokers on whether they can use our software to block trades while the pit is closed.
Sharon: I want to add to that and take it one step further. What we can do with blockchain is really unique and absent from the physical world. On the settlement side where we’re building, you can use the blockchain itself as a trusted party without having to rely on any other third party. This is a really unique characteristic of the space and is one of the biggest reasons why I’m here building.
Meltem: Sharon, we didn’t even talk about what you did before. We have 3 traders on this call, but you’re the anomaly. Tell us about your background.
Sharon: Yes, I’m a Computer Science Professor at Boston University. I teach cybersecurity and cryptography. I’m the odd duck on this panel.
Meltem: But what’s so funny is that you actually explain the issues in settlement and how settlement should work better to me than both traders I’ve met.
JM: Well, to be clear, most traders don’t know anything about middle and back office.
Darsh: That’s true. It was only since I started trading crypto that I’ve become my own back office. Before that it was someone else’s problem.
Sharon: Back office is super fascinating to me. The idea that we’re replacing people moving files from filing cabinets to this immutable ledger that can’t be changed, that’s not controlled by anyone, that can’t be censored by a government or taken down -- that’s an incredible technology. My team is really excited about doing settlement without having to rely on any trusted third party. And this is especially powerful when trading bitcoin because bitcoin doesn’t have any value if there’s no bitcoin blockchain. So if we’re using the bitcoin blockchain to back our settlement then we’re using the blockchain for what it was designed to do. And that’s what keeps my team going every day in this space.
Meltem: But Sharon, you don’t want to trust me to hold your money and give you casino tokens to trade?
Sharon: I actually love that question. This is what we call colored coins. The simplest example that everyone knows is Tether. In Tether, you’re giving dollars to the party that issues Tether (Bitfinex) who then prints you some Tether. You’re then trusting Bitfinex to hold onto your dollars. And this is a really understandable use case because you don’t have a blockchain for USD.
Sharon: Right. So what we have in cases where we would take an asset and issue a token for it, but there’s no direct binding of the two, that’s a very natural use case for an asset that’s not issued on the blockchain. Like dollars, or tokenized art, or real estate.
Meltem Asks for Leverage
Meltem: Hold on. I have a question for Darsh. Can I post part of my mortgage as collateral so that I can get leverage?
Darsh: Sure, if some custodian will take it.
Meltem: JM, would you accept part of my home mortgage as collateral for leverage?
JM: … no Meltem. I have to stay firm on this one. It’s a no. (laughing)
Meltem: Fine. Let’s try again. At CoinShares we have recently been involved in the issuance of an asset called DGLD. DGLD represents ownership of gold bars in a Swiss Vault represented on the bitcoin blockchain. Could I give you some DGLD and would you let me use that as collateral?
Meltem: Darsh, how about through X-Margin?
Darsh: Absolutely. If Sharon will lock it for me then we’re good to go.
Meltem: Alright so we’re good there. Anand, can I plug into Paradigm and request a quote?
Anand: Absolutely. Let’s do it.
Meltem: Wow. Look at that. The gold bars in my bank vault are now collateral for me to lend. So we have all of these assets that banks want to put on the blockchain -- I’m not sure how far out real estate is because there’s still a pretty big price discovery problem -- but I think commodities can definitely be represented on a blockchain as long as you have a way of digitizing it in a way that a custodian and counterparty will be willing to accept. There’s this really interesting world where right now there’s this huge market that’s emerging in the crypto world around lending and the potential for people to use different types of assets to lever up -- I think that’s really interesting. JM, you’ve spent some time in the lending market. What are your thoughts?
JM: I think using new assets for collateral is fine, it’s just a question of liquidity of the collateral which is going to drive your haircut. What you tend to see in the crypto lending market is you have some very educated people in the market, like market makers in Chicago, for example, and they know exactly what the borrow rates are and what the collateral should be. But then you have other, less educated participants trying to borrow that want to do things like deposit less collateral than the value that they want to borrow. This creates a mismatch of expectations which is pretty intense. There are, though, some market participants who will do these trades which in turn creates some big asymmetric risks.
Meltem: OK. We’ve been talking for a while now, let’s open it up for questions from the audience.
Q&A from our Discussion
Question: How far away are we from building all of this software? What’s 12 months away vs. 10 years away?
Anand: For us, this software already exists. We’ve been in business for 7-8 months. We’ve done a ton of volume, have 120+ counterparties, and onboard more and more every day. What I think will change is the market as a whole will gravitate towards these types of solutions. That is happening gradually right now, but I think this health crisis will push people towards these types of automated solutions, similar to how the financial crisis and regulation pushed people towards centralized clearinghouses and digital trading. I think this crisis only accelerates this process. 12 months from now, I think the vision of a true single point of access to multi-product, multi-dealer trading will be materialized. I think a significant portion of the market will be using us (hopefully) or another solution like us. Because consumer expectations have changed, the market will demand this.
JM: What’s interesting is that Anand, your background is in commodities and traditional markets, probably using ICE Chat, and you saw the problems with these products and built a better solution in this “new world”. You learn from the past, and then improve.
Anand: That’s exactly right. It was also the most interesting time to be in the financial markets because the financial crisis had just happened. There were all of these changes happening -- cleared vs. uncleared, Volcker, Basel, capital requirements, etc. You had to learn all of this stuff in a very compressed amount of time which turned out to be very helpful.
Darsh: For us, our v1 product will be released in the next month which will include OTC bilateral trading and integrating with Paradigm. So you’ll be able to RFQ bilateral derivatives. Later down the line, we plan to integrate with exchanges, starting with crypto derivatives exchanges, then move to traditional finance over the next 10 years to tackle all of the bespoke products in the traditional markets. Broadly speaking, this might include bespoke equity derivatives, commodities derivatives, contracts for difference (CFDs), and various other cash or physically settled bespoke derivatives that have no central clearing solution. I also think we will change the dynamic of central clearing. I agree with JM that there is an important role for central clearing. Right now, the calculation part is done as well as the central counterparty part, right now they are both bundled together. What we’re doing is unbundling that and focusing only on the calculation part. The central counterparty part will then become more competitive. You can imagine a situation where we can provide cross-margin across several different asset classes, and then a central clearer who provides the backstop if trading entities go bankrupt -- this becomes a competitive process.
Sharon: In the next year we’ll be rolling out our settlement product which we plan to make the standard way of settling bilateral trades, so people can settle bilaterally without having to trust anyone other than themselves. In 10 years, what I’ve been thinking about recently, especially over the last few weeks, is the rise of stablecoins which is accelerating. What’s really interesting is that we have the opportunity to use blockchains and stablecoins to settle and clear more than just digital assets. So the future we’re looking towards long-term is to see how much of the financial industry will transition into using blockchain technology and blockchain-based settlement technology. My hope is that this will just become a much more efficient way of moving all types of assets around.
JM: We need to start where we were 10 years ago, where we weren’t doing a single trade electronically in crypto, to where we are now at CoinShares where 99% of our trades are done through API. In the long-term, where CoinShares can really help is by opening our OTC desk to institutional counterparties, integrating ourselves with software like Paradigm, X-Margin, and Arwen, and then helping to marry this software with institutional demand.
Question: How cooperative or uncooperative have the exchanges been given that these products and services overlap with their customers and markets?
JM: We speak with a lot of exchanges on a regular basis. These exchanges understand that the market is changing rapidly in crypto and are ready to adapt and embrace this change. I don’t think exchanges see this type of tech as competition, but more as a way to bring more liquidity and more flow to their venues. Anand: I agree. The exchanges are super opportunistic and they know that people want choice and ease of execution. They understand that we are a platform that will ultimately drive more clearing volume to their platform. We’re working with 2 exchanges right now but will hopefully be integrated with 8-9 by the end of the year, and they’ve all made it super easy to work with them.
Check out the video for more!