So, I just met Thibaud Maréchal. Thibaud has been working in the Bitcoin industry for more than two years as VP Growth and Head of Client Development at Knox, a Bitcoin company focused on insured custody. Knox answers a basic need: how to safely hold Bitcoin for others?
Relax and enjoy the podcast.
- From Real Ventures to Knox
- Custody, a problem that isn’t solved fully
- What is custody?
- Trust minimization
- Becoming the industry standard
- A new risk market
- How to sell complex things?
- A fundamentally flawed shared belief
- Paper money, Bitcoin and stock-to-flow
- Bitcoin in 10 to 25 years
- Are we prepared?
- Stay humble, stack sats
Here’s the lightened transcript.
From Real Ventures to Knox
Thibaud had been working at Real Ventures which was building an internal thesis for the crypto space in general. They were interested in venture opportunities and in a finding a way to hold the assets. Indeed managing capital and holding assets physically are two different businesses. The idea was, where do we find people who will hold the asset safely?
Custody was nascent at that time and it happened that Knox was being launched and precisely focused on custody. So, there was a good match between Knox and Thibaud: he was very interested by what the team was working on and joined in April 2018.
Custody is still a problem that isn’t solved fully
Custody is not easy for both consumers and larger corporate structures. There are different employees with access to funds who may move from one structure to another. There are different governance schemes. Consumer grade solutions such as Ledger, Trezor or Coldcard are not made for institutions.
Knox provides a solution to custody coupled with insurance to cover the risk of collusion and brings a compelling solution to the market.
What is custody?
With Bitcoin the user has the private keys that allows to send and receive Bitcoins. The whole problem of custody is: how can I safely store these keys in order to be able to send the Bitcoins later?
For example: Desjardins holds millions of dollars of Bitcoin. If someone access their keys, they could steal the funds. Addressing the problem of custody is working on providing peace of mind for large players.
(T.M.) It’s all about private keys management. Those keys have a life cycle to handle: they are created, used, stored and backed up. The whole life cycle brings attack vectors to the custodian (Knox).
Knox analyzed the risk vectors across the keys life cycle in order to mitigate them using technology. And for the edge cases that technology cannot cover, they hold an insurance policy against the risks still outstanding. They are isolating the risks and transferring them to insurance markets who receive premiums for the risk.
In short, if you are holding Bitcoin on behalf of others, you can rest assured that Knox has an insurance policy against theft and loss up to 100% of the holdings.
We do custody respecting the Bitcoin ethos of trust minimization via the insurance policy. Besides we are locking ourselves out of the funds. Customers have hardware devices holding special keys letting them issue withdrawal requests. Nothing can be done by Knox unless it was asked by the customer. Trust minimization is the goal. Losses occurred in the past due to collusion in other custodial exchanges, which is what we designed against.
The search of trust minimization is a way to increase the value of the custody because it prevents the theft of funds. The custody is safer, as the clients do not hold on-chain Bitcoin keys so they can rely on you for safekeeping while remaining the legal owners of the Bitcoin holdings.
There is never a zero chance of collusion. Saying that Knox is “100% collusion-free” would be selling snake oil. What we have built is a collusion-resistant Bitcoin key management system. And if internal theft were to happen, our insurance kicks in making our clients whole. The whole point of Bitcoin remains to dis-intermediate, to remove trusted third parties, that are security holes.
On that note, we support sovereignty in Bitcoin holdings. It is desirable that individuals hold their own keys but for complex internal governance models, with multiple individuals responsible for holding funds, with employee turnover and organizational changes, holding Bitcoin on behalf of others can become complicated. Knox helps to bring consistency and continuity against these types of risk.
With these tools a bank could focus on providing its services. They would not need to provide the service and at the same time be responsible for the whole security setup.
Yes exactly, it’s still early for banks but there is a path forward where those services are provided to legacy players.
Becoming the industry standard
Is there a lot of work to do or are you ready to talk to the customers?
Always more to do haha. We work on making secure custody accessible to non-technical operators and administrators while providing high security. While some existing open source cold storage frameworks are great, they are hard to deploy and maintain. There is the risk to lock yourself out of your own funds. (see Glacier)
You could become the default security providers. I imagine competition is not abundant right now and you are early. Do you aim at becoming the industry standard? With good practices and proven experience.
We are trying to raise the bar and we publish articles on risk mitigation techniques to explain the internal doing. This explains the level of complexity that is at play.
We want to commoditize custody. There are different risk tolerance levels on a spectrum. Unchained Capital and Casa provide one type of solution where you remain sovereign with multisig setups. Other players do not want to hold any liability related to key management (Knox).
At some point, it will all be open-source with known procedures and a standard emerging. It’s great because we want to get rid of Bitcoin losses. It will be a beautiful future when it becomes very hard to lose your keys.
A new risk market
We want to build additional layers of financial services on top of custody to create more value to customers. We are looking at insurance at the moment.
The infrastructure we have today is able to originate up to a billion dollars of Bitcoin risk that we can transfer to insurance markets. When this spreads geographically the size of the insurance market increases and the amount of insurance coverage as well. Therefore more Bitcoins can be safely stored. Correct risk pricing will generate revenue streams out of that market.
The collaboration with the insurance markets gives confidence in the storage. We know there is no zero risk but you can price it and therefore hedge against it.
It’s all about confidence in your ability to actually price the risk. Today it is still early stage and risk pricing is quite rudimentary. Traditional insurance categories such as health or home insurance have more data and robust models. They have previous claim events and lots of other data sets to more accurately price the risk.
So, basically, you are creating of a whole new industry for which you are laying down the base infrastructure.
Yes, one could see it that way. Insurance carriers with a lot of capital are looking for diversification just like other fund managers who are investing across different asset classes to build diversified portfolios.
Bitcoin custody is a new risk category. It is an interesting opportunity for insurance carriers: they can charge a premium to the entity that originates the risk, like a custodian, for example. Every dollar of Bitcoin we custody is a dollar worth of risk that we originate. And we transfer that fully to the insurance market that backs up that dollar of risk with a dollar of insurance, in exchange of a premium. We hope that better models of the risk assessment will drive these premiums down over time.
How to sell complex things?
How do you explain all those concepts to the customers? Bitcoin is fairly complex on its own, how do you handle that?
It depends on the customer, their level of understanding, and ultimately what they care about. We try to keep it simple at first. A big challenge is to make these security principles relatable. When talking to private wealth managers, they may be looking for a solution that is a “comfortable risk exposure level.” We are happy to dive deeper in our threat model, our technology primitives and insurance program as they see fit.
So, you can simplify by saying, you have for example 4% of your capital in Bitcoin and each year you will pay 0.01% for insurance and you know you will never lose that 4% basically.
Yes make it simple and start from there if there are questions.
A fundamentally flawed shared belief
Thibaud posted on twitter an intriguing sentence “Bitcoin breaks a fundamentally flawed shared belief: money needs government to exist.” I asked what he meant by that.
Looking back in history, money has been emerging naturally in the free market. Usually a commodity people found useful was used as money. Then different properties make that commodity useful as money: scarcity, durability, portability, divisibility. We forgot all of this today and are stuck in a fiat paradigm since 1971 when Nixon stopped the convertibility of dollars into gold.
Instead of having a money emerging from people, for example people chose gold as the unit of exchange, we switched to a system of money that is not emerging from the bottom but that is governed from the top. Instead of being dollars of the people, it became dollars manipulated by singular entities.
Voluntary choice on one end and coerced utilization on the other. We have fiat currencies because of legal tender laws who demand people living on a certain territory to use a certain money. Bitcoin breaks that by being unaware of territory and tender laws.
Money used to be divorced from the State. Gold was money but had some flows: hard to transport and divide. So, we used paper money which originally was gold certificates but later they became detached from gold. That notion today is completely gone. There is still a shared belief that the US dollar is backed by gold. It is an interesting collective delusion.
To summarize, we had gold in vaults which was not practical to transport from one place to another. So, we emitted paper certificates which were first fully backed by gold. It was easier to transport paper across the country.
Then at some point, the backing by gold was left behind. Each individual unit of paper became less and less worthy: one unit used to give you one unit of gold but after the switch you could not redeem as much gold as you used to.
The material aspect of gold made vault storage more convenient which lead to paper money. With Bitcoin it’s different, it’s not hard to divide or to send across the planet. This is why to some people it is a superior form of money, among other reasons.
This leads to the idea that Bitcoin is competing as money against other forms of money with humans selecting what form is the best. This is a process we are witnessing right now.
Gold failed as money. Bitcoin fixes money. Gold is still used as a settlement instrument between central banks. What makes Bitcoin different, besides that it is harder than gold, is that it is essentially existing in a network which allows almost instant transfer across the entire world at the speed of light.
harder: in that context, means that for Bitcoin less new units are discovered each year (relative to the total amount of units) than for gold. For more details about this concept of hardness and stock-to-flow, have a look at PlanB’s article
Paper money, Bitcoin and stock-to-flow
Gold lead to paper certificates because the “gold layer” was not usable for exchanges.
My question was then, is there a risk we replicate this phenomenon with Bitcoin? If the network is congested or becomes too expensive to use, people will need to use parallel networks to transact Bitcoins without touching the base layer. This parallel network is the Lightning Network (LN). Can we trust it?
To use lightning you have to commit Bitcoins on-chain to make them available on the LN network. There you can transact faster and with less fees. But it is trustless you don’t need any permission to go back to the Bitcoin network. It is not an IOU, in my opinion, with my understanding of the technology.
LN works through a peg with the original network which prevents the emission of additional Bitcoins on the second layer. So, when you commit a Bitcoin from the base layer to LN, you will not be able to duplicate it there due to how the protocol is built.
Yes exactly. There is no one to trust when you peg in or peg out. It is different from the Liquid network which is also a side network with different trade-off for different people: traders, exchanges.
If we imagine some trusted third-party proposing its own personal network to use their services, it’s up to you to use it but you don’t need it to transact. Because you already have LN if you want cheap and fast transactions. I’m trying to understand if there is a way for Bitcoin to replicate the problem that we had with gold.
With LN as a privacy-enhancing layer with a greater throughput, what else do you need? Why would you need a third layer? There is RGB for example Giacomo is working on it (see here). The goal is to handle asset issuance and more reflexivity in the contracts you want to code. I’m not that familiar with those layers yet. As long as the engineering is done in a conservative manner (unlike with Ethereum where you promise the world to have PoS in two weeks), you’re fine.
So, we can envision a future in a few hundred years where people can exchange a few sats without trusting anyone. My great-grand-children could be paid 10k sats a year for their job for example and could still trade and buy things easily without relying on anyone else.
sats: stands for Satoshis, the smallest unit of Bitcoin, 1 BTC = 100M sats
Today it is. We don’t really know how the topology of the LN will evolve. There are operators who will help you manage your LN node so maybe there will be a degree of trust at some point. The companies building those networks are sensitive on these issues to avoid those weaknesses and bottlenecks.
LN is mysterious and requires additional digging.
Yes for example there exist atomic swaps that allow to not pay an on-chain transaction fee to access the LN network in a trustless manner: you can transfer your on-chain btc to LN-btc. There also are providers who make things more convenient. There’s always a balance act between convenience, cost, privacy and so on. No finite answer on that, LN requires heavy digging.
Next subject on the plate!
It is a whole industry emerging with their own conferences.
Bitcoin in 10 to 25 years
I asked Thibaud what will Bitcoin look like in 10 to 25 years?
It will be the only money that we use. It will change a lot of things because it changes individuals and their relation to consumption. We will witness geopolitical and individual changes.
With the amount of monetary expansion we are seeing today and the fact that this becomes political tools to support people, there is no stopping. Fiat currencies are basically designed to go to zero and Bitcoin is designed to go up forever.
I very like this notion of lowering the time preference (two cookies later better than one cookie now) and forcing people to think twice before spending their sats or put their sats in other investments.
For example, if as a portfolio manager you start to measure your risk-adjusted returns not in USD but in BTC, your portfolio looks completely different. It is actually down for the last ten years. That will fundamentally change how capital is used and how money is consumed for products and services. Less focused on consumerism and malinvestment.
How do you see this transition? It changes our relation to time, to capital, to investment. We saw fast changes with the coronavirus, in a few months everyone had their vision of the world changed. With the riots in the US, it took two weeks to see every city protesting. How fast do you think this change could happen?
Hard to tell. In the short-term, we overestimate and underestimate on the long-term. Speaking in halvings, we may have two more halvings before we get to hyperbitcoinization.
halving: 4-year epoch where the daily issuance of Bitcoin is fixed, when moving from an epoch to the other this daily issuance is divided by two
So, 8 years.
Yeah, I think so. We’ll see major adoption in the next few years. In 2017, everyone got aware it existed. Now people will realize this thing is still here and actually going up. With all the uncertainties out there about fiat currencies, keeping your job, pay your next bills, at some point people will realize that maybe there is something going on.
People are getting aware that there is something weird. That all this money printing is not necessarily normal or doesn’t feel normal. You’re working every day of the week making 1.8k a month in Canada and then all of sudden the government is giving away 2k every month because of COVID. You start questioning why am I working that hard if I can be paid more and stay home? Where is this money coming from? Why do I pay taxes if the government can print billions and trillions of dollars?
There is a lot of questioning starting on the question of what is money? I saw a few graphs on Twitter this week about a little spike on the search of “what is money”.
We all know that’s the first question to the rabbit hole. What is money? If you ask yourself this question long enough, you become interested in Bitcoin at some point.
Are we prepared?
If all humanity is on board within 10 years, do you believe the industry is ready for that influx of new users? Will the demand be too high for entrepreneurs or on the contrary entrepreneurs will emerge to tackle the issues?
That’s the beauty of Bitcoin. As it goes up in price, it incentivizes people, builders and entrepreneurs to have exposure to the space and find solutions. I’m not too worried about that. I’m quite bullish on humanity overall. When you let people find their way, they will find their way somehow. Especially when incentives are aligned.
I wish the transition to be peaceful and without hiccups but I don’t necessarily believe that this will be the case. We may see a few hiccups down the road: central exchanges being much more regulated or countries banning Bitcoin, for example. But on the other hand we see countries accepting Bitcoin as legal tender like Japan. A phenomenal achievement. Entrepreneurs will be incentivized to do jurisdiction arbitrage and switch to territories where the best opportunities are lying. We shall see.
I think Bitcoin is on a really good path, much more infrastructure was developed during the last bear market. We have bigger on-ramp and off-ramp in the US, we have custody providers, we have future markets, we have open-source multi-sig products. The LN came out during this bear market as well. I’m curious to see what will happen around block space market in 2021, let’s say.
Maybe what sets this transition or revolution apart is that here if the whole humanity is transitioning to Bitcoin, every individual has the power to provide the services required. Bitcoin is totally decentralized and anyone in any country can build the tools that are missing.
This allows the transition to be smoother. Lots of minds are equipped to address the problem. It helps that the protocol is open-source because this means that humanity, on which both of us are bullish, can attack the problem from many many different angles. We will see solutions emerging, there is no barrier to entry.
Exactly and it is a pure meritocracy: if you can create value on Bitcoin, you will be rewarded.
Stay humble, stack sats
What would you tell to your past self?
I would have bought Bitcoin earlier. That’s the case for everyone. First time I heard it was 2012 with SilkRoad. I discounted it as a weird dark web dirty money.
Besides that I would remind myself to stay humble and focus on Bitcoin, learn about money and monetary history, what happened in the past that lead us to where we are today. That would have prevented me from shitcoin trading. A cycle through which a lot of bitcoiners went. I learned things but in the end I have less sats than what I could have had before. I wasted time I will never get back focusing on projects that will never bring any value to the world.
So, yeah focus on Bitcoin, stay humble, don’t trade. Don’t do leverage margins. Just stack sats and learn about how to contribute to Bitcoin whether education, development, joining a company in the space. It is a very humbling journey.
Why is humility important?
There is so much to learn and the more I learn about Bitcoin the more I realize I know. It really tramps the ego. It forces you go one step at a time. To go from the fundamentals and accumulate knowledge.
Humility also due to the fact that all the people I met so far as the most generous, curious, open-minded people that I have ever met around Bitcoin. When you are part of such a strong fundamentally drive movement, it’s not really about yourself. It’s way beyond that. It’s beyond Knox. It’s something I know will bring a lot of good stuff to the world.
The reason why this thing matters is because I have seen so much changes individually. It’s funny because it’s beyond the self, but at the same time I have seen personally a lot of changes around myself. Whether it was nutrition-wise, consumption-wise, going out to restaurants and bars versus doing more simple things, or rethinking relationships with friends and family. People will tell you dude this is a complete cult. It just opens up a lot of rabbit holes.
When we talk about the red pill (a reference to Matrix) or Alice in Wonderland, it really is that. You are economically incentivized to change your world view on a lot of different aspects. It is humbling in that aspect. I had a whole bunch of premade conceptions whether it was on politics, food, business and life in general. All of these have been shaken up in a way.
It’s the cult where your duty is to question things. So, it’s the good cult. You are pushed to question reality starting with the question “what is money?”, then you keep digging all the time. You learn that by asking questions, you actually understand more and more and less and less at the same time. You are on a path of questioning that is never ending because there is no finite answer. So, it is humbling in that sense. Your ego decreases because you realize wow I am not as important as I thought I was, there are lots of things outside of me that exist and I don’t know them. I don’t know lots of things. I have a lot to learn and every time I learn I realize how few I know. It has a fractal aspect: the more you learn, the more you realize the little know and you keep on going this way.
Exactly, keep on going the bottomless rabbit hole.