The Big Whale: To get straight to the heart of the matter, why do you think lending and borrowing on a public blockchain is more attractive than in the traditional world?
Paul Frambot: Lending and borrowing is essentially about bringing together two types of user: those who have surplus capital and want to earn interest on it, and those who need funding. The aim is to use an optimal algorithm to match each lender with each borrower as efficiently as possible.
A shared infrastructure, based on a public blockchain, is essential. To aggregate all the flows and players around the world, regardless of regulation or identity, it is crucial to have a neutral, shared settlement layer, where everything happens in a credible and transparent way. The main reason is efficiency: we are looking for a single place where everything converges and is settled. This centralisation unlocks greater efficiency.
Accessibility is another interesting aspect. How do you enable people to build on these infrastructures in an ultra-simple way? Currently, setting up a lending and borrowing business outside the blockchain is very complex, particularly from a technical point of view. Ethereum offers a much more accessible approach to developing these solutions.
Finally, after efficiency and accessibility, comes resilience. Although this is not yet obvious today, due to the numerous hacks suffered by decentralised finance (DeFi) in recent years, in the long term, having open, massively audited and formally verified code is the way forward. This offers far greater resilience than centralised servers, which can be deactivated at any time. Completely open and decentralised software offers guarantees of security and availability unmatched in the traditional financial world.
What was your initial thinking when designing Morpho Blue ? Because previously it had been a different protocol, operating as a yield aggregator.
We took a very entrepreneurial approach: what was the best way to enter the market? At the time, the main player was Aave , a very large lending and borrowing protocol on Ethereum. We noticed a big gap between their lending and borrowing rates, due to inefficiencies in their protocol. Our solution? To optimise this gap. We built a kind of intent aggregator that allowed better rates than Aave, while preserving the same liquidity and risk management.
We deployed this protocol in 2022, and it grew rapidly, at a time when DeFi was almost at a standstill. We reached around $2 billion in deposits, becoming a major lending protocol. Our success was such that we became almost systemic for Aave. The size of Morpho relative to Aave raised questions: people wondered how it would end, whether we would represent the whole of Aave.
We then realised that we had reached our market limit. Our growth was capped by Aave itself, since we were built on their protocol. It became clear that we needed to move away from that and build our own, totally independent and innovative lending and borrowing protocol. This is the current version of Morpho that we call Morpho Blue, but you can think of it as the real Morpho V1. It's the one we're currently developing, launched a few months ago. Today, it's already approaching two billion dollars in TVL (total locked value).
If we trace the history of loans and borrowings in DeFi, first there was Compound with its 'liquidity mining', then Aave, and now Morpho. Can you explain the main differences between Aave and Morpho?
Aave works like a bank or a fund. You deposit money with Aave, and they manage it. They look after the risk, the code, and update it periodically. Recently, they updated all their instances. They manage thousands of risk parameters every day, but not in an automated way. They have to perform the calculations off-chain before integrating them into the protocol. In short, it's a decentralised broker to whom you entrust the management of your money.
Morpho, on the other hand, is an infrastructure for banks, not a bank in itself. We don't offer a standard yield solution. Our role is to provide the tools to enable everyone to create their own lending and borrowing use case. In this way, we can recreate systems similar to Aave v3 or Compound v3. Morpho is not a product, but a platform on which players like Gauntlet or Stakehouse Financial can develop products that compete with Aave.
Morpho is therefore closer to Uniswap than to Aave. Its core is immutable, permissionless and totally decentralised. Anyone can create the lending and borrowing use case they want, with their own compliance and risk parameters. Morpho's vaults, managed by risk curators such as Gauntlet and Stakehouse, offer an experience comparable to that of Aave. In the end, on the Morpho website you will find a broader range than Aave, with various risk profiles - from the lowest to the highest - and various compliance profiles, depending on your needs.
Can you explain in a simple way how Morpho's infrastructure works?
Morpho is based on two main components: markets and vaults. The Morpho markets are similar to Uniswap pairs, but for loans. Each market includes a collateral asset, a borrowed asset, an oracle and a liquidation ratio (loan to value ), which serves as a risk parameter. We have hundreds of these pairs, such as BTC to borrow USDC, ETH to borrow USDC, or stETH to borrow USDT. The possibilities are almost limitless, even including pairs like SHIB to borrow ETH. However, as on Uniswap, some pairs can be risky and should be avoided.
For end users, managing hundreds of markets or doing your own risk management is neither practical nor desirable. These markets, which are totally immutable and governance-free, operate like micro-primitives that will then be re-aggregated in the vaults. It is the latter that provide the curation. The Morpho markets focus solely on lending and borrowing, with no risk management. Although difficult to use directly, they are extremely efficient and totally reliable from a code point of view.
Liquidity management and interest accounting are carried out in a fully programmed way at market level. However, for a user, choosing the right market can be complex. For example, should you lend in USDC against BTC or against stETH? That's why we've introduced a second layer: the Morpho vaults . These vaults, which can also be managed by anyone, are completely decentralised. Any player can deploy a vault, whether it's a native crypto risk trader like Gauntlet, a fintech, an exchange platform, or any company looking to create a return solution. Once the vault is deployed, they can begin to manage and select risks for their users. The management of these vaults can vary: some are administered by DAOs, others by algorithms, or even by specialist risk management companies.
Interacting with a Morpho vault is similar to interacting with Aave. You simply lend your USDC and wait. The vault then takes care of market-level management, a complex task for the average user. It acts as an intermediary, facilitating interaction with the market. This approach is reminiscent of Uniswap: providing liquidity directly on Uniswap can be risky and complex for an individual. Standard practice is to entrust this management to specialist players. With these complex players interacting with the Morpho protocol, we can offer more complexity and expressiveness at the primitive level, a unique capability compared to other market players.
With your vaults, if I'm a bank, I can launch a vault with my own parameters, my own liquidity and my own players around that vault, right? This is not yet totally possible with Aave, with the same transparency on the management and risk of the markets or assets that I am going to lend or borrow.
On Aave, Aave manages everything. On Morpho, anyone can manage. That's the fundamental distinction. A Morpho vault can be managed by a very opaque algorithm or be completely open source. It can comply with French regulations or American rules. We make no assumptions about how it will be managed.
Here's how you should see Aave compared to Morpho: Aave controls everything. The human beings behind Aave manage the risks, the code, etc. At Morpho, even if the team disappeared tomorrow, the protocol would continue to operate autonomously. That's really not the case with other lending solutions.
With Morpho, users bring their own compliance and risk management. If you're an institution and you want to use Aave, you come up against a lot of regulatory constraints, which makes it very difficult to use. On the other hand, with the Morpho infrastructure, you can create your own lending solution, incorporating the appropriate KYC (know your customer) compliance or any other requirements necessary to use Morpho vaults.
But isn't there a risk of liquidity fragmentation on Morpho? For the moment, it seems easier to manage liquidity on Aave.
That's not exactly the case. On Morpho, the markets effectively fragment liquidity, with each market having its own share. However, the vaults re-aggregate this liquidity for both lenders and borrowers via a contract called public allocator . It's a bit technical, so I won't go into detail. To simplify, Morpho offers great granularity and then re-aggregates this liquidity. As a lender, you benefit from liquidity that is equivalent or even superior to that of Aave.
Let's take a concrete example: if you deposit in a vault, say Gauntlet's, which provides liquidity on 10 markets, these markets re-aggregate as if they were one. Our competitors often criticise Morpho for fragmenting liquidity. In reality, this is not the case at all. Morpho is just as liquid as Aave thanks to this re-aggregation. That goes for both lenders and borrowers.
And that's not all. Not only do you have the Gauntlet vault, which works in a similar way to Aave, but also the Stakehouse vault. These two vaults have different risk profiles, while sharing some of their liquidity. For example, Stakehouse manages real-world assets and crypto-native loans, while Gauntlet focuses solely on crypto-native collateral. Some of their liquidity and risk profiles overlap. With Morpho, where risk profiles overlap, you share markets and therefore liquidity, in proportion to that overlap. This is an advanced concept, but crucial to understanding Morpho. Not only do you re-aggregate liquidity, but you also share it with other players. It's like two forks of Aave sharing their liquidity for common collateral. The mechanism is similar.
All right, but how do you ensure that there is no unwanted aggregation of liquidity between Stakehouse and Gauntlet vaults in some cases?
In some cases, isolation can be total. Take Maker, for example: his Vault on Morpho is completely isolated. In Morpho Lens, they use separate markets from everyone else. However, for Steakehouse and Gauntlet Vault, the situation is different. I believe they both use WST East as collateral. In practical terms, this means that USDC is deposited in the wstETH/USDC market from both Stakehouse's and Gauntlet's vaults. As a result, Stakehouse benefits from the liquidity provided by Gauntlet, and vice versa. It's a unique concept, not found on any other lending and borrowing platform.
Recently, Aave launched 'isolated markets' with Gnosis and Etherfi, replicating a mechanism similar to how Morpho works. Why do you think they chose Aave?
Six months ago, Aave was opposed to the deployment of other instances. Their approach advocated a single solution aggregating all liquidity - that was their thesis. Then Morpho started to gain significant market share, and Aave changed course, creating other instances - a strategy they swore they would never adopt. Now they're launching various Aave marketplaces in an attempt to stay competitive.
The problem is that each vault, each instance of Aave, has completely isolated liquidity. For example, if EtherFi created a USDC vault on Morpho, it could share liquidity with the Gauntlet vault. Let's say Gauntlet develops something close to the main Aave v3 instance, while EtherFi collaborates with Gnosis on a separate instance. On Morpho, they would share liquidity with Gauntlet's main vault. On Aave, on the other hand, it's completely siloed - two separate protocols, with nothing in common. That's the fundamental difference.
As for why they opted for Aave rather than another protocol, I honestly don't know. We are not in talks with Gnosis and EtherFi at present. I wasn't even aware of their plans. Generally, when an actor is in talks with both of us, I can't disclose anything publicly.
I beg to differ...
Honestly (laughs), I can't remember an occasion when we lost out to Aave for a specific partner. But when we're not in the loop, it's similar to what happened with the Trump protocol being launched. We had no idea. Sure, they're launching it, but it wasn't a competition where we could have presented Morpho. Aave has the strongest brand, with eight years of existence. Morpho is only three years old. It's possible that the Trump team hasn't even heard of Morpho.
What type of player are you currently targeting? Which traditional, more institutional players are you in discussions with?
The essence of Morpho is that when you build on it, you own the whole thing. You own the code and you own risk management. The code belongs to you because it is immutable. We don't control it, we can't change it. That's not the case if you file on Aave, where the code can change, as we saw last week.
Certainly, but on Aave, it's the DAO that changes the rules.
In fact, with seven days' notice. To give you an idea, our Morpho Optimizer protocol, built on Aave, still manages around $600 million. It's a very complex protocol. When we update it, Morpho's main protocol remains unchanged, unlike the old one. An update requires several audits, taking around four months to finalise.
In comparison, Aave typically gives a week's notice after a vote is passed, only revealing the code a week before implementation. This leaves very little time to examine changes, adjust dependent protocols and carry out audits. This is why no serious institution builds on Aave - it's too risky. They can't check the thousands of Aave-based protocols to identify any critical changes. Of course, the vote is public, but it's last minute. Seriously securing a protocol takes several months and multiple audits. One week is far from enough.
As a result, by building on Aave, you own neither the code - which can be modified - nor the risk management, which is controlled by the holders of Aave tokens and their service providers with their opaque algorithms. With Morpho, you own everything. As an institution, fintech, exchange platform or traditional financial player, you can deploy your own vaults and markets on Morpho. You are the sole owner, with no need to trust or even know the Morpho team. This is not the case on other lending and borrowing platforms.
This is crucial for institutions: they refuse to cede control to unknown token holders. What's more, this is the very essence of Ethereum: a permissionless and decentralised infrastructure. Without these features, you lose Ethereum's fundamental advantages. This complicates access to builders (builders). Admittedly, you can easily reach individual users, as Morpho does with its various vaults. But to reach the real market of builders , those who will create the best tools for mass adoption, you need to respect decentralisation and permissionlessness. Few protocols actually achieve this. To my knowledge, only Uniswap and Morpho really adhere to these principles, as Ethereum promotes them.
So what type of player are you currently targeting? There are entities like fintechs or banks - take Revolut for example - that might be interested in offering access to Morpho. This would enable them to generate returns from stablecoins or offer businesses simplified cash management.
I can't disclose which players we're in discussions with.
I'd like to stress again...
I would say, however, that it has become clear in recent years that many fintechs and centralised exchange platforms have built the necessary infrastructure around accounts. I'm talking about account abstraction tools, such as smart wallets or paymasters to manage gas charges, partnerships with Layer 2 solutions, or even the development of their own Layer 2 solutions. In this way, they bring together all the building blocks needed to interact with a DeFi protocol.
If you're a fintech, you excel at distribution, but not at manufacturing. I often use this image: good companies master both manufacturing and distribution - they make cars and distribute them. Fintechs, on the other hand, only do distribution. They do not create financial services, but rely on traditional financial institutions. To emancipate themselves and have their own financial infrastructure, they don't need to turn to traditional finance or build a traditional financial infrastructure - which would be extremely difficult. Instead, they can use DeFi to create their own financial infrastructure.
From my point of view, you are the protocol for institutional players, whereas Aave is more focused on retail users.
Not quite. Morpho is a versatile infrastructure. We make it possible to create use cases for both institutions and individuals. Our flexibility means we can build a variety of solutions. That's the essence of our approach.
You've reached about a billion and a half TVLs now, haven't you? Is TVL still relevant to describe the growth of a protocol?
It's a complex question, in all honesty. I'm not sure it's the best metric. We feel that DeFi still functions as an echo chamber, where we constantly recycle the same users. Around 80 users generate the majority of volumes and deposits in DeFi. As a result, we don't find the current situation particularly interesting. We don't focus too much on metrics. It's true that we post impressive figures, with billions in deposits, but I wonder about the relevance of talking about total deposits, TVL or other similar indicators. In my view, we are not yet targeting the right market, but that will take time.
Currently, why is collateral not reused when borrowing on Morpho, unlike on Aave? On Aave, you provide collateral, receive rewards and can then borrow an asset against that collateral. But on Morpho, that's not the case. You can provide funds and win, but to borrow, you have to provide collateral again. When will this feature be introduced? And what is the logic behind this difference between Morpho and Aave?
First of all, it is quite possible to do exactly the same thing as on Aave. However, when you earn yield, it's never free - you're taking on extra risk. If you provide collateral to borrow and earn interest on it, it's because that collateral is itself at risk. It is lent to others, which is what we call rehypothecation. At Morpho, we let the market makers decide whether the collateral should be an inactive asset or rehypothecated, i.e. lent to others.
The risk is that if the collateral becomes illiquid, liquidations cannot take place. In the case of Aave, this could lead to the bankruptcy of the protocol due to a high use of capital. On Morpho, we let users decide what kind of market they want to create. If someone wants a collateral like ETH that remains inactive, that's possible, although I personally think it's too conservative and won't gain much traction. But some people want it, and they can do it on Morpho.
Let's take an example: my personal risk profile is that I want more yield on my ETH. I can therefore use assets such as stETH or ETH on Aave as collateral. It is also possible to use a vault or a wrapped version of a safe as collateral. This is an argument that some competitors use against Morpho, saying that we don't have native rehypothecation, i.e. the Morpho DAO doesn't control rehypothecation. This is true, and we do not wish to do so. It's up to vault managers to decide whether their markets accept rehypothecated collateral, which carries more risk.
But could we consider such an implementation in the future? I suppose you're going to tell me that it will be up to the curators or vault managers to decide.
Exactly. And I think we already have several markets that use some form of rehypothecation as collateral.
However, the reason it's not popular on Morpho is because rehypothecation is not, in all honesty, an ideal form of collateral. It is an illiquid asset that mainly benefits the lending protocol, as this liquidity can be reinvested elsewhere in the protocol, artificially inflating the TVL and other indicators. Let's take a concrete example: would you prefer, as collateral, inactive stETH - where your only risk is related to Lido 's smart contract, with a staking return of 3.5% - or AWETH, Aave's deposit token, which exposes you to all of Aave's management and smart contract risks for a lower return? Rehypothecatable assets as collateral are generally disadvantageous, especially for borrowers.
In practice, safer forms of rehypothecation are preferred: for USD, we use SDAI, an excellent collateral, or sUSDS . For ETH, you opt for stETH, which offers a better return and greater liquidity than the other options. Ultimately, rehypothecation primarily serves the interests of the loan protocol.
Currently, Morpho does not generate any revenue. Rather, it is the curators or certain vault operators who benefit, but not the protocol. What is your long-term strategy on this subject?
Morpho does not currently generate any fees. The fee mechanism (fee switch ) exists, but it is not activated. The vault managers, on the other hand, have made a lot of money - around $23 million in the last nine months, which is pretty good. There are a number of reasons why we're not charging at the moment.
First, let me explain how we could generate money if we wanted to. This would be by activating the fee switch , which involves taking a portion of the interest paid by borrowers to lenders. The Morpho protocol could therefore take a share of these interest rates.
However, if we enable fees, vault managers could question the value we actually provide. They could simply take our code, copy it and recreate the same thing for no fee. Code, in DeFi, has become a commodity. The more we move towards web3, the more all code will be open source, and SaaS models will disappear. That's my prediction.
Are you in contact with players like Uniswap to discuss this idea that code will become a commodity?
No, that's a personal reflection. I have no information on Uniswap's position on this. However, I am convinced that the move towards on-chain will mean that all code will have to become open source in order to remain competitive. It's a question of survival. In Morpho's case, activating the fee switch in our licence would automatically make the code open source. Similarly, if Morpho CAD starts generating revenue, the code would immediately become available to anyone.
This means that anyone could copy Morpho's code and waive the fee?
Exactly. That's why we have to justify the added value of our fees. And that value lies not in the code itself, but in the state of the network and its liquidity. Why choose the Morpho network? Because it is already operational: if you are a lender, we have borrowers, and vice versa. We already have liquidity and an ecosystem of tools. If you simply copy the code, you're starting from scratch: you'll have to find your own lenders and borrowers. Morpho can therefore generate revenue by capturing some of the value created by these network effects.
Why not activate the fees now?
We are convinced that the network can still grow considerably without it. Morpho currently has around $1.5 billion under management, but we believe its potential is much greater. Thea second reason is the lack of regulatory clarity around fee collection. We are keen to ensure that we are compliant with regulators before we start generating revenue.
Which regulators are you in discussions with? In Europe? In the US?
I can't divulge the specific details of these discussions. Just know that we are taking a cautious approach on this subject.
Recently, you announced a partnership with Binance Web3 Wallet. Not directly with the platform, but with Binance. Can you explain the logic behind this partnership?
Currently, Binance offers an 'earn' section where you can lend and borrow various tokens. Everything is centralised and managed behind the scenes by Binance, which is not developing efficiently. The more assets there are, the more complex and difficult risk management becomes. So it's natural for Binance to evolve, especially with the development of portfolios like Trust Wallet, a kind of Binance derivative. They have all the account abstraction technology needed to enable their users to interact with DeFi. In this way, they offer their users access to vaults in the DeFi to generate returns with different risk profiles. In short, it's a way for them to outsource part of their financial infrastructure.
Can we envisage the multiplication of this type of partnership with other centralised exchange platforms in the future?
We can envisage many possibilities.
Is this coming soon? In a few months, a few days?
You're free to imagine whatever you want.
Currently, you're only deployed on one layer 2 , which is Base. Why this exclusive choice?
First and foremost, Morpho's culture has always been one of focus. We are all about lending and borrowing. No stablecoins, social networks, games or wallets. We only do one thing, and that's lending and borrowing. We focus all our efforts and resources on this single objective. We don't spread ourselves too thin. For a long time, Ethereum was our main choice, because we could focus on it and build.
However, over time, we realised that we were missing out on some of the potential adoption, because layers 2 offer new opportunities for users thanks to lower gas fees. It's also crucial for large integrations and channel partners who want to eliminate gas charges for their users. They can't afford to pay $30 Ethereum gas fees for each transaction.
So it's not primarily to access Coinbase's large user base, which Jesse Pollak, Base's creator, says has ambitions to go on-chain?
I think it's natural for exchange platforms to go on-chain eventually. And not just for them, but for all the players, starting with these platforms. They are the technology companies that are closest to the on-chain universe. They already have the necessary infrastructure: Coinbase, for example, has an intelligent wallet, paymasters , and dev kits accessible to all. I can't predict Coinbase's specific actions, but the general feeling in the market is that many players are moving towards on-chain. It's a really exciting time to be building in DeFi.
Let's talk about your token, which is about to be launched on the market. What was your rationale behind this token? What would be the usefulness of the DAO?
I'm going to share my initial vision of tokens. This is my personal point of view. Currently, the Morpho forum is the scene of many discussions on the distribution of liquidity and the launch of the token. The exact process has yet to be defined. However, my initial thought was that the search for utility in tokens was mainly aimed at getting round the regulations. If we compare tokens to shares, nobody questions the usefulness of a share. It serves to solve incentive problems for early investors and contributors. Tokens offer an additional advantage: they can attract early adopters thanks to their programmability. In my view, a token does not need a specific utility beyond its role in adjusting incentives to stimulate the protocol. Tokens represent a form of network property, created by the interaction between lenders and borrowers. This dynamic generates value in a decentralised way.
But how useful would the token be for governance in the DAO?
Actually, what you can do with the token is activate the fee switch and decide the percentage of fees charged by the protocol. You can also decide on the treasury budget, for example to grant subsidies to certain players.
When will it be transferable?
I can't give a specific date. It's currently a topic of discussion within the community. Our aim is to make this launch as decentralised as possible, and we've received a number of proposals regarding the method of launching the token.
A final question about the "conservative" option that appeared in the application. Why did you use it?
It was a feature we tested for three days. We withdrew it, but I can explain the concept. The idea was to simplify the user experience by highlighting the most conservative safes. However, following complaints from users, we quickly removed it.
What was the logic behind this approach?
Our aim was to offer a smoother experience, where the user wouldn't have to think too much. Unfortunately, the feedback hasn't been positive. That said, we are currently working on a complete redesign of the application which, in my opinion, will deliver the best user experience DeFi has ever seen.
What do you think of Uniswap's initiative with Unichain? Can we imagine a "Morpho Chain" one day, along the same lines?
I think Uniswap's approach is relevant overall. While I don't agree with or fully understand some aspects, Uniswap seeks to create the best trading experience possible. Their aim is to reduce what is known as 'loss versus rebalancing' (LVR), which measures the loss of value of liquidity providers and the outflow of money from the system. To achieve this, they need to speed up the chain, optimise maximum value extraction (MEV) and reduce gas costs. A smooth user experience also requires pre-confirmations, among other things. Unichain effectively addresses these user experience issues, which is why I find the idea interesting.
What I don't fully grasp is the DeFi chain approach. If you're building on the Superchain with Optimism and others, you should believe in layer 2s interoperability. In that case, why need a specific DeFi chain? Why opt for a generalist approach? I understand the point of hooks other features, but in my opinion - and I may not have all the information - they should have focused solely on hooks for Uniswap. I doubt that the whole of DeFi is built on Unichain, it doesn't make much sense to me. But I could be wrong, and I'd be delighted to be proved wrong.
