Morpho: Developing Foundational Infrastructure for DeFi Landing

Juan: Welcome back to another episode of StablePod, a podcast where we host conversations with the individuals who are shaping the future of decentralized systems. I'm your host, Juan Esquivel. Today's guest is innovating in the world of DeFi lending and money markets. Paul Frambo is the CEO. and co founder of Morpho Labs, a team developing foundational decentralized protocols underpinning the future of lending.

In today's episode, we dive into everything from why Morpho is innovating over incumbent lending protocols like Aave and Compound, what permissionless lists really means to Paul, Morpho's unique approach to governance, and their insightful plans for the Morpho DAO. 

Gustav: Yeah, I think, having been in DeFi for a very long time, I really connect and respect Morpho's approach, right? You can say they pay respects to what has been built before them, But something I've seen with a lot of people who's worked on DeFi for a long time is that they can recognize, some of the flaws of the existing protocols, 

And then trying to approach the problem in a different angle, where like Morpho, they're kind of saying, okay, you know, the tech, the tech is completely permissionless, right. And we cannot change anything there. And then we create a very open system where anyone can build what they want on top of it.

I really enjoyed, hearing, how they also wanted to solve the, scalability issue that you could say we have in DeFi to some extent, DeFi has been scaling quite well, but it also feels like we've hit. A little bit of a wall, 

And now, a lot of the, OG protocols have been looking at, adopting, real world assets as the means to scale, right. where I think like Morpheus approaches is very unique and we don't see a lot of protocols in the market that has been built in the same fashion.

So it's really interesting to see, how do we take and actually improve upon, the model we have at hand, right? 

Juan: Yeah, I think that one of the reasons I was excited to get Paul on was, he, a lot of the content they put out and their messaging from the Morpho team is that, they want to take this more modular approach.

they're bringing a different. Way of thinking for lending protocols, which I think is very interesting. and in my mind, my take is that maybe there's room for both to be possible, right? the Aves of the world, as Paul mentioned, as Paul describes them as more of like an on chain fund, a different way.

His, Morpho, Morpho's approach, excuse me, is more of like an actual protocol. He speaks about internet infrastructure, for the decentralized purists out there. You know, that's kind of maybe what we should be actually striving for. So, you know, speaking to all these different builders, it's good to get on different perspectives because everyone sees, this spectrum of decentralization quite differently.

I wanted to bring him on to speak a little bit about Morpho Dao and what, He expects, what he sees. How does he see that developing over time? And of course, the mechanisms that, will be governed in the future. So yeah, I think it was a great conversation.

If you enjoy, lending protocols, if you enjoy, a different perspective on how to decentralize protocols, I think you'll find this conversation quite interesting. So without further ado, our conversation with Paul. Welcome to the podcast, Paul Frambo. Hey, thanks for having me. Paul, I think a great place to start is by explaining this new era for lending protocols, you know, as they shift from monolithic to more modular, which I think your protocol is taking this approach, you know, from app to more like an infrastructure layer.

how is Morpho Blue fit into this, new, way of building lending protocols and how are you, leading this new era? 

Paul Frambot: Right. So I think there's a couple pieces there. I think the modularity, like the modular narrative is interesting, but I think it's much more than this, in general.

And I think the best way, in my opinion, to capture what's happening in DeFi and not just in lending is like the shifts from being, financial products to financial infrastructure. And so what we've seen over the last, five years or six years building in DeFi is the first protocols that emerged, whether they're swap V1, V2 or compound or Aave were.

Small contracts that were directly usable and sort of like user friendly, right? the user base was very not complex, right? lenders and borrowers and traders and liquidity providers like were very primitive. And so we needed things that, could work on its own. And so this is why the protocols like more peer to peer lending, like Eastland or Dharma, were not able to, work in the first place because they were not sufficiently autonomous, right?

they required too much complexity from users. And this is what enabled compound to thrive because the essentially like the idea of compound was like two things. First, like enshrining the risk management of things and say, Hey, we, we will handle some of the complexities. on behalf of users, and, and, and for risk management and also for interest rate.

And this has been like really a breakthrough in the sense that users could deposit their capital and earn interest, without thinking. And, that has been like over the last, you know, three, four years, been a great success, compound grow a lot. Then Ave, you know, took the path of compounds and, and we're able to improve the compound model by adding like new.

new features, new monitoring and engineering all around Aave. a lot of like the great things that Aave has built is around the protocol itself, not in the protocol, but like everything, the, the, the governance processes, the engineering behind the scenes that has been able to streamline like the, the updates of risk parameters, et cetera.

and we've been building on top of those protocols for a long time. So before we're going into like more for blue and, and, and this sort of like, In for a protocol, we were building what we call morph optimizer, which was an optimization layer for those protocols, for RV and for compound. And so we're actually, and I think we still are today, the largest at scale protocol builder on top of lending and barring pools.

So we know a lot about using lending and barring platforms in a way that I don't think any other, you know, protocol or team has, has gone through. the reality, when you build. a multi billion dollar protocol, like MorphOptimizer on top of AvinCompon themselves, you realize that they mainly have three big limitations as financial product.

The first one being, the, security and autonomy of those protocols. Like those are financial products. They're not as autonomous as they look like. you have DAOs like voting parameters on a daily basis, et cetera. And those are, this has like some, some, some limitations and then efficiency and flexibility, we can talk about later.

And so to come back to your question, the big question we're asking ourselves is like, how do we grow DeFi? How are we able to capture such a large market as the finance, users in the world beyond what crypto is today without compromising on the security and the reset of things.

And you have really two approaches to this. either you try to have one product that fits everything. and this is what is trying to do, which is like having one monolithic approach where you're able to cover all use cases or collateral, in one single pool of capital, aggregating all the liquidity or the capital, et cetera, and then you have another approach, which is essentially doing much less. And let people build the specific experiences that tailored to one specific use case, essentially. this is the more for approach, which is, okay, we're going to build lending in layers, and we're going to just build what's common to every single.

lending and borrowing engine, which is the core lending engine. And then on top people will, build the abstraction layers that they need. it's always different from one use case to another. this is about. Not building a financial product that, is super easy to use, for lenders.

Like when you go to more food, you don't know which food you should be using. you have to do some risk decisions, or delegate this to like some risk management layer that we have built on top. but, we don't do it for the user. And so I think this move is like just moving from being a financial product.

App to a financial infrastructure on top of which you can build specific experiences, et cetera. And then this layer can be modular, like can have modularity, et cetera. but everything at the end of the day comes down to the same base core contract, which has a shared state with everybody. And this is not necessarily modularity.

That's like just building finance, like the internet layers. but, but yeah, so. Okay. That's the long answer to your initial question, but yeah, happy to elaborate on, yeah, so some of the aspects of it. 

Juan: Paul. So, so I hear you. It sounds like, you know, taking this modular approach definitely allows, lending at least lending in the specific example to, to actually scale as to what you envision.

so. To clarify, maybe for our listeners who, who actually maybe are on the opposite side of that and think that, you know, why can't Aave scale or why can't similar lending protocols actually scale? Can you clarify that you did kind of mention like one is an app and one is infrastructure and your, you know, mental model or what you, your narrative you believe is that, you know, infrastructure is the actual thing that can actually scale To the size that we actually envisioned.

Can, can you just clarify for the, for the other side that thinks the, the other approach? 

Paul Frambot: Of course. so first I think that Aave can still scale a little bit, you know, like, Aave is like, you know, probably 15 billion. I believe Aave could go to like 30 or 40. but lending it over collateralized lending and borrowing is trillions in traditional finance.

So we're not even, you know, we're several orders of magnitude behind what we should be expecting from T5. And What does it take to get to trillions? Well, you need to adapt to a bunch of different risk profiles, a bunch of different compliance profiles, and you can't be hacked. Right. That's like, you need flexibility.

you need the efficiency of trial time and you need the security, like the guarantees that your protocol, like when never ever get hacked essentially. And so if you're taking the average approach, which is essentially like. a large pool of capital where everybody's going to deposit their money and everybody, is going to be treated equally.

and the funds are going to be, managed by the Dow. So the Dow is like setting those risk parameters. So what is the risk parameters like the collateral factor, or the liquidation, Parameters, like at what point do you get liquidated? How much can you liquidate from position? What is the record pricing, et cetera.

And I think you have more than 700 of those. And the reason why you have 700 is because have as different markets for each market. It has like, no, probably a six to 10 different risk parameters. then I have is on multiple chains as multiple versions, et cetera. And the user does not choose on any of that.

The users let the Dow manage it, for themselves. But at the same time, users want a bunch of use case. They want to be able to borrow with a bunch of different collateral. and they want as much flexibility as they can. And so the job of the Avid is like to trade between the flexibility, the security and the efficiency.

It's always like this kind of trilemma where, you know, they will want to list some assets. but not others because it will, be too risky to list some assets and others. But the more assets you list, the more parameters, the more management you have to do, the more complexities the Dow has asked to handle.

And so the Dow has to sort of handle all, the complexity and this is the least scalable, right? so that I would say is the bottleneck. I still think we can do better when, and they're doing a great job at automating more and more and putting a lot of the hard work off chain, whether it's on the Oracle, automations or the risk parameters, automation, they're doing a great job.

It's one very specific risk reward profile. When you deposit your funds on Aave, everybody's treated the same way. Same risk, same compliance. it does not reflect the reality of the risk profiles of people. People have different risk and different compliance profiles.

in my opinion, it's not possible to think that one single pool will rule the activities of all lending and borrowing in one single product, essentially. 

Gustav: Yeah, I think it's, really interesting, the process you guys are taking, right.

Because I also agree with you. and I think that if we look at someone like maker, for example, they have also taken a very different approach today compared to five years ago, where I believe that, the shutdown model, it's also kind of playing into this narrative, right.

We want to, remove. some of these, individual use cases from the core of the protocol and allow, other people to set up what they want. Of course, I think the incentives in Morpho and in Maker are quite different. And where I think Morpho brings a little bit more flexibility around the lending itself and where Maker focuses more on the stable current liquidity.

but this is also an interesting match, right? So I think like one of your most successful. you know, partners so far is maker. So maybe you can talk a little bit about like, how do you see connect more for fitting into the current landscape? Right. Because I do believe that can like one of your biggest modes today is to work with existing protocols and teams.

Paul Frambot: just to make sure I get this right, like you want me to clarify how we plan to work with all of those different teams 

Gustav: Yeah. I think before we were talking about okay, you know, more folk can like, you know, attract is like huge, massive market, right.

potentially in the future with the trillions of dollars and overclutters lending traditional finance, but maybe you can tell us a little bit about how you guys are engaging and interacting with the defy ecosystem today. Because I think this is one of your.

biggest use cases as of today, right? 

Paul Frambot: Right. so MorphoBlue essentially unbundles the AVI landing board, right? Into two pieces. You have the purely core protocol part, which is MorphoBlue that externalizes any form of management of code, upgradability of, you know, risk management, et cetera.

and you have metamorphosis, which is the, risk management layer, which is, specifically interesting for a wide variety of players. So the risk experts themselves, like Gauntlet, Block Analytica, RiskDAO, Bprotocol, Stakehouse Financial and more on the real world asset side of things, R7, all of those are vaults.

On top of, of more full, to, you know, basically have like, risk management that sort of do the hard work on, on behalf of lenders. And you also have that like, like maker, for example, building, you know, with spark Dow, using the more for infrastructure in order to be able to, to go from like, lending and borrowing on, on, on, on their platform.

and I think. If you look at the maker, so yes, the short answer is yes, more first providing infrastructure for all those players and each, the more players we have in the ecosystem, the more they self reinforce just because at the end of the day, all those different pools are all coming down to the more for blue protocol, right?

And they share liquidity to the extent that they share risks. I'll give you an example. gauntlet and stakeouts are doing two very different, things. gauntlet is more into blue chip assets while steakhouse is more into the real world, assets. but the two share some markets in common, they share the S St SDC, pool, for example, in common.

And so they will share liquidity to the extent that they share risk. A bit like, you know, if Compound and Ave, could share liquidity for their ST collateral because they both list STEs or something, whatever the asset. so anyway, so we're building this common infrastructure with a shared state where, Any player that has lending and borrowing needs can build a use case they want, on top of Morpho without having to trust us without having to, trust the code, and that's very relevant for risk experts and DAOs in particular stable coins. We also have the angle DAO. with the angle stable coin, I believe frax team also has deployed some markets as well. so yeah, that's the current state. 

Gustav: Okay. And, so like with all of these kind of like individual actors, like inside the protocols, like inside the markets, right.

To like which extent do they still have like flexibility after the market to like deploy it at like any security concerns or like how does Morpho basically kind of like secure the liquidity? because I can imagine that like, you know, like let's say you're like Gauntlet and you want to like deploy kind of like a vault.

you potentially want to make changes over time to the vault, right? so maybe you can speak a little bit to like kind of like how, like with so many different parties kind of like engaging in the protocol, how do you kind of like ensure that there's some kind of streamline process around security?

Paul Frambot: Great. So this is where we do have a very strict difference with any other protocol is that we externalize risk management from the core protocol. So the core protocol in itself has is a fixed point in the space of like risk reward profile. And then as a risk manager, I can decide to allocate some liquidity in one market.

And so the way gauntlet and others are doing risk management is essentially by allocating, you know, maybe USDC in the vault that is, depositing into the St. SDC market, right? Or into the, CB east, C market. and by choosing a market, you choose the liquidation loan to value and the oracle.

the approach we have to risk management is that lenders and only lenders, I mean, lenders and bars like market forces should be pricing risk and not a central authority, like a Dow or something like this. And we don't see, to us, that's the most scalable thing.

You just provide the infrastructure and then lenders and borrowers can decide which markets are comfortable with in terms of risk. And this is how they allocate the equity. So I'd say the downside of this approach is that you can't just change like one parameter. If you want to change the Oracle, you can't change the Oracle, 

The upside is that everything is priced, right? Like, for example, if you have a collateral, in this market, when the rate that you get on that market prices, the risk of that collateral, which is not the case on average, and then volts have to do liquidity and risk management to essentially get exposure to the right oracles and the right, markets. so yeah, I don't know if that answers the question correctly, volts can decide to reallocate liquidity and change if the market condition of an asset changes over time.

They changed the oracle that way. By depositing capital in the markets where they feel like they are comfortable being borrowed from and not changing the market structure itself. And I think it's, it's very important in like the Morphe vision to, to like completely remove this from, from the market structure.

Juan: Paul, coming back to the Spark Vault real quick, I just wanted to get some clarity. Why does Spark, let's say in this scenario, come to Morpho? Why do they deploy this vault on Morpho? Let's say, as opposed to going through maybe something like Aave. Maybe for our listeners that maybe aren't too familiar.

Like, why is that more attractive to them to come to you? 

Paul Frambot: Right. So it's a few things. So first Spark does have an Aave free fork that they run. Right. So they could, you know, if they had the flexibility and the confidence of using that instance, they could 100 percent do it, but they're not doing it.

Right. And there are many reasons for that. the first reason, and this one, I'm just guessing, you know, I don't know from, from the maker team specifically is that if you look at the number of incidents that I have a had over like, I have easy free and I have easy to even had over the last year. It's quite scary to be frank.

and, so that's the first consideration. Like there is a lot of features in Ivy free. Most of them are not used. by, by the protocol at all, 

So that's the thing. Second thing is that on Aave, each collateral, as I was mentioning, like each collateral If you have high, very high yielding collateral, they're not priced, the, the, the rate is not differentiated from one collateral to another. Say you have, SUSD, or, LRTs that you start listing, on, on the Spark Land instance, right?

Well, people will start depositing some LRTs and borrow a lot the, east, on the east market. So the east rate is going to go up. But what you have on Ave and Spark is a lot of STs, East loopers, right? People that borrow STs in like very, very large quantities. We're talking about like, billions.

And so if the rates of the borrowing rate of East goes above the STs rates, well, those guys are right, right? Essentially. and so they have to make sure if they list an LRT, That's the first don't pump the right too much. So that's additional complexity, additional monitoring. That's completely off chain, not controlled by the protocol that has to be monitored by someone that you have to trust somewhere.

We don't know who that is, but essentially that has to do it. And then the Dow has to green light this proposal. And obviously the Dow, does not have three PhD in risk management. They have no idea what's going on. So there's just going to green light everything that the Dow says 

Work to the external, like, you know, management and a lot of trust as well to, to all those people. And so what you can do then is list the LRT, but up to a certain cap, but then your USDC depositors, suddenly they get exposed to under RT, right? And then you have to monitor the interest rate curve of with plus the cap of your LRT deposit in order to make sure that, you know, the rate is not going to go too high, et cetera, et cetera.

And, you know, you kind of get the idea of very quickly that this is not scalable and that if you want to add like, you know, 10 LRTs, essentially the trust assumptions are going to grow bigger. and, you know, eventually a problem will happen, like so many times in the past it almost happened, right?

so, yeah, and I guess those are the reasons why Spark Lands felt very uncomfortable, listing S-U-S-D-E as collateral Multiple, liquidation went to value like SUSD is a risky asset. So they had multiple, liquidation onto value, multiple caps.

for each of them, which is not something you can do in the ABU, for instance. And then they are able to have markets like being re aggregated with their MorphoVault on top of that. Right. So that's very like flexible for them. and I think all those reasons combined were the reasons why they decided to deploy the market.

Gustav: Paul, how do you guys deal with the interest rates in this scenario? Right. Because I think you know, if we look at. maker had some isolation over the individual tokens. But at the same time, they also had the problem with the kind of scaling that.

I remember back, when I was working there, onboarding new collateral was such a headache, even though you actually had the ability to set interest rates. So I think in this case here, how do you then put the price in because I know that you guys also use market dynamics.

Right. And I completely agree Where the problem often comes in, especially for the first generation of lending and borrowing markets is when we have these natively yielding tokens, right? Because then all of a sudden there's kind of like a part of the token, which is not possible to represent in the system, right?

Because what they really only are geared to work with is kind of like price and liquidity, right? so once you add kind of like native yield on top, it's just like a third dimension to the token. They're not really able to handle. So like, how do you handle that with, interest rates and the market rates?

Paul Frambot: So that's a very good question. So in our past to building Morphe Blue, which is a completely permissionless and governed immutable lending and borrowing core protocol, the hardest challenge has been the interest rate model. how do we have one interest rate model that's.

whatever the assets, whatever the market condition will self adapt and give you a good rate, a rate that makes sense. Right. And so interest rate models are great. They were, I think, I believe were invented by compound. they're not perfect. and you know, they're not used in traditional finance for many different reasons.

but they're great for a blockchain environment. the thing was having an incumbent approach is that, they have the Dow vote for each asset set, the different, parameters. in the case of SOSDE, that's even more complex because they have to monitor how much the rate is pumping.

every week and then change the parameters, go through garbage port proposals. what we have is a PID controlled interest rate model, where essentially, we have this curve of when the utilization of the capital is high, the rate is going to spike, if the utilization remains high for too long, meaning that you have a lot of borrow demands, then the rate is going to go up on its own.

Right. And so the more for blue interest rate model is completely autonomous and is self reacting to any market, condition. And it works, and it has proven to work very well, in all the different instances that we have. so whether you have SOSDE, LRTs, et cetera, it takes two, three days for the interest rate model to essentially self adjust to either the market conditions or the assets.

And, the DAO does not have to do anything, does not have to vote on anything. And that's extremely convenient. and so, yeah, we're very, happy about this interest rate model. I still don't think it's the end game. I think there's still improvements to be made, but, 

Juan: So Paul, as you start to think of Morpho, Blue as this ecosystem platform that is enabling other things to be built on top of it.

could you speak a little bit to that point? And like what you envision, like, What else can be built on top of Morpho Blue besides, metamorpho how do you envision this ecosystem developing?

And is your team thinking about it from the grants and the incentives, perspective, or are you just letting the market decide? where things get built, 

Paul Frambot: So I really like the idea of living complete freedom for people to build on top without imposing any constraints.

We can suggest metamorphosis, a great addition that we've been suggesting, which is essentially this risk management protocol where anyone can deploy vaults and allocate liquidity. to the market, which is essentially a proxy for cleverness for lenders, right?

Morpho Blue is hard to use in itself. It's not a financial product. It's a financial infrastructure. You have to make your own risk decisions when you supply on one pool or another. if you don't know what to do, you use a metamorphic world that is going to do the hard work for you. But you have a bunch of other protocols that have been building great abstractions, whether it is for the UX of it, for the risk management of it, for the automations and, yield optimization, to build auto liquidation mechanisms, 

Right. And so, so I think we don't want to impose lines of code on users for us. That's extremely important. just because we've gone through an old transparency, we've gone through all those different war rooms. 

If you have too much lines of code, if you're building like too, too much unnecessary, you know, features, et cetera. So I'm a big fan of immutability, formal verification, hardcore, like security practices, but that only comes with like minimalism, right? If you don't have, like, if you're adding features just, you know, to be able to be feature complete.

While it could be built on top, then it means that some people that are using your code base, they don't need feature I2Z, but still they have to suffer from its presence like in the code, right? So I want to have the core protocol completely minimal, and then people opt in to the layer they want, to the abstraction layer they want, depending on the use case they want, with the minimal number of lines of code, always immutable, 

Juan: Gotcha. I want to double click on risk management. Paul, I want to dive a little bit deeper in there. Cause I think it's quite unique, your approach, the protocol takes in this, in this case. So could you explain for the listener, you know, Well, what the role of the risk curator is. I know you did, you explained it a little bit, but like, why do you, why did Morpho take this approach or was it decision?

Why do you think this is the optimal approach to risk management, in the lending protocol? and what have you seen from users? I believe, you know, Morpho Blue has been around now for over six months. Like, what have you seen in the activity? That validates that this idea is actually, you know, what people want, what the market wants.

Paul Frambot: Right. so first Morpho Blue has been live for three months only. so we have more than a billion dollars of deposits in the protocol in just three months. Morpho Blue is a take that Enshrine risk management in market structure is not the most, the safest, most flexible and most secure approach.

And so far we've seen a lot of players adopting it as risk managers, as builders on top of it, but also as users in an extremely short period of time, and in just three months you've got to 1 billion. So it's hard for me to say 

You know, we'll have market validation in a year from now, we'll outgrow like the Ave model or the compound with them. I have a ton of respect for what they've built. even if, this respect is not always mutual, I believe that, you know, like the compound and other approaches is really great.

But, But, so yeah, I, you know, I'd be cautious and just wait for a year or two to see how things have been going. but we feel very strongly that, the free market approach of letting people decide what they, were and essentially taking their own risk. And if they don't know what risk to take, then the delegated to a specific layer and rebuilding finance in general, like the internet, I think is just the way forward.

I don't see a world in which like, the entire financial system is run by, you deposit your money and everything is running like huge monolithic pools that are DAO controlled. it does not seem realistic to me at all. Right. Like to be frank. And when you talk to people outside of the DeFi echo chamber, they're like, come on, this is not serious.

we're never, ever going to use that. Right. Like, especially on the lender side of things. And I mean, people depositing capital in like a serious way. They're like, how can we trust the Dow? And they started asking relevant questions that we don't even ask as, as DeFi lenders and borrowers, which are.

How has those risk parameters decided at the end of the day? What's the algorithm? Is it open source somewhere? Well, the answer is not right. It's just like, you know, a bunch of risk parameters are pushed every day in DAOs and, and they get voted and green lighted. but this is not what the financial market beyond DeFi is looking into.

And this is why we've had so much of a hard time getting adoption outside of DeFi. And it's still, you know, very focused on, on the current crypto user base. So I think the real test Is this, like, how do we break out of crypto and this only because this, this only will prove like the, the, the, will prove some value in the longterm.

We're not interested in just capturing like the other market or the compound market. Like we're going from 

Gustav: much bigger than that. So Paul, I mean, I guess like a counter argument here could be that like, you know, DeFi has still brought kind of like a level of transparency. I know that you say like, you know, like the individual, like risk algorithm, right.

Might not be like open source or like, you know, I think like sometimes, you know, you have to wait like a bit for like the risk reports to come out. but, but couldn't you also say it in like, you know, you guys are kind of like bringing back these kind of like individual actors who will sit on top and like manage risk, but, to what level of like transparency, transparency informed upon them.

Right. Because you know, at Dow, you know, we work a lot with Dow, so we know how, you know, Slow, inefficient, and, and, you know, like the problems you can have around like that model itself. Right. Because of course it's kind of like trying to run like a small democracy on top of a financial product. Right.

But this also can give some security, right. And it can also give, yeah, like a different level of like transparency. So like, maybe you can talk a little bit. It's like how promoting like a transparency culture, right. Because you don't want to deviate completely from, from the reason why we got here. Right.

Paul Frambot: 100%. 

Gustav: And, 

Paul Frambot: and, I think in Morpho specifically, we're completely externalizing risk management for the protocol. And so you see a lot of different approaches. You have completely open source and transparent approaches. You have DAOs like Spark, you have like closed source algorithms, and more or less centralized, decentralized like approaches.

The one thing is that it's a free market and it's an open market for risk management. So anyone is free to, and then users will decide what they think is the best, right? should they use like, do they value transparency? Is it useful for them? My guess is that eventually it will be one of the biggest arguments for risk managers to choose their vote over another one.

They will have to open source just because open source will be used as a competition. tool to compete for like between different risk managers. And we've seen that like, I think Rizzo and B protocol have been open sourcing all of their formulas and how they calculate LTVs, and I sometimes have like this remark that, you know, because we're pushing the risk management to above layers, those above layers are going to be closed source, et cetera.

And that's factually not right. Maybe some will get it closed source for their own specific reasons. Maybe some others not. And in practice, like I've written the world piece on this deal based risk management. I think at the end of the day, token holders, you can't assume any knowledge nor any activity from a token holder.

And so if you're a token order, and you have to decide on risk, of a pool. And when the decision is infinitely complex, what you get is slow, less like you're some kind of resiliency because it's slow and political, but there are definitely other ways of getting this right. Like you don't have to force yourself into some kind of quote unquote, liquid democracies, just to perform such complex tasks.

Right. like more food just leaves the door open for any of those solutions to be built. And then the market to decide what it thinks it's the best. 

Gustav: So another just question here also a little bit to the same point here, right? It's like, I think one of the kind of like key struggles for, you know, like, kind of, you know, we've seen a lot of different protocols over the last six, seven years.

But it's really just kind of like build up like proper liquidity, right. Liquidity means, you know, in most cases, stable coin liquidity, right. That's like the most important thing. you know, I was, I think, I don't know the exact numbers you guys have now, but a lot of this has come from kind of like more.

You know, exotic use cases, for example, with the Athena, collaborations you guys have had, right. How do you see like Morpho competing in like a, stable coin liquidity market? Right. Like, because I believe that if you guys want to scale, of course you're going to have traditional finance coming with the collateral, but you also need somehow to get, a lot of stable coin liquidity.

What, what, envision there? what is your thoughts on that? 

Paul Frambot: Right. So obviously there is a strategic side of it that I won't be able to share, but I think you're right. stable coin liquidity is, super important, often linked with distribution.

how do you, because passive, stable coin depositors. Are usually not willing to perform complex, actions on chain, And when you look at, some of the default protocols, not just in lending, but it's very often currently coming from, big venues of distribution channels or extremely high networks, individual family offices that represent like very large style part of the pool.

And obviously all those players are Not scalable like you, you can't scale with like high net first individual because you, we just have a limited number in crypto. We have a lot of high net first individual, but not, not enough to get to thence that we should get to in, in, to compete with stratify and, and so, so yeah, and, and you know, when, when you think through how, as a distribution channel, if you want to make sure.

If you want immutable, like if you don't like strong guarantees on what's going to happen to your capital, then this is where this immutable stack makes a ton of sense, right? Like, you don't have to trust me. You don't trust staff to trust the more for that. You just have to trust the code, right? At the end of the day, like the, the, you know what those, I mean key distribution players, I'm talking like much bigger than, than, than what we have today are, are, are looking after. Right. so yeah, I think, you know, for now it's hard to compete with the, Avi brand or, or, or even the compound brand that has been here forever and is trusted by like all the, the, the billionaires in, in crypto that have some, SDC to park somewhere.

but I'm hopeful like we can compete, otherwise like, and, and, you know, and also that in time people will learn how less risky, it can be to deposit in the metamorphic vaults that is, you know, run by Gauntlet or Stakehouse or Block Analytica, where you only get exposed to three collateral. While on AVI, I don't know, you have, I don't know how many collateral and how many risk parameters, et cetera, et cetera.

And, you know, probably like 11 times more lines of code and upgrades and everything. So, yeah. And, you know, I'm confident that people like will learn over time. It's just, you know, Lindsay effect and it's going to take time. 

Juan: Paul, I want to make sure we get to this. you, we mentioned it earlier about, you know, the Morpho DAO.

I want to dive a little bit deeper into that. So there is some governance, right? I know we're shitting on governance a little bit. could you explain how Morpho governance is currently structured? you know, its role in decentralizing the protocol further and also the role of the token, which I believe is still non transferable.

Paul Frambot: Right. So I guess our general take in governance is that we want to use governance only when this is. That's the best option that we have. Right. And that we know that there is no other way of automating things or no other way of building an incentive compatible mechanism that let people or the free markets.

answer the questions that we have, for example, like risk management, et cetera. And so tokens are, and token governance is a great way of decentralizing very political, decision things. I think in particular, everything that comes down to finances is great because you can think of like token governance, more.

Like the way I think of it, honestly, it's like more of a decentralized multi sig with like a bunch of players involved and extremely resilient, et cetera. So it's just like a, a very resilient, multi sig that can, for example, like get treasuries or able to take like decisions, like turning on fee switch, you know, and, and taking a specific cut out of a protocol, et cetera, And when you think about like, more for his governance and how it's structured, it's exactly this, like taking very political decisions that it's very hard to automate, like, in particular, like, like the switch and like the treasury management, et cetera, et cetera. but the more we can put in the codes or to the free markets, the better.

In our opinion. And, and again, you know, some people have different opinions on this, but we do believe that if there is no, like governance should, should come last in the, be the last part. 

Gustav: so, so Paul, just on this topic here, right, I think like, we've kinda like in the past seen the like, you know, very like governance minimized protocols.

They have one key thing, which they struggle with, which is upgrade ability, of course. which I think, you know, like if we look at like, you know, some of like the, you know, big technologies and web three to kind of like today feels very outdated compared to, you know, what we see today. but like to kind of like this ability to like upgrade or like not upgrade.

Right. I think this is like a pretty Key part of the protocols, right? how do you then deal with this? Is this something that you guys are still able to do? can you make changes to the code or would you need to redeploy? what basically happens? how do you innovate on Morpho over the next five years, you know?

Paul Frambot: Right. So I'm a bit of an extremist, on this topic because we started with an upgradable protocol. morph optimizer is upgradable built on top of Ave, which is in compound, which are upgradable. The thing was upgradability that people don't realize until they've built like large scale products on top of upgradable protocols is that first, if you build a protocol with an upgradability pattern in mind, you will build it in such a different way than if you built it immutable.

Right. Like if you build a protocol and you have in mind that it will never change, the approach you're taking is very, very different. so that, that's one thing. The second thing is that if there is a bug in your protocol and you decide to upgrade it and change the line of code to fix the bug, then you can do it.

Have has done it a bunch of time. But how do you know that the line of code you're changing is not going to break people integrating your protocol? how can you make sure that every single smart contract that is pointing and integrating with your smart contract, you're not going to break their logic somehow, right?

You simply can't, right? It's extremely hard, infinitely hard to do. And I think the power of immutability in like giving like strong guarantees that you can build on top of a smart contract over the longterm is like. It's very, very, very powerful and understated. and it's hard to realize until like you've actually done on top of a product and of a protocol that is upgradable.

And so the way we were going to evolve about this is frankly, like the same way as Uniswap. So this is just like, okay, we've done this version, and if we have another version, it's going to be deployed and most likely immutable as well. and it's going to be like a V2, V3, et cetera, and yeah, at least people have the strong guarantees that they have, Like they don't have to trust us. All right. And then there is like, upgrade to add some, some code and add some features.

even list in favor of this, not because there was like no upgrade to fix bugs and a great to add some, inject some code and add some features. I think this is very dangerous, in general, especially in financial applications in non financial applications. I think there's great, like it can be used a lot, but I think the order is a great example of, A protocol that has a great team.

but no matter how smart is your team, how much effort you put into security, if you have too much complexity, then you get progress. And even if it's like one chance out of 100. you can't afford to have one chance out of 100 when you're Planning to run trillions of dollars and the entire world is going to, rely on one, smart contract, essentially.

Juan: Paul, last question on governance. I know we're coming up on time. so me as a, let's say I'm a more full token holder. what value do I have in being a token holder? you mentioned things like the fee switch, which may come in the future, token transferability, could be another thing that I can vote on.

I've also heard you describe owning the token as like Having ownership over the protocols network as the network effects are to start to be triggered across, Morpho. So can you maybe just quickly elaborate like what is the value as a Morpho token holder that I have? 

Paul Frambot: Great. So that's another hot take that we have on like token holders.

I know it's different from what, you know, other DeFi funders may say, but really the token for us is a way to own a share of the Morpho network. And by the Morpho network, I mean the share of the protocol states. And the protocol brand, right? What does that mean in practice? Morpho creates value by bringing together lenders and borrowers at one specific place, which is the Morpho protocol.

It creates value because if you fork the protocol, if you copy the exact same code, and replicate the exact same use case, you're not able to lend and borrow because you don't have other lenders and borrowers. What has value is the fact that all those people are regrouping at one single place and this creates value.

And so then the question is like, how do we bring people to essentially build the network, build the network early, fund the network early, contribute, like, you know, contributes, et cetera. And this is where the token. Pleasant. It's like tokens get distributed to incentivize the early growth of the network effects, right?

And then the network effects create value. And then this value gets redistributed to token owners. And it's as simple as that. We don't have like a fancy utility, like a safety module, which by the way, I'm still having a hard time, understanding how you can cover the losses or protocol when the token price is down.

Like we have simple decentralized value creation through protocol, network, and this value gets redistributed to the network itself, like through the token, essentially. So that's the general idea. And how you should think about the Morpho token, which I think is a very general mental model that can apply to many different protocols and blockchain networks in general.

so yeah, but I don't think people should expect from the Morpho token to have like very like fancy utility. it's just a way to own. Share of the network, essentially. 

Juan: Well, gentlemen, I think that's a perfect place to wrap up. Paul, what can people expect from Morpho over the next six to 12 months?

what do you want to plug? 

Paul Frambot: Right. 

Juan: So I think,

Paul Frambot: yeah, I think the Morpho protocol in general, like we're going to do our best at, improving the security efficiency of flexibility of the protocol in the coming months we'll have a bunch of updates that are going to be released in the coming months, a bunch of partnerships as well.

but, yeah, nothing I can share, at the moment. 

Juan: Awesome. 

Gustav: yeah, just wanted to, really thank you for coming on, Paul. it's really interesting to see, I think this is the first wave of new DeFi primitives. I used to work at maker back in the day.

and I've so many times seen new teams come in and try to challenge what already was going on. every single time it just seemed like a very short lift, path to try to do that. But it does, it does seem like the approach you guys have here has a lot of modes, so I'm just super excited to see where Morpho is going to be.

developing in the coming years. Well, thanks so much for joining us for this. Thank you.