Morpho Labs CEO Examines Drawbacks in Current DeFi Risk Management

Transcript

I guess you know we had two problems with you know the first iteration of defi first iteration of defi breaks and it's still remains within the circle of crypto for crypto users for high networks individual def phones Etc if we're able to get a defi that is safe truly safe right and I'm talking about orders of magnitude and that is sufficiently General to tackle all kinds of use cases not just one specific risk profile or compliance profile but any risk any compliance profile we're going to progressively reach Mass [Music]

adoption hello and welcome to the protocol podcast I'm Brad C here with my co-hosts Margot nykerk and Sam Kessler excited to dive into Today's Show with the latest news and developments and technology behind crypto and blockchains first please do not forget to subscribe to our Weekly Newsletter the protocol on coindesk.com and let's dive right into it we've got a pretty incredible guest today here Paul from boo he's co-founder and CEO of Moro Labs a research and development company responsible for building and growing the moro protocol which is you know at the Forefront of kind of this evolving and crazy you know

landscape of Defi and just sort of changes every week or month um and we're gonna go deep with Paul a little bit later um but first we're gonna talk about uh eth Denver which actually all three of y'all were out there um Sam Margo both moderating panels and Paul uh was out there just is now back in Paris mark ago you know you had a pretty interesting panel and you've also been writing about the Denon upgrade on ethereum which is coming next week which is pretty amazing just to how fast this stuff unfolds of course like everybody

talking about Bitcoin alltime high this week is the dominant story but you know it's just like the news in crypto never sleeps Margot you know tell us what were what were some of your experiences out at at eth Denver last week and some of the panels that you moderated yeah I mean what was really noticeable to me is that the presence of like layer 2os and like da Solutions and a lot of infrastructure layer like layer solutions they they were all there they were roaming the halls and what I also noticed is that like the eth core dev's team was not really present so I think that was like an interesting observation just because like it was almost like they were competing and fighting for like marketing and users and things like

that and of course like everyone was talking about um what dun means because it's like right around the corner and you know layer twos are at the center of of that they're going to be benefiting the most from that because of Proto D sharting which will lessen like transaction fees uh eventually for users of layer twos but primarily for like da data availability for Layer Two um so that was really interesting to sort of see what their take was I think like high level overall all the teams sort of agreed that it's all speculative at this moment like we don't really know what that like change will look like post canun but that I think there's going to be some kind of War you know between these layer two teams uh for you know who can deliver the best like price point in terms of you know transacting

on these layer twos and and the cheapest price point let's say and they all had different takes like it all really depends on the economics um of these layer twos in terms of like how much data they're willing to consume from layer one or a layer two so obviously you know last week uh we had Jesse poock on uh of Base coin base's Bas layer talking about kind of his breakdown of the of the fees and how those might shake out post denune um and Margo actually has a really good story out today uh talking to some of the other teams at arbitrum and polygon and starware um and that is on coindesk.com so check that out um and Paul you know I'm curious what were some of your

highlights or any surprises anything that like you did didn't know before or it's anything that really kind of changed your views on things no I think look that was my first sist ever uh it was the first time I went there and I genuinely uh really loved it you know like the quality of the conversations I had there uh were were extremely high and um I guess you know the the one narrative that I would you know keep and that has emerged from from this conference for me is uh you know the panel I I gave is unisoft uh and maker uh about you know Finance being moving from apps to to

infrastructure and essentially how we're rebuilding Finance like the internet right instead of having like mon Financial products on chain we're more having like a layered sort of like structured approach and I've been discussing a lot because like the unisoft team was the maker team during the events and a bunch of other teams as well everything seems to be converging in that direction and that's also the case in like socials when you think of it like f foraster when you see like how the protocol has been built and uh layered as well uh just still like we're getting there uh and we're you know truly rebuilding the internet and extending the internet uh by

transferring data that is now valuable thanks to the the properties of blockchain it was the first time for me like the three years that I'm working in the space I could still concretely uh that we're deeply transforming the current state of the internet today uh and not just you know building you know shortterm e FM Ro products uh it's like true deep infrastructure that we supports like very like you know massive grow in adoption I think that's really interesting I mean I guess if the concept is you know a lot of people talk about um you know what blockchain is doing is kind of Reinventing

Finance but maybe if this is what you're saying Paul is just kind of thinking about it as the next it's the next iteration of Internet development yes um exactly so the way we think about it is in the early days especially in the context of defi think the first experiments were about building Financial products on chain right how you know in in in traditional Finance we have FS okay how do we bring them in chain and make it decentralized with a smart contract with a decentralized governance and you know this is what compound inovated on for example by having you know one one big pool of capital where people can deposit and then the fund gets managed you know

in a decentralized way and and you know forming very great financial products but then you know as soon as you want to scale and you want to add more assets or more features well you kind of have problems because you have to Trin all the complexity in One Financial product and eventually if you keep that road with that route well most likely you're you're going to end up you know taking too much risk which which happened in 2022 and 2023 essentially and the the the new approach that we have now and that's you know uh many protocols actually most Protocols are looking into is doing much less to be able to enable more on top and instead of BU Financial products building Financial Wells

infrastructure on top of which you know Financial products can be built right and at the end of the day this is how you know kind of the internet has been built uh around networking protocols so if you think about SMTP or HTTP those are like very unopinionated protocols that are Global you know they don't have any opinion about the web page that is contained in HTTP or the compliance or the regulations or or anything it's just like a very pure simple and scientific objects on top of which you can you know rebuild abstraction layers until you get to this very nice Fen web page on Coes for example that completely abstracts away the Internet Protocol and and I

think this this is the same route for for like defi is like having extremely minimalistic simple and fundamental Primitives on top of which you rebuild you know maybe a risk management a a user abstraction ER gas abstraction cross chain abstraction etc etc until we reach like you know uh mass mass adoption that make sense yeah if I can chime in just on what you were talking about around how these um how we're kind of rebuilding this infrastructure layer one thing that I noted at eth Denver and I wrote an article about this that we published yesterday the day before I forget um I I I noticed um kind of like you were saying Marco there were a lot of teams there it wasn't just layer twos but it was also um a sensibly competitors to ethereum in Bitcoin

salana and so on and it does feel like ethereum um you could interpret this in in two ways one is maybe ethereum is seeding its lead in the um L1 race in the layer One race it's you know lagging behind maybe some of these other competing platforms data availability layers like Celestia and so on in in terms of providing certain functions but you could also look at it as ethereum widening its umbrella because we are reaching a reality where the apps that we're talking about are built across several different ecosystems so I wonder if if you have any thoughts from what you saw at East Denver around what this means for ethereum which until now has by far been the strongest largest um blockchain in terms of developer

ecosystem liquidity and and so on and so forth as far as smart contracts are concerned right so I think the very you know powerful thing about Assa in terms of community and ecosystem has been the evm and all the code that has been dealt already especially the safe code that has been bued times and times again that's a lot of money that been that hasn't pour into an e system that sort of like strengthens strengthens the uh the network effects of theum and makes it extremely hard for teams like hours to consider like going to rust for example you should require having completely new Talent base completely new audits etc etc however this has

pushed like competitors of eum to take very op pated BS to be able to challenge like you know the current state and I think salana specifically had been extremely like to brilliant at taking those deaths whether it's like okay launching a phone or taking you know Sala with you know with DVM and rust which are you know very concentrated uh concentrated bets as well as the way they're thinking about scaling and approaching uh the long-term vision of sonana which is like more of this Global State machine whereas you knowm is more this ler to uh Vision where the state is a bit you know fragmented and uh but you

know scales on top of this very unopinionated base a and I just feel like um the competition right now is super healthy because we cannot have the feeling I would say like maybe a year or two ago that the serum uh you know Community was a bit relying a little bit on on you know uh it's previous success uh well you could tell in the last six months that you know is your researchers and Community you know in my opinion I I have the feeling that things are accelerating because they see that there is pressure coming from like there is there are other opportunities out there uh so not having probably the the leading one at the moment and um yeah you know overall I just feel like that's

extremely healthy competition for this space that is super interesting Paul I mean that you know the the the the network effects um but that you know that point you made about how there is a lot of money that and time that's been spent on kind of perfecting you know the these early movers you know uh that it's you know once once those Network effects are sort of establish it is does seem to get much more difficult to knock them off uh and you just kind of got to go with the flow if you're jumping in now all right let's let's go to the Break um and then

when we come back Sam is going to lead a discussion into some of the stuff that he's been writing about and talking to Paul about uh with sort of the the new look of of defi um okay we'll be right back calling all developers score a consensus 2024 developer pass for just $19 but Act Fast only a limited number of these passes are available you may have heard that consensus ain't for devs but here's why you're wrong consensus is the only place you can fully immerse yourself in a multi-chain environment and learn directly from 20 plus chains including arbitrum chain link salana and

more enjoy 3 days of intensive learning with technical talks 40 plus expert speakers and 20 plus in-depth workshops including dedicated half days for ethereum and Bitcoin and three full days of programming on our protocol Village stage consensus 2024 is happening May 29th through 31st in Austin Texas don't miss your chance to network at curated developer meetups find new career opportunities and explore hundreds of side events and hacker houses around town grab your $19 developer pass today but remember this exclusive offer is limited visit consensus. coindesk.com now to secure

your developer pass before they're gone explore the epicenter of blockchain innovation at consensus 2024 and we're back uh with Paul fro of Moro and uh Sam I'm just going to turn it over to you um to lead this discussion and this part this portion of the of the show on don't she TI it yeah um yeah thanks Brad so so Paul I I kind of want to just get into a discussion um with all of us here about defi what you're building um and what it says um you guys have a thesis about the the state of dii and particularly risk

management on defi that um is ambitious um but also um you know you'll have a lot of people I think disagreeing with your and we've talked about that as well and I want to kind of get into it so Paul how about you um explain to us the the conventional model on um in crypto we we use we talk about T5 we talked about lending can you uh briefly if possible like just tell us how these you know incumbents like a and compound work how they manage people's money when they deposit into those systems right so so the way we think about defi is that you may you have two different kinds of protocols uh you have actual protocols and you have what we

call unchain fonts and really the way to think about having compounds is like onch funds right yeah the mor okay yeah the mor for like you know the mor for and uh so so yeah like so arent Farms are essentially like um large PS of capital are managed right and managed in a decentralized way uh so in the case of compound and a essentially you have like one one one p of capital where everybody is going to deposit their money and it's going to be super passive for users right they don't get to uh you know do risk management themselves they don't get to care about

user experience it's a very know flexible uh inconvenient way of lending and borrowing right and um and essentially this requires uh Management in the background to make sure that you know the money is not going to be lost that the interest that are generates uh you know come from proper collateral uh that you know the collateral is safe that the code is safe Etc and so this Management in the background is handled the following way essentially the Avid da so token ERS are voting and and compound them as well are voting to basically list new collateral adding you know maybe uh DTC Asal or mkr Asal to enable borrowers

to to to borrow money from the large to of capital that is common and Global to everybody and um essentially token orders have to decides on collateral liquidation parameters uh collateralization factors Supply gaps borrow gaps close factors you have lot of those parameters essentially on a you have 700 of them right and those are fair Fairly nont two sets you know should you send the wbtc liquidation NTV at 85% or 82.6 really you know it's it's it's not that Tri there's a lot of data and simulation to be done in the background and obviously toen orders are usually

like you know you you quite passive to be frank and they don't necessarily are knowledgeable about risk management uh because it could be anyone like you know I could go to deck right now and and and buy a tokens and so the Theses we have is that uh you know this first is does not seem right to us and obviously you know in practice they don't know how to set those parameters so they're going to contract with external parties that are going to tell them what's the way of of of doing fs and so what the avidar has been doing over the last you know four years is Contracting with uh players like gaunlet or chaos that are going to make proposals and are going to propose like Risk parameters so you know moving

STS from 72 to 75.6 whatever on you know on the daily basis they're going to you know propose those different changes and then every day like the talken ERS are going to look at the current proposals and they're going to approve it or disapprove it they have something to say but usually it's you know 100% of the time gr lighted VAR cases was not the case for different reasons if I can pause here and maybe like summarize what I'm hearing so the a model just at at a um you know at a fundamental level is just people can take money they can supply these pools or they can borrow from these pools and the a da a bunch of a token holders are tasked with managing the risk of those pools they're tasked

with voting on whether they should increase decrease Supply borrow cap so on and so forth and they defer to these um firms Gauntlet we'll talk about this no longer but that was one of them chaos is another one that's still there they defer to them on many of these um decisions they they being the Dow they ultimately vote but they use these Risk Managers that they contract that they pay to make decisions and you think that this is um you know a problematic system can you talk about like what are the the risk I I mean maybe to to frame this in the context of like events there was this big thing that happened over the summer when we first met Paul I reached out to you because curve um was this like there was this big D5 blowup that happened and it looked like um a a a

huge pool on a was at risk it looked like they were lending out to or they were accepting too much of a certain token for um a certain kind of collateral that's why we were talking about things I wonder if in the context of something like that why you think this model um that we're talking about is not the right model right so maybe first I think it's worth mentioning that those decision that token ERS have to take well they have no no actual proof about how the risk parameters are computed right so downlet and and Chaos uh were paid by the avidar and in order to be able to you know to charge a fee uh the algorithms that they're using internally are close Source right so the

token order has no way of reconstructing how you know the risk parameter has been decided how it has been computed because if chaos was to or gantlet were to you know opensource their model where they will not be able to charge such a high PR that's the sort of like the idea just clarify Gauntlet and Chaos they'll make a proposal for hey we should do this to this limit we should increase this cap we should decrease this thing and to back up their proposal they will site research that they've done and you're saying that the research that they've done the methodology that they've used is opaque so the Dow is voting based on like how much they trust these people not how much they trust the metrics to

be clear like the methodology is open but the algorithm is close Source right can you explain to me the difference between um you know you said their methodology is open source but the algorithms are closed Source like briefly does that just mean that the um you know we we know at a high level what they're doing but we don't know exactly how they compute these numbers exactly right right because e offens all their mosts well you know economically for them you know it just does not make sense because you know most likely this is going to go get lot of negotiation power to beow to be able to lower the fees because it would you know anyone could run the algorthm and don't need essentially chaos or or gaet anymore because everything's happens first essentially we we as more for you

know like to call those like unchain funds because the protocol is managed it's not the fact that it's close Source or you know it's it's open source or or it's just the idea that inside the protocol there is this management component that's going to change you know the risk model that if this management goes wrong the instance is going to you know could potentially break and go bankrupt uh which is not the case of UN right there is nothing in the case of unop that can there is no concept of management the protocol cannot go bankrupt it's just you know a pure protocol you know with like few lines of code that assemble like a very uh Financial perative like and so this is how we make the the the the essentially like enshrines or

externalized risk management so what moro's doing is You're Building lending pools just like a except you're saying that instead of a Down man in Risk we'll have these risk curators manage the risk where it could be a dow it could also just be um it could be a dow it could also just be Gauntlet or whomever else one of these risk stewards so I'm curious obviously there's a lot of advantages to that kind of a system it takes away a lot of bloat it can make things more efficient we we didn't get into all the disadvantages of the a thing um in terms of your viewpoint on how those systems work but um there's a lot of disadvantages there and that's what you're trying to solve but with your system um I think people will come and be like hey what's the point of defi if um and I know you'll

have an answer to this but like what is the point of defi if you have you know a pool that is managed centrally just like in our traditional Financial ecosystem right so so first morha blue is doing much less than a the sense that it is just the Ling and boring engine so that's 600 lines of code completely inable super efficient whereas like a is doing like is more 7,000 lines of code doing risk management and training the protocol but so what Mor blue is is more an infrastructure protocol on top of which you could rebuild those you know high like instances with the mor blue the mor blue prent and so the idea is

that Mor does not externalizes completely risk management and does not have any opinion on how risk management should be done right maybe uh you know risk management should be toen based and decent centralized through toen government right that that could be a way or maybe it should be like a completely centralized you know SEC registered fund or maybe it's a mixed of both and mor does not have any opinion on that and it would just let the you know the free market decide what users will consider you know the safest and the most convenient for their use case and does not have to be one single winner by the way some those you know will will have their own Landing pool and maybe they

want to keep it doing you know the a way and some others like were withing with like gunet or RDA or block are experimenting ense they're mainly taking the task of like Automation and trying to remove completely the human components and through maybe you know like more Z oriented technology try to automate as much as they can but also you know you have approaches like I think steakhous is doing this and probably block analytica as well where the users themselves not the topen ERS but the users themselves can veto uh the risk management decisions so the design space is huge and to be frank I I don't pretend I know uh what's going to be the

best solution out of this I just know that you know the market will eventually decide and I think it reminds me of a a quote a tweet from Aden from unop and the way he thinks about you know how to decentralize F and he said like there are really like three uh different ways of doing it and the most powerful one is code like smart contract code the second is incenses and the third and last is government right and so you should always favor code first and this is what Mor BL is doing a completely inable code if you conso it through code well maybe the incentives is the best way of doing it so no Marketplace is you know uh probably better suited than one mod G

gar Po and then finally there is if the decision you know making is too political say for example activating a fe switch that fly be automated through code it's very hard to have you know a free market decid what what what is the fee switch of a dow that you know should be taken and so in that case you know when it's natur Poli it makes sense to decentralize with the token uh in my opinion Paul I mean this is so fascinating you know I mean one of the things that we focus on in this um podcast is you know we we talk a lot about the infrastructure and the the uh you know the various blockchains and the

the race for relevance and Supremacy it's so fascinating to hear you talk about you know you're you're really working on you know the protocols a top of all of this you know and just to hear how you talk about it the 600 simple lines of code and real quick I think we got to wrap it up here I feel like we could talk to you for a really long time but um you know defi summer 2020 and then we've gone through these waves of defi Evolution over the past few years and now here we are in 2024 you know do you have a sense in your mind of kind of when we're going to see maybe the ultimate evolution of of

what defi is going to look like is that going to emerge this year is that a few years down the road several years down the road I don't know what what do you than real quick yeah so I guess you know we had two problems with you know the first situation of defi first situation of defi brains and it still remains within the circuit of crypto for crypto users for high networks individual def funds Etc if we're able to get a defi that is safe truly safe right and I'm talking about orders of magnitude and that is sufficiently General to tackle all kinds of use cases not just one specific risk profile or compliance profile that's any risk any compliance profile well once we have those base layers in place then it's just a matter

of like how many Builders are going to build on top and on top of those Builders two layers of obstructions we're going to progressively reach massive option and you know this I'm not able to tell when exactly will happen but I think when you look at you know unisoft probing unisoft before I think around Q3 maker progressively deploying maker core this year and more for blue going l just few weeks ago uh I think you know um probably end of the year will be a time where you know all those different new models will will start emerging in my opinion so that's that's I think by my quick calculation that is five years in crypto time uh if you assume that one

year is seven years okay all right well uh Sam thank you so much I think we got to wrap it uh we're getting a nudge from uh Michelle our producer but anyway um Paul thank you so much uh for joining us it's really been fascinating I hope this will be the first of many conversations um Margo and Sam thanks thanks to Michelle uh and to our Booker M mananas and that is it for this week thank you for listening to the protocol podcast if you have any questions about any stories or comments please reach out to us at podcast at coindesk.com subject line the protocol you can listen to US Weekly on coindesk podcast Network or wherever you get your podcasts also please subscribe to our

Weekly Newsletter the protocol on coindesk.com thank you and see you next week