DeFi-native Innovations in Lending Protocols — ​TokenBrice | The DeFi Collective

oh so welcome uh I'm talken Bryce I've been doing defi since it was born pretty much six seven years ago now there is a plenty room in front I can go pretty deep but I will try to level with you tonight but feel free to tell me if we're going a bit too harsh the idea of this presentation is to look into lending protocols and we're going to Deep dive into three of them that are doing interesting thing at least in my perspective on on Main net and at the same time um discuss essentially uh some of the challenges for Lending protocols today so uh straight into it first one so yeah we have three one is live two are soon to be released so this one is a live one and is operating pretty much on the mainset of um in the arena trying things so the life but they're changing the mechanism quite often uh but it's essentially what we call a CDP protocol collateralized debt position we'll go into this a bit more later but the core idea is you you have your own little Vault within the protocol where you post some collateral and you can borrow a stable coin from it so it has um various use cases but you know the main one is of course you are able to get some leverage you put you probably know of maker that would be a first one on liity so you put some s and you can borrow stable coins and they have um different challenges but usually the main one being that they create a stable coin um out of CA if I may not of on because it's calized but each borrower create their own stable coin as a borrow so they have a big theme of making sure the liquidity on stable coin is sufficient and the stable coin is actually worth $1 and um the other big concern for those prodor is to make sure that their stable con is always worth at least $1 the approach is to have what we call over ization and so simply put depending of the protocol the number change but essentially you have more in dollar amount of collateral that you can borrow um so this is a quite common ratio about 150% uh of dollar value of collateral compared to the B so that's just for the basics not for what diad is doing uh interesting new different than the other protocol that could Lear into something um who knows potentially valuable in the future is um the core idea they have is what I call here the liquified protocol of collateralization so what happen on those CDP protocols is because they have all the system in place to ensure that you have more collateral than you have dep you end up being having a form of capital inefficiency uh where you simply have much more cat than you have of Deb so um it's useful for the safety of the stable coin but it's also inefficient so there's a bit of a game of scale here they can push a parameter lower demanding you less collateral per dollar borrow but then they put the stable coin at risk and it's essentially yeah a game of finding the good balance well here the comes with a kind of no idea at least not seen before um where just like other protocol they would have a pretty high collateralization ratio of 50% but then they liquefy that over collateralization so meaning you have a token that is not the stable coin that's a kerosen token that represent the over collateralization of the protocol and that people can use to further secure the loan or borrow from it given they have sufficient uh um external collateral so collateral is not kerosin so that's essentially an interesting mechanism you know so if you have a large over collateralization of the protocol that tokens essentially has this uh deterministic value that grows allowing you to borrow more or secure your position more and yeah it's a bit Rocky but uh it could lead into something interesting other features that could be of not is no interest borrowing uh they're trying to stabilize uniquely with the kerosen emission uh so that's essentially the mass of that kerosen s for those that are interested but it's not so hard it's essentially how much depth you have in the protocol uh how much collateral you have so you have the over collateralization and then essentially your share of kerosen determine your share of that over collateralization so if you have $10 million of collateral in the protocol 5 million of depth you have 5 million of over collateralization and then you know if you own 10% of the Kiren you will have 500k uh of potential potential additional collateral that you can use for your position uh so this is kind of how it looks like um from from from them I think um essentially as the uh surplus of collateral growth the kerosen has a bigger deterministic value that allows uh people using the protocol to essentially leverage more out of the same amount of of initial collateral so as I was saying this is kind of experimenting Tech and um essentially what happen so on top left you have the T of the protocol in is so essentially what happened is pretty much nobody noticed for about six months then you can see a discovery let's say to 10,000 s and then it's pretty much stabilizing now where it's a bit more Rocky is the stable coin price where it used to be pretty stable but in that craziness phase it dep to the upside quite a bit and lately it's deping to the downside so they are really trying things this is the kerosen value um I'm sorry I couldn't find you a chart with the deterministic value but you have interesting Tendencies here where initially was undervalued compared to theistic value then it went massively overvalued compared to the deterministic value and now we're entering an area of undervalued compared to theistic value so it's quite new people are grasping their head around it but I think this idea of uh you know you have your cdb protocol it has excess collateral and then you find a way to enable users to tap into that excess collateralization without putting the stable coin at risk it potentially solves one of the problem which is the capital efficiency of the model uh second up is tapa so this one is not live yet um should be um potentially before the end of the year and the interesting thing here is it essentially merge together the two main model you have to build a lending protocol so those are uh the CDP that we just talk about because diet is one maker CDP model and also liquid um the key idea with a CDP is that essentially it's like an infrastructure where each user has his own little Vault he puts collateral in that Vault and he can borrow mint a St coin against this collateral the other side is money market so that's more like a or compound for those who know them and essentially it's pool based model where you put collateral people on the other side have supplied some stable coin or whatever you want to borrow and uh the the whole game of the system is to essentially make sure the PO are sufficiently supplied and it's a matter of utilization rate and so on so this is quick recap of the difference between the two um but yeah maybe just a main main one so you can follow me so uh in terms of interest rate uh on CDP protocol is very Diversified like liquidity version one has just an Initiation fee no interest rate maker an interest rate determined by governance um it can really vary but essentially it's depending on the protocol parameters and as a b you have a form of predictability uh over those interest on a money market it's completely dependent on the utilization rate of the PS so you go on ay you put some you borrow usdc on the UI you see that you borrow at 5% you're happy but if right after you one guy come in and borrow 200 million usdc and pretty much explode the utilization rate you might find yourself not paying 12% interest rate so it's really uh more PVP in that sense the uh money market um collateral it can vary pretty much also they both can pretty much do what they want but overall it's more complex to have more collateral Types on a CDP that it is on a money market where for money market it's mostly a matter of risk management and not really a matter of infrastructure challenge uh the main and core difference is really important one is what you borrow on a CDP you borrow and you mint at the same time only the stable Co of the protocol so on maker you borrow die the stable Co on liity you borrow L USD on diad you borrow diad D stable coin on a you borrow whatever you want usdc e Bitcoin what whatever is enabled as borrow you can borrow they are not minting coins for you when you borrow and so from this main difference also stems kind of the the main challenges of the protocol which is for a CDP you have users literally creating stable coins in a unpredictable manner tomorrow you could have a guy showing up with 1 billion collateral meting 750 million stable coin you need to be able to stomach him so the question is liquidity for the stable coin and making sure that um the stable coin actually stays at $1 on a money market like AI the uh the problems are more around um pool utilization as we discussed and onboarding new collateral and borrowable Assets in a manner that is safe because on an ay model the risk is shared by everyone you know if they had one bad collateral that turns into rock pole the whole Avy Market is affected no they have limits and and things to hch for that but that's something to keep in mind so for tapa and that's why it comes interesting is essentially the the two model into one so uh on the first hand they have a CDP model where you can mean the stable Co the usdo against essentially the highest quality of collateral the S the stake s and the gas tokens of the chain they're on and uh so this is what backs are stable on the USD but then at the same time they have the other component of the protocol which they called Singularity which is which is essentially an isolated uh lending Market uh where you can borrow USD or against more risky collateral so you will have users who use S or other high quality collateral using it to minan some usdo and then they can either use the usdo as such maybe they want to leverage their position or whatever or if they want they're looking for year L the stable coin they will go essentially Supply those isolated uh Landing PS um that are on boarding more risky collateral like let's say State USD format now or things like that so yeah I think this is an interesting thing I kind of anticipated that before I see them saw them going live so uh they're not live yet but they're in testet but I'm curious to see where it goes I think it can actually um scale better both sides so it gives an edge to scale the stable coin the CDP side because now you have a native year Source on the stable coin and on the money market it ensures you have constant fresh supply of stable coin which is kind of the challenge for for the AV and compound of the world they have all interesting idea want to keep it Focus mostly on this merging of the two main model but just to go quickly they use an option model for the Covance token so essentially any form of liquidity mining or whatsoever you don't get the raw token itself you get an option to buy the token at a discount 50% or in that range the discount varies uh but it's a way for them to essentially have external Capital that they can then use for uh protocol on liquidity so the idea is the model is designed to capture the liquidity over their own stable coin over time so as a protocol is used they ACR some usdo and pairing assets and then the protocol itself will be a big supplier of those pair so it's a way to ensure that you have a solid Baseline of liquidity on usdo which if you remember from the CDP slide it's a big big focus and matter for U cdb protocols uh other things to note uh without going too deep is they are building using layer zero Omni chain fible token standard so this allows from some really cool things like you have on arbitrum you're borrowing usdo on optimism if you want and um pretty much essentially more interoperability between the chains and um for the token they have um Revenue sharing to the Token U with a locking system that is a bit uh a bit more complex to tww twap so I won't get too much into it but um interesting to not because it's rare to see protocol of potentially this scale sharing Revenue directly so third one is uh liquid two B stated for November so in a botom month if all goes well and here uh same thing I'll focus on the main ideas the one that are the more uh challenging let's say compared to what we used to and so I want to tell you about uh user defined interest rate because that's the first in defi uh and uh a few other things but mostly that so the core idea here is that the borrowers will get to decide his interest rate within a range with a very low minimum of like let's say 05% and pretty much uh I think a maximum in the 50% range whatsoever uh it has consequences this choice of interest rate we we'll get into this in the next slide uh there is another interesting thing which is this uh interest rate management can um be quite demanding for the user that must be active decide to which interest rate to put to pick when to change it and so on so they have also this concept of interest rate delegation where you can have your own position where you borrow but you essentially delegate the management of your interest rate to another entity like let's say defi saver for instance uh the other interesting things of not is uh what I put here as self-contain and essentially all the fees collected by the protocol are redistributed within the protocol to incentivize different behaviors there is no external token capturing something uh and one of this um area where those fees those interest rate collected are redirected is um for the liquidity of the stable coin again obviously a big concern and they have an interesting mechanism here we go into so uh quick overview should know by now but essentially maker is a CDP interest rate is defined by governance maybe you borrow at 5% and then you know a be dead guy in the middle of Nevada wake up on a bad foot and no thanks to a governance vot you know pay 6% interest on ay we talked about the situation it's only dependent on poor us AG so you don't really have predictability on home you pay here on liity will be set by users uh and so you decide to pick you get to to pick how much interest you want to pay uh of course there is a consequence because you know else everybody will be borrowing at 05% yearly and be happy so um what happens is um the B stable Co the stable Co outputed by the protocol has a Redemption mechanism against the is used as collateral within the protocol so from protocol perspective what this means is essentially there is a way to Arbitrage a stable coin is a stable coin is for whatever reason say trading at 98 cents you can buy that stable con at 98 cents and redeem it for $1 off p as you do so you buy a lot of bold and so you push the price up and you bring it back to Peg uh for the borrower though this has consequences because a Redemption if you are the one that is redeemed against it's like a repayment of your depth but you don't get to decide when that happen it's just you wake up one morning you have less collateral and you have less depth so the way this decision is made is based on interest rate essentially the lower your interest rate the more likely you are to be redeemed the Redemption goes in order of interest rate so if you accept to pay a higher interest rate you Shield yourself from those redemptions if you pay a lower interest rate you are taking more risk of getting redeemed but you also pay less interest rates so it's kind of a choice to be made uh so yeah that's kind of what this side say front f it but yeah it shows to you visually how it would go essentially you know there is 1 million of B being redeemed it will go in order of interest rate paid by the borrowers uh until it has 1 million uh to to redeem against so obviously the higher the interest rate the safer you are from Redemption but then you pay more interest rate so it's a it's a game of finding balance uh now for the other interesting part is the revenue flows on the protocol so you have this uh interest rate collected on on the the borrowing for a space of virus interest rate and then the protocol has it redistribution system for um the ball coming from this interest rate collected so three4 of them go to the stability P which is essentially um the main liquidation mechanism of the protocol if there's liquidation to be made no external liquidity it's needed it's simply taken from the stability pool and the depositor can potentially and E it allows for fast and smooth liquidation you can dive into liquid V1 it's similar mechanism if you want to understand exactly how it works uh the other more interesting part is this uh liquidity initiative things where essentially you have a force of the interest rate collected at the protocol level that can be directed uh by the liquid uh stakers to various liquidity initiatives and liquidity initiatives are quite flexible they essentially a contract that would receive B so they can be a liquidity mining contract for a Unis swap Bal usdc pool or a curve pool or they can even be uh other protocols that are doing collaborations at the stable con level or even um initiatives like uh defi Collective where will be doing things to help the liquidity of Bal on layer to um so a cool feature of that model is that uh because you get 75% of the interest going to stability Po and the 25% remaining for initiatives it means that essentially uh as long as you have more than 25% of the B minted overall that are not in stability pool the passway you just mint at average uh interest rate and you supply the stability p is already profitable more or less but profitable and so this is a interesting property say compared to what you have on a money market because on the money market you have a spread so on IV for instance you might all have seen if you supply usdc IV will give you maybe 8% y but you see that on the other side the borrower paying 99.5% for instance this spread here doesn't exist and it can actually turn into an amplification depending of the share of B inst stability P um yeah it's kind of the overview of the whole system so you have the T of the Vault this is your own little Vault where you put SS St you can borrow Bond there's interest payment to the stability po the liquidity initiatives and then if there are liquidations the uh B needed for that liquidation will come from stability por and stability POR depositors in exchange and E so it's like essentially a way to buy the dip depending on if there is another dip after or not it might be profitable or so this is how it articulates over all all together in situations of over Peg above or under Peg so if you have an over Peg of the stable coin it goes to $1.1 for instance $ 1.01 then essentially uh the Redemption risk is pretty much nullified because Redemption is an Arbitrage mechanism when it's under Peg so what happen is borrowers will obviously lower the interest rate because why why pick a high interest rate where have no risk of getting redeemed anyway as they lower the interest rate the yield of the stability pool and the liquidity initiatives decreases which reduces the demand for B and therefore push a price down as people will be selling B let's say the sell B into a oversold territory and we go to 099 or 098 and the price decreases you have a higher risk of redemption so first we have Redemption that keeps clean and helps push the price up but as you have just the risk of redemption people will raise the interest rate to protect themselves from Redemption as they do they increase the UN of stability pool and liquidity provider therefore creating more demand for B therefore pushing the price up so you can have a nice self-contained self-balancing system that uh could spin into Infinity at least that's kind of the goal so yeah I don't know how long I was I hope not too much but if you have questions uh I'm listening you have a question question please repeat question because of microphone so when someone ask question please repeat I all right tell me um you definitely explained this I just didn't understand it so what does tap let you do that you can't do right now or what does it make easier exactly uhhuh so you yeah so what's unique with tapa repeat the question yeah what's unique what's special with tapa as a prodal uh from a user perspective what's really cool is this like natively Omni chain you know this idea of you have collateral and arbitrum you want to borrow on optimism or you can repay dep from any network without having to bre because the USD stable coin can essentially be teleported from one chain to another because it's a omn fible token using the the target standard so I guess that's the main benefit for for the users on top of you know potential yield may be interesting and so on protocol Lev uh my perspective it has a decent capacity to scale essentially because it's merging those two model together where essentially this money market side creates demand for the stable coin creat by the CDP and so both can grow in harmony if they find a good balance so um yeah it could be a a protocol AV size if one goes well for them that's why it's interesting for me because it's always you know the challenge in those Protocols are it's easy to have a protocol that works well you have 10 million borrow 20 million collateral we have 100 of these but a protocol that has like 20 b of TBR like AB we have only one of these great thank you any other questions looks like you explained everything Perfect all right thank you guys