The Expanding Spectrum of Stablecoin Yield Sources in DeFi and Its Risk Implications

Transcript

Yeah, okay. Uh um Hello everyone. Is it working? It is. Okay. Hello everyone. Um welcome. My name is Nicola. I'm coming from Block Analytica team. We're doing risk management mostly in DeFi. Um we are the risk team behind Sky Protocol, formerly known as Maker. Um and today this is not going to be a talk of shilling any product. We're going to uh see the risk implications of something that has been happening for year, year and a half, let's say, uh within DeFi and that is the increasing number of um off-chain yields yield sources being tokenized and presented uh on chain.

Uh okay, I think I would need to do this. Okay. Uh >> [clears throat] >> So, the agenda will look like this. Uh The first thing we're going to go through is basically just uh see which yield sources we're talking about. So, um the tokens that are essentially powered by uh off-chain yield sources and what is backing them. And then what expectations we as a risk curator, uh as well as uh the potential investors or suppliers, uh are expecting our you know, disclosure expectations and some of the best practices we we think uh we should push for the the same way, for example, we

kind of had a push like around 2 years ago uh when it comes to centralized exchanges and their their solvency proof of reserve. Then we're going to try to tackle the second problem. So, the first one definitely being transparency of those yield sources. Then the second one being pricing of RWAs um and the redemption process of those uh on DeFi in DeFi. Um And then to wrap up, we're going to just briefly mention some of the native yield sources uh on the Ethereum protocol level we could explore more, arguably, uh that can kind of uh limit the the exposure to the increasing number of off-chain yields

we're seeing in DeFi that um without a doubt increase the overall risk. Okay, so obviously, probably a lot of you may know about Athena, which is a project that kind of was one of the first to successfully and um tokenize uh strategy that requires active management, which is in this case basis trade. So, they kind of proved that this kind of approach to tokenizing such a strategy is uh scalable, first of all, and then um that it's modular, meaning if you tokenize it rather than have a do-it-yourself approach, you allow for um that token to being used as

collateral, so to borrow against it. You allow for um risk trenching on top of uh this kind of asset. You allow for yield speculation, for example, as Pendle Protocol offers. Uh and of course, you can even short like the whole strategy or the execution team behind it by by shorting the the token. Um then after that came basically a wave of new tokenized projects tokenizing different kind of strategies. So, for example, now we have um tokenized Odyssey arbitrage, uh AI infrastructure financing, which is being tokenized.

Credit credit lines in DeFi, which is also something we've seen recently. Um And then uh even more increasingly and now highly adopted recently in DeFi, which is tokenized private credit funds. Um and even tokenized hedge funds now. Now, all those all those uh projects uh this is all the kind of been an interesting um to see, to to track is that the yield how those projects that tokenize such strategies do the yield distribution.

And the the realization was that a lot of them tend to take the very similar approach of what what something that Sky Protocol actually and at that time Maker did, and which is the PSM approach. So, basically having a single smart contract that acts like okay, you deposit uh it supports a deposit token such as USD USDC or USDT or uh some else. And then you get back the receipt token of the protocol, which is basically uh the token of the of the strategy itself. And then on top of that, you need to kind of stake it to to receive a yield-bearing stablecoin. So, the approach is similar to like what what Sky's has with S uh sorry, USDs and SUSDs.

So, similarly to Athena to Maple Syrup, uh but of course, the backing of those tokens, the transparency, the redemption process, uh it's uh completely different and should not be taken for granted. And yeah, today we're basically going to just raise some I don't want to say concerns. I want to say things that should probably um we should be probably more vocal about when it comes to showing this data. Like to to to investors or to anyone using using them using DeFi.

Okay, so as we may know, like the average DeFi user from uh before 2 years ago was really just used to having everything they can put their money into highly verifiable. Uh so, everything was on chain, everything was fully disclosed, and we even didn't ask question of okay, is this really backed or is this really uh delta neutral or is this really market neutral as a strategy? And then uh came those tokenized strategies. So, those teams that execute those and of course, us as a risk managers and investors and potential suppliers to to those projects needed to come up with a

solution um to basically minimize the trust you put in such such tokens. So, here I just wanted to give basically a waterfall of best practices of what you as a potential supplier or user um uh should expect. And uh to the teams that are doing this, what they should present uh to their users. So, I categorized it in three categories, basically. The first one being you need to know who is your counterparty. So, you need to ask for them to disclose incentives and any relative uh parties, so you avoid any conflict of interest. Um And then the second thing is

rehypothecation and admin risk. So, of course, this is what you want to be limited, of course, maybe even no rehypothecation ideally. What I think um the second category is actually addressable uh by the third one um with proper transparency, for which I wanted to comment. So, this is basically a waterfall of transparency best practices. So, the first one is obviously having everything public. So, that's that's the best approach, right? But this is usually not uh feasible by those kind of of um projects. Whether that's because of the the risk of being front run, uh whether it's maybe even security, uh maybe even competitiveness.

Uh so, not uh not um possible always. Uh but the best solution, of course. Then the second thing you should uh consider is ZK uh verification networks. So basically you could allow uh or enable your investors that in any moment they can do that they can generate a ZK proof themselves uh that uh the project you are considering into investing is whatever you need need it to be like delta neutral, uh market neutral, transparent, backed. Okay?

Um Then the third thing um is oracle attestation. So, here the trust is just kind of migrated to the oracle network itself. Um but yet still better, you know, from the if you go down the line. Then auditors. So, the next topic would be okay, now we there is an option we have multiple independent third party attesters uh to which we are giving our data um and they kind of provide us attestations to increase the confidence in the in the potential depositors.

Um but yeah, here again, the trust is just put to a certain parties. And of course, as much frequent those attestations are and as much attesters they are, um the better. Uh but still there is some level of trust. Then self-reporting. This is actually what we have been you know, seeing across DeFi um mostly. So, projects running those kind of strategies also just um do the attestations themselves. So, trust is just stayed with the same with the same counterparty or issue of trust.

And then obviously, you can do just nothing. So, there is a blind trust. Now, the second uh issue with the like the upcoming off-chain yield sources are the RWAs and two things related to them. So, the first one is um pricing of those. Where curators, if they want to, let's say, deploy a lending market, they in most cases don't have other option rather than using a redemption rate price.

And the the reason is very simple because most of those are still very almost illiquid on chain. So, very low liquidity or no liquidity. So, you can't really rely on any market rate to to to price the asset as collateral within a specific lending protocol. So, what happens or what could happen like a let's say a risky turn of events is that okay, you use the the the RWAs are illiquid on chain. You use the redemption rate pricing and then on top of that you get some fancy loan structuring or yield product that is not very transparent. So, you basically

have end users aping into the such a project such a product without or with very [clears throat] little knowledge about what is powering it as as an underlying yield source. And then, let's say we have an RWA depeg. So, this world asset loses value. So, what happens? Well, bad things happen usually. So, holders of this RWA lose proportionally to to the the loss of the the amount of depeg.

Then borrowers are kind of okay. They may even profit cuz they're literally on the other side of the trade. Loopers of such an asset get their equity trapped within the lending protocol. So, it doesn't really matter if they're solvent or not anymore because they simply can't get out since the redemption of such an asset mostly happens off chain. So, they're kind of dependent on that. And then lenders, I think, are actually the ones that lose lose the most because you're lending against an asset with a fixed oracle.

And then you basically lose as much as the borrowers profit. But there is another layer of potential loss, which is a bad debt. So, um I think lenders are the party that is that loses the most here. So, to tackle this, recently, and this is actually very recently, there were some ideas of non-atomic unwinds. So, now we're speaking about if someone wants to unwind a leverage RWA position assuming even that everything is okay, it's not has not depegged.

This is still an issue, right? Because those are essentially not liquid on chain. So, you would need to find a way to finance this kind of unwinding. Usually in DeFi, what we've been used to that unwinding is atomic, right? So, you have but that, of course, requires on chain liquidity. So, what are the options? The first one we saw was financing via Morpho. So, this is kind of like just-in-time liquidity. So, and this is what Free F is doing. So, they get a loan out and if someone wants to unwind a position, they use that kind of liquidity and deal with it afterwards. And here actually Morpho lenders are acting as unwriters. So, if actually something

goes bad they're the ones who who take the loss. The second one is credit lines as a standing liquidity. So, the difference is this is not just-in-time liquidity. You you open a credit line and there is actually standing liquidity acting as unwinding liquidity, so to so to speak. Here in this specific case, credit lines were taken from the Cap product Cap Money protocol. And basically the risk takers on the on the protocol as acting as unwriters. Then there is an option of direct collateral seizing. So, this is what Silo, for example, is doing and that's collateral just automatically goes to the lenders and they they bear the risk from that point.

Another just a spark discussion. So, there there could be more complex systems and this is probably something we are going to see in the near future for async redemptions like limited duration risk if you find a way to automatically hedge your position if you're providing liquidity for unwinding an RWA position. But yeah, maybe maybe too soon um to to discuss that that part. Yeah, now so um to to wrap up, I think this is also very important and that is to explore another native on chain yield sources not to get too compressed by the

the off chain the increasing number of off chain yield sources that basically require off chain redemption process. And those being on the on the Ethereum protocol level. So, what we have done so far and what DeFi has as basically a base rate for ETH across all the lending markets for years now is transaction ordering being financialized, right? So, we have a staking yield and this is the part of this trifecta that has actually been massively financialized and acted as a as a good yield source.

And the second two things are storage and compute. So, compute being computation, you know, required to actually execute the smart contracts on Ethereum. And storage meaning the permanent storage that Ethereum offers or data availability, if you may. And these two parts are actually something that we are yet to financialize. So, let's say, and I think this is actually what also the EF the Ethereum Foundation has been vocal about. So, I I'm kind of echoing that.

Um So, currently the infrastructure like L2s are kind of um like the old revenue from the from the compute is going there. But for example, this can kind of be tokenized as revenue from from L2s and then again use used with across DeFi as collateral. Then, for example, there could be gas futures to buy, sell, and hedge computation costs in advance. And then probably the most interesting one is if we actually see those AI agents running most of the executing most of the transactions on the

on protocols such as Ethereum, I think it would be the the importance of data persistence, data availability, and the execution or computation would become something they would need more than the transaction ordering. So, this is probably something that has a potential to be to be financialized. So, yeah, I think I think I covered it all. Yes thanks. >> [applause] >> I'm not sure if we have any time for questions. Yeah, 2 minutes are free.

Okay. Yeah, if any questions, feel free. Okay, great. Thank you.