DAPPCON 2025: Fireside Chat: Aave V4 - Marcelo, Marcus, Stani

Transcript

Who wants to start? Yeah. So, well, thank you for for coming to to our talk. We're going to discuss uh the integrations of of balancer and the new architecture of of a before which this uh innovative way to to separate assets uh from risk. So I think the the first uh questions I would I would like to to to ask you Stani is um what is the the new role of of go into this uh new architecture of liquidity habs and and spokes

yeah it's kind of um really fundamental um I think um the a protocol and and the the infrastructure is quite agnostic to various different stable coins and um and and and the same kind of like a principle applies into uh the AV v4 um architecture and and the configurations. But Go specifically has a really special place in the AV ecosystem. And in the first place um Go was incepted from the idea of like creating um more predictable borrow rates for for the users. um um given the offer rates are uh fluctuating between the supply and

demand demand essentially variable uh rates um and and and going forward um um just to think about like the whole architecture of of a v4 now when you look at the v3 um obviously as an architecture you can use it in different ways so you can create a market of set of uh different collaterals and and parameters um and every time you want to create this type of a new market or a new risk profile. Essentially, u one of the key problems we we see not just kind of like um um in in a but across the lending space is the um the the the the seeding of of liquidity uh issue and the

liquidity network effects. Um so every time you create a new market you have to figure out how do you get that liquidity there so you can establish interest rates and and and the borrowing activity and for the um a v4 we uh introduced a new type of a architecture where um you have a so-called uh liquidity hub um where all the liquidity is always stored uh regardless of of the asset and then you have a uh liquid liquidity uh spoke and these spokes can be there can be multiple spokes and and spoke is simply a configuration of um uh borrowing and and and the way you supply liquidity

into the protocol. Um so from a user perspective you always come into the u a protocol from a spoke uh but then the assets are stored in the liquidity hub and what we want to achieve is that um anyone from the a ecosystem or the developer community can create their own type of a configuration whether it's uh some sort of a risk profile uh rwas or other types of uh assets or parameters um and then request uh liquidity from the liquidity hub in form of a credit line. And when you would do when you do that, you fundamentally um remove the kind of like a chicken and egg problem

that you have when you create new types of uh lending uh use cases and you can access as a developer to that deep liquidity that has. And when we think about go is that um we want to use go in the same way where um accessing not just um any type of an asset in the liquidity hub but also being able to access uh go as well and I think that's going to be interesting because those rates are configured by the aid. Yeah, this it sounds very interesting.

I'm trying trying to understand how this is going to match with the with the new architecture of balancer. We know we had in 2021 we had this uh market on mainet with BPT tokens basically the tokens uh representing when somebody provided liquidity on balancer and in this moment well I remember it was very difficult to to liquidate assets. So I wonder what's the the main difference and and why it's different now to have these uh yes composable you know yield getting uh fees from the swaps and also lending fees from from a how do you match this

and yeah what's the difference in uh the previous markets well uh I would say that one big difference is we already are integrated with a so uh v3 was launched last year uh end of last year and about 75% of all the liquidity that we have on V3 is actually boosted on a so we're talking about 150 mil that is sitting on a balancer pool is already on the landing market as a um getting yield from from there uh and with the new architecture for for a v4 what it will open up more is you'll have access to a

leaning market for tokens that maybe you wouldn't have before. So you'd have uh a DAO token paired with a go token and not only go be on a of a market but also the Dow token could have a a spoke there uh and be able to be used uh or even even provide um supply or be used later as collateral. So as we we talked before uh another thing that it opens up is using again BPT as collateral because the their new architecture is much better on analyzing the risk of each type of uh token. So when we talked about we talk about like legal Lego and composibility

we can have a boost pool with both tokens getting yield from a but also the BPT that the liquidity provider uh puts on that pool can be used as collateral for funding for example. So, uh, I think it's it's going to be it's going to be great to have it. Yeah. I mean, it's um part of the reason that why we um um wanted to make a lot of these architectural changes is that we don't really know what's going to be the future of colorization of of a um so a lot of that has been across D5 over the years. Um you know, the the native assets. So, you have basically eat and and then you know from there more derivative assets like stake it and

liquid restaking it and rub bitcoin and and um other tokens. But um down the line like one of the benefits of um biggest benefits of decentralized finance is the ability to to actually um achieve really high capital efficiency. Um and that's basically where um we want to have an architecture where uh whatever that idea you might have um in in in DeFi and you need some sort of a component of access for credits uh and liquidity u v4 should be able to cater um your idea and and using the um essentially the LP tokens as a

collateral is is is a great example because you have users that want to provide liquidity for market making across MM and at the same time they want to unlock that liquidity have better strategies um leverage those positions um and you can do that easily by just configuring a a new spoke why this is really u fascinating is that we've done in the past some uh similar experimentations with LP tokens and and obviously like one of the challenge typically is that uh you will have to see the liquid liquidity. So either with some sort of a rewards or um u you have to contract LPS and you don't have

really like a native instrument uh to do that. So with this type of integration you can uh simple uh create the spoke uh request the the funding uh from the main of a liquidity hub and then that liquidity will be available in in the borrowable assets or or go for example and I think um that creates more orderflow u deeper liquidity um and and creates better execution you know one thing it's I've been always been a fan of these um LP tokens used as a collateral because from a DAO perspective as capital allocators if we

have the option to use a native token native DO token as a collateral or a BPT I would always choose BPT because usually it's tied to Ether or tied to to stable coin and that creates a buffer if the token dumps still we have a buffer and it's much less riskier. So it means it's a much better collateral. So it's yeah it's um you know as capital we want to avoid liquidation at all cost. So so yeah I'm very happy to see this this in innovation here. Um changing a little bit the topics I'm I'm wondering with the next wave of on boarding new assets. So the one we

believe, you know, the next trillion dollar of assets on chains are likely going to be real world assets. I'm I'm wondering how these spokes would play a role in in defining the risk parameters and creating credit lines for this for these new assets. I think um it definitely simplifies uh the whole um the whole actually process of of understanding kind of like a what types of collaterals you can borrow against. Um typically on a you have like a one very deep uh liquid market and and then a secondary market. Um and and and and with this new architecture um you can

create configurations of uh of spokes where you have RWAS and not all the RWAs will be in the same kind of a configuration which makes sense because in traditional finance you have uh um you have essentially T bills you have also um credit funds um you have um equities um and and other types of uh uh real estate for example. So um what it enables is that you can create a interesting uh combination of let's say real estate assets uh that you can fund against uh through the liquidity hub and and then you might have something that

is more liquid on the uh curve uh which can be let's say some sort of uh uh ETFs uh credit ETFs as well and I think the the uh reason why it's really interesting from RWA perspective is that we don't really know what RDBAs are going to take off and I haven't had a really like a single conversation where I realize that there's some some sort of a angle that it will take off. First of all, it's a really big bet that the whole kind of industry is looking at the moment. Um, but there's still chance that like none of this will actually kind of like come on chain necessarily.

So just saying that out loud. Um but we want to be in a place where we can cater all of these experiences and being able to um allocate liquidity whatever that type of uh um activity is as long as it's credible in the sense that you can collateralize there's liquidity of chain um and then um it actually fits well with the whole a v4 vision. Maybe this is a question back to you is that what is your uh thinking about RWAs and um whether they're going to grow in the next couple of years?

Well, I think it's likely going to be driven by regulation because this is a um a big thing. So far I'm very concerned with um with the liquidity of these assets and because you know using um um an asset as a collateral that is not liquid can can very risky. So I would say do you mean like nonlquid on offchain or Yes. Not liquid onchain. Yeah. and and then the thing we have as as a capital allocator is really difficult to to to see what's the um yeah what's the liquidity of chain.

So for us something that is chain offchain is basically just a black box. So anything that is on chain is much easier to to understand the risks. But then of chains we start you know uh starting have to to trust on on a lot of assumptions that are tend to be riskier. So I'm not a fan of looping on real world assets. I think that's um that's a ticking bomb. So I'm not a fan of that for sure. Um, I know I'm also not a fan of how other protocols have been combining AMMs and and money markets.

I we have some some examples out there that they they tried it and they had some catastrophic losses, right? So, but yeah, o overall to me it's not a matter of of if for real world assets, but but when eventually it's, you know, it's a lowhanging fruit to get some extra juice of these uh assets out there. So, yeah, it's it's definitely going to happen. What's your take? Uh I I I share a bit of your your vision, but since I'm not allocating capital, I think it's going to be great when when it comes uh um it's it's it

will have to have liquidity. So for for AMMs, it should be a good thing. uh it talks to our new product uh that Fabio mentioned an hour or two ago uh that we're launching a readjustable uh concentrated liquidity fungeible at balancer. So this RWA is coming on chain uh we'll need a place to be with liquidity that can be uh um easy for everyone to to have access to. Um and again talking about the integration will be uh good to to to have a place that that liquidity sits that can be uh traded and swapped but at the same time

it's not uh sit just sitting there. It can be used as capital for uh collateral and and and making it uh uh the money work let's say. Um yeah, but you mentioned and you talked about RWA, but for for KPK, uh how do you see this integration and this changes for a v4 on a capto efficiency uh target as you guys use it? So I I remember when we started having um new layer twos a few years ago, I have to say I was a little bit uh frustrated and disappointed because DeFi is all about liquidity and and you know

having a centralized global financial system which at the time was yeah mainet. So I was very frustrated when I started seeing all of these replications of DeFi. Now I think with AB before it feels like we're starting starting to tackle this problem. You know we you should fragmentate risk not assets and that's critical. So uh hopefully in the future we're going to have yeah much much deeper uh defy uh yeah ma much deeper liquidity basically.

So uh for us it's all about capital efficiency getting better utilization ratios having better oracles you know making sure in a flash crash on a black Thursday things work. So, so yeah about uh stability, security and capital efficiency is the promise of the Legos we have how can we can stack uh yields while u at the same time not creating some systemic risk. So that that's that that's my view. Yeah, I think uh you always kind of like have to balance between like how do you segregate risk and um you know and you still kind of like you you don't create capital um

fragmentation and I think it's a um it's it's really hard problem to tackle because um you will have to always pick a direction. So you know either um you you basically don't fragment risk at all and you kind of like have one entry market um you can establish deep liquidity and and list a lot of assets there um but at the same time you can add risk and obviously risk can be capped really well and um a uh a protocol has been actually uh being quite a great example of being able to do that consistently over the years and being able to have uh 40 um 5 billion

worth of net deposits there and and seeing kind of like a trusted end that that scale. Um but then when it comes to these uh different asset classes, uh it is it is clear that um you know these cryptonative assets they're they're going to exist and that category will grow. But there's just like a really significant category of assets in the traditional finance. And if we think about the whole evolution of um you know where we the whole like financial system or if you can call it a system in the early days was like written in like clayboards u as a ledger moved into paper and then later to digital and and

now slowly it's moving to onchain. every single kind of like um uh new medium creates new types of properties and with with onchain that that that provides the the security mitigation of uh single point of failures, resiliency uh transparency and those things what basically makes DeFi uh quite amazing and uh that said like we will see a lot of these assets come come in. So um when you start seeing different asset classes and that diversification will just uh multiply uh over time. Um the kind of like a balance that we are trying to pick is that um how do you ensure that

you create um deep and efficient markets but still you can create this different types of uh uh risk profiles and and um enable innovation uh at the same time. So that's a kind of like a vision how uh we came up with the the architecture. Um and then also the pricing of of of risk is is is really important. So we always think about in terms of interest rates and how much is the cost of capital but rarely we think about um how accurately we we we price the risk. And in V4 we're introducing a a feature called uh risk premiums. And with the risk premiums, we're able to allocate a base rate for

um the the most basic asset uh collateral uh u uh composition essentially. So if you're supplying it, that's the less like a least riskiest asset, you're paying a base rate. But if you are starting to um expand your collateral composition and adding more tail risk um you basically are paying a premium on that. So what it means is that if you are creating the least riskiest position on a um you are going to pay less uh in the future than for example now. And then if you're bringing tailend risk um you're going to pay the the the more accurate price on that. So that's going to be interesting because

then when you're borrowing let's say stable coins uh from the a protocol um you have more kind of like a personalized uh risk uh profile and this prices the the risk and reward more accurately within all the uh users of of a and I think that's going to be interesting for also RWAS and and and future assets because you have to start measuring like the the offchain liquidity of these uh assets until Everything is in balancer at some point. Hopefully. Yeah. Yeah. Yeah. But I uh that's that's uh I think that's a big difference and and and that's that's how you tackle it. As you said, a got until uh this big being very conservative on

what it would put on the market uh as a lending product. Uh and we are proud to say that a lot of times we'd knock at a and say, "Guys, can you raise the cap? We have to put more money there. And we need to like especially in B and Nosis. Um and and now you open that for uh everyone to be able to use it but you're saying okay you can use it but there is a higher risk. So you you have to pay for that higher risk but at the same time you are giving access to everyone.

Yeah. And that's that's a really good feature. Yeah. And I think access to liquidity is really undervalued. Um, and I I do think that there doesn't need to be just like a one, you know, one main liquidity hub. There could be uh multiple and they have different purpose. Um, and all of these primitives that we're uh launching uh very soon um are also kind of like composable. So obviously you can have one liquidity hub um that is launched and that can seed funding to another liquidity hub that has spoke. So it's a really kind of a uh module um liquidity management architecture as well. So um liquidity

hubs are for liquidity management, spokes are for collateral management and these two together and combining them is how do you create a product in in V4? Yeah, I'm very excited for this credit lines. I I wonder um who's going to define the the accessibility to these credit lines? Yeah, it's a it's an interesting question. So, I would say that in the the main liquidity hub is um and you need to have some sort of a governance uh for liquidity hubs. They can be um you know 0x 00 can be a governor uh it

can be the it can be um a third party as well uh down the line. Um but basically from the main liquidity hub um which will be the the core market um uh you would request that from the aid essentially. Why this is really uh exciting is that um first time ever you can just build something really interesting uh deploy the contracts and request for credit line um and after review and verification if that's product is is u uh uh flawless you get you get access to to to to credit so you can start

building products on a and you don't need to focus on marketing and boost bootstrapping and all that that developers need to do. You can actually just like build something really cool, make it available and get access to credit um and and and have a really good product. And I think that's that's really what's something that will power like the next generation of of DeFi builders. Yeah, we're very excited for that. Perfect. I think we run out of time. Yeah.

Just one last question. Do you have a date? Well, it's very soon. Okay. I can say we're working working every weekend uh and and weekday. So, uh we're getting very close. Perfect. Thank you. Questions. Questions. Questions. Questions. Okay. So, um thank you so much, Stani, Marcos, Marcelo. Uh let's open it up for questions. One question there on the front, please.

On the front. Thank you so much. This has been excellent. Um I kind of wanted to ask maybe a little bit more of like a regulatory question with now that the US market looks like we're moving towards a period of time where um at least the US administration sort of considers that DeFi might be a core element of the sort of like United States policy going forward. How are you thinking about maybe engaging frankly like Wall Street and the sort of like wider uh market that maybe exists outside of crypto and sort of like what ways does that reflect in the protocol itself or not?

Yeah, I I think um that definitely there is a kind of a 180 degrees uh turn on on on on the kind of regulatory side especially in the US uh compared to going back just you know one year um uh earlier and I think that's obviously there's a couple of things so so there's a lot of interest in stable coins so the Genius Act is um in in a place where it's going to pass. There's going to be a lot of stable coins uh from big tech, fintech, commerce, e-commerce, and um and other types of use cases. Uh and some of them we will find into DeFi

because DeFi is essentially an an environment for best financial opportunities. Why? Because that's where the best algorithms uh exist uh that can be matched instantly uh with liquidity. So there's those kind of like uh um perfect conditions to um to find these opportunities. Um and then I would say there's another piece besides the stable coins is is uh uh the institutional interest in tokenization. Um and one of the things that I kind of like try to um when I talk to institutions and there's a lot of actually inbound at the moment than than ever before um is that like once you tokenize you have to figure out

like what the hell you're going to do with the asset on chain. Yeah. And and there's like a different reasons why people tokenize and um kind of like convince their organization uh to do so. But I think that all the roads will lead to DeFi. uh eventually because um RWAs are a way to scale the the DeFi opportunity um stable coins are the way to access and consume this um the D5 liquidity and also extract the u yield opportunity as well globally. It's the best technology to distribute it um across the world. Um, so I I think now the climate is probably uh better than ever before. And uh last week I was um

in DC. Um I was at the White House and I was in Capitol Hill. I was at the SEC. Uh surprisingly they let me in there. Um but uh it's it's been like a really interesting um because in the past and many of many of you who's been building D5 for few years now um has a little bit like anxiety about the regulation and even if you're deep down the down the bottom you're building kind of like a freedom technology empowering technology at the same time um you have this uncertainty and the uncertainty is what basically creates the distance between institutions solutions um and and consumers and and accessing uh uh DeFi.

So I I think now uh specifically it's really good timing to build DeFi and we probably see a lot of uh institutional uh integrations. Um so it's been really good at the moment in terms of like the inbound. Do we have other questions? No. No. One last question. No. Okay.