Coming off the back of our recent partnership announcement with Ion Protocol, we were joined by Chunda McCain to discuss lending on Ion and liquid restaking.
In case you missed it, here are the top questions from the event.
What are Swell’s liquid restaking plans and how do they fit in with Ion Protocol?
"Swell is looking to build out the best possible liquid staking and liquid restaking experience in DeFi. We currently have our LST live, and in the past 6 months we have become one of the fastest growing protocols in the space having just surpassed 100M+ in TVL. Swell is looking to to extend our products into a liquid restaking token (rswETH) and were very excited to launch that soon. LRTs are in many ways an extension of the LST and their use in DeFi is enhanced through the LRT. That naturally comes with a bit more complexity, risk and more rewards but it nonetheless is an exciting opportunity for us to be able to deliver to the community." — Daniel, Swell
Generally speaking, we saw a similar problem emerging in the current restaking environment where in effect what happens is those restaked assets on EigenLayer are non-fungible so you lose that DeFi functionality. Swell is looking to liquify some of those positions and give people the opportunity to go ahead and use that in DeFi. As part of our plans, we wanted to approach the development of the liquid restaking primitive together with people in and across the ecosystem and found Ion protocol to be leading the charge in the area of lending, borrowing and underwriting. These are all incredibly important concepts to get right in the development of an LRT, so we partnered with them in addition to other AVS and operators.
One of the key outcomes we are looking for is how best to collaborate with people that are thinking about collateral underwriting frameworks. Those very frameworks are going to be the ones that will eventually lead to some maturation pathway for the LRT to harden itself as the collateral, and be the pristine collateral that we are starting to see. ETH has been the most pristine, liquid and accepted pledged collateral in the space, and it is quickly being taken up by LSTs as they are yield bearing instruments. This line of thinking can be extended to LRTs given the fact that they are a double yield bearing instrument, or in some cases a double + yield bearing instrument. Just getting that flow right and having the right parameters and frameworks is top of mind for Swell and I'm sure top of mind for the team at Ion as well." — Daniel, Swell
"Ion is a lending platform specifically for staked and restaked assets — from LSTs to LRTs and everything in between. Our goal is to create the most capital efficient market for you to borrow and lend with your staked and restaked assets.
The LSTFi ecosystem is becoming more mature with a wide variety of yield instruments that can provide users access to ETH denominated yield with very minimal risk. We are seeing products being potentially double digit nominal or real ETH returns. The problem with that is when you’re looking to build LRTs and bootstrap capital on EigenLayer, you're using the same type of super liquid, universal accepted asset (LSTs) as the lifeblood of EigenLayer liquidity. The limitation with this is you effectively have to fight against the opportunity cost of participating in the greater LST and LSTFi landscape.
Our goal in providing a performance lending platform, specifically for the sub class of staked and restaked assets, was to fulfill the goal of being able to scale capital allocation in EigenLayer by switching the paradigm of users having to choose to deposit their LSTs in EigenLayer or in LSTfi/DeFi. Instead we can do the risk underwriting to turn that into an opportunity where users can deposit in EigenLayer and still participate in DeFi with a risk adjusted representation of that LSTs value.
The core ethos for us at Ion is answering the question: 'What does it truly mean for staked or restaked assets to be solvent?'. In traditional lending architecture, solvency is a matter of whether or not you can liquidate a position fast enough before a bad debt enters into a system, whether that be a CDP, base system or lending market. In reality when you think about LSTs and LRTs, the risk that you are incurring core to the protocol is whether or not there is going to be the same quantity of assets and the same amount of ETH in your underlying position after you take into account the risks of potential slashing or penalties that you can incur. We focus on using trustless systems to view and take a look at these underlying balances and predict the propensity of these systems to be slashed or to have vulnerabilities or penalties that could lead to a lack of solvency. Realistically, stepping away from this liquidity oriented mindset of how we underwrite assets and more towards this infrastructural mind, our methodology has to be the way that we can enable capital efficiency at scale otherwise you're going to get a lot of fragmentation of liquidity across the spectrum. This depends on liquidity incentives to provide a robust secondary market liquidity, and liquidity for all these assets is going to be very difficult." — Chunda, Ion
Is Ion’s approach to restaking new or is it considered in relation to other existing approaches in DeFi?
"What we are trying to ensure when we set LTVs, mint caps, borrow caps, and supply caps, is the solvency of anyone in these positions if the price of the underlying asset falls. The inherent constraints with this system is the dependency on liquidity. The more liquidity you have, the more prices are stable, and the less you have to worry about the situation causing bad debt to enter into one of these markets. At Ion the question we ask is “What does it mean for a LST to be solvent?”. We realized that if you are taking on risks that are infrastructure derived, we can't depend on liquidity because there won't be a lot of it in every single one of these markets. If we want to allow users to select any AVS operater they want and borrow against any one of these assets, then we need to think about underwriting this infrastructure risk and that's what we focus on doing.
The concept of underwriting or pricing out risk return for a liquid restaking token is definitely a major area of focus for Swell. Chunda alluded to the difference of combinations users are going to have across the different AVS’ and operators of those restaked assets which becomes more complex when you have delegations mechanisms as well. Being able to understand how those positions change not only at one point in time, but over a continuous period is an interesting area for us to understand how best do we use, manage and set up these liquidity restaking pools, and how best do we express what the receipt token representation of that pool is going to look like.
We expect this space to be continuously evolving and over time will become more complex but also potentially more rewarding from a capital perspective for restakers. Over time, I would expect that as more information comes to light and there is better understanding of slashing, correlated slashing and what those frameworks or standards will be, that you will see this stratification across the various restaked positions which are then liquified. The importance of underwriting is just appreciating that risk, lending appropriately and then being able to liquidate those positions in an orderly manner without having those positions flow on systemically." — Chunda, Ion
"For Swell, whilst we are in the early stages of the overall liquid restaking ecosystem, these are some of the things we are thinking about in the medium to longer term. It will definitely be a step through risk management approach as we look to set up the pools, but over time we are expecting the situation to play out such that you'll see low risk, low return and high risk, high return categories start to come out. A lot of that will also be composable with these entire borrow lend markets." — Daniel, Swell
"One thing I really want to highlight is this idea of DeFi being visible and transparent and being able to understand risk. I think we take for granted how transparent risk is in DeFi especially when we talk about liquidity dependent underend models. When we think about risks entrenched in hardware dependencies in concepts such as slashing risk, there are a lot of things that aren't necessarily transparent and that is where topics such as — how well our node operators are securing their systems, how reliably are node operator running their hardware setups and are there any internal dependencies on node operator hardware setups between AVS’ — come up. One of the biggest focuses for those who are taking on the role of financializing these positions is there is no impetus for all of these stakeholders to make this data visible or easy to visualize. A large responsibility we have whilst building out these primitives is to ensure that users understand the risks that we and the users are taking on by entering into any one of these positions." — Chunda, Ion
What is Ion’s approach to the risks highlighted above and what is your framework for zero knowledge machine learning?
"At Ion, we look at quantifying and underwriting risk as a two step process. The first step is understanding and visualizing risks itself. A lot of people don't know about the pre-existing indicators that lead towards certain validators being slashed. It can be as simple as how long the validator has spent in the market, the historical income data and attestation accuracy or which client you are running. It takes a lot of effort and infrastructure to be able to visualize all of this data and understand where all the weaknesses are for any one infrastructure provider.
We started by building a proof of concept ML model that takes in all of the infrastructure data about Ethereum validators and hardware operator systems. Based on all of these different points of data, we then spotted the trends and outliers that drove insights into what validators might be more at risk than others. It took months to do this level of due diligence because a lot of this research is something that researchers haven't spent a lot of time working on due to there being no application for it yet. We then took that data and plugged it into our ML model to get an idea of extrapolating the actual slashing risk that any subgroup of validators would experience over a given time period. So far our air margins have been sub 1% of being able to predict the propensity of validators to get slashed on the beacon chain." — Chunda, Ion
Ion recently announced their upcoming integration with Swell. When will we be likely to see swETH on Ion?
There are 3 main things that users will be able to do with their swETH on Ion:
- Leverage staking looping
- Borrowing against swETH
- Hedging against swETH exposure
We are currently going through audits and security reviews for our mainnet deployment, which will be completed in mid January 2024, and you can expect to see Ion on mainnet with Swell very soon after." — Chunda, Ion
Thank you to everyone who attended the event!
To stay up-to-date with Swell, join our thriving community on Discord, DeBank, and X.
