Staked ETH Withdrawals: Expectations & Associated Risks

hello everyone um thank you so much for joining us on this panel about staked eat withdrawals we've got a great set of panelists today who are experts on this topic let's start off with around a very quick introductions please say your name and what you do I'm Nick zorkish or nixo for short I work for East acre East acre it's just a unbiased platform that helps people figure out how to stake and I'm Pablo villalba I'm from Diva and diva is doing Liquors taking fully powered by distributed validation so I'm langis I'm from rocketpool and we are a decentralized liquid staking protocol I'm Alex Stokes I researcher at the EF and I've been doing a lot of the spec work around withdrawals so it should be a fun panel it will be a fun panel and I want to start with you Alex because we're all talking about the Shanghai upgrade and you've been so close to the ethereum like the whole process for determining what goes into Shanghai what doesn't so my question for you is why was steak to eat withdrawals prioritized for Shanghai and what does the prioritization of state to eat withdrawals over scalability over Mev censorship resistance all these other things say about whose interests developers are really prioritizing yeah that's a great question uh I think from the outside this might be like a confusing or opaque process but ultimately there's like a group of uh client developers and researchers who uh I suppose do the prioritization here and yeah I mean there's a lot of debates sort of internal to that Community around like yeah what feature goes next uh this one or that one obviously like hopefully there's a lot of feedback between the broader community and like that set of people which I think there's like a healthy back and forth right now and yeah so what that looks like is you know we had the merge which was like a big event where now you know some months after that and the question is okay what's next and uh yeah where we've landed is like shipping withdrawals there's like a number of really big upgrades like you mentioned some another big one is like 4844 and like the scalability and starting road map of ethereum so you know for example let's say why withdrawals over like 4844 um 444 is like a relatively big change you know compared to say withdrawals and the idea kind of where we landed was you know it draws this list is like more compact smaller thing we could kind of do quickly um it's something that we've been promising the community since like the launch of the beacon chain right like this has been a long-standing sort of promise to like the broader Community eat sticker and otherwise and um yeah so seems like the right time to do it and yeah then it unbox us to do a lot uh more exciting bigger things down the line I feel like there's going to be continued contention even with Cancun coming up like there's a lot of things that we're hoping to pack in um but yeah I mean validators are really one of the core focuses for Shanghai because steak eat withdrawals is such a big Focus um I wanted to ask you Nick so because I feel like you're really close with validators like what validators are thinking what validated node operators are thinking for Shanghai um tell me what are your thoughts around how validators will start to unstake after Shanghai do you think there's going to be a big unlock do you think what percentage of rewards do you think validators are going to sell or how many validators do you think are going to restag what do you think is going to be the staking kind of activity happening after Shanghai I think it's going to be impossible to predict but it's going to be really interesting to watch and I think that a lot of the solo validators that we're watching don't seem super interested in withdrawing their whole stake a lot of them are very committed to the chain committed to staking it's always been a long-term investment to stake it's going to be really interesting to see what the bigger organizations like centralized exchanges do because that's going to be harder to predict because those people weren't necessarily banking on a long-term investment when they staked any predictions on the staking rate for what ethereum what's the staking rate is going to be for ethereum at the end of the year I guess that depends on how many people withdrive a lot of people withdraw it might go up but I if I had to guess I wouldn't say I wouldn't guess that it's going to change significantly for very any long period of time interesting interesting um yeah I guess the other big picture that you were talking about is you know what our centralized exchange is going to do let's also talk a little bit about decentralized Protocols are going to do what rocketpool is going to do how much of a technical lift has it been trying to prepare for Shanghai prepare withdrawal prepare rocket pool for withdrawals do you foresee any technical risks potentially being seen on the upcoming quarterly test net tell me a little bit about how hard it is for so so drawers are actually relatively simple for Rocket pool in fact our contracts kind of already handled withdrawals and we just need to tweak it slightly to handle what's known as partial withdrawals which are essentially like the skimming of rewards above 32 eth so we just need to kind of add that in so we've worked quite hot you know closely with the EF to make sure that you know it's compatible with the way that we work and so well they reached out to everybody it was just us um so uh yeah so it's been relatively painless and in fact we've been participating in the Xiang test net and so that's that's actually been great fun actually kind of exiting a validator watching it come back you're watching skimmed rewards come back and everything it's uh it feels complete it feels feels very very good uh no so there's no real issue in terms of that ritual stuff we do have to do an upgrade which is always you know a slight risk but um yeah that's we don't have any concerns around how the withdrawal mechanism Works no concerns around like withdrawal credential changes as a process additional process on top no so actually with rocketpool there are no withdrawal credential changes because all of our withdrawal credentials are at zero zero x one which is the the standard ethereum ritual credential nice yeah so so basically from as at the moment Shanghai releases uh our Awards will start or our node operators rewards will start um skip being skimmed uh they'll be able to exit um and and kind of do everything they they can do one of the things that I'm really excited for is the fact that like you said Alex like the validator life cycle is going to be finished is going to be a big part of um one final piece of the merge finally like getting complete and I think it marks perhaps a new era for validating and um Pablo I'd love to talk get your thoughts on how you think distributed validator technology is going to change the validator landscape moving forward after sharing High well actually first like where is DVT right now in terms of its development and how do you see like the competitive landscape between the different players in DVT panning out after Shanghai yeah great question so in terms of um in terms of distributed validation I like to split the problem into three different domain areas there's really liquids taking like rocket pool is doing and does economically Superior for many different reasons and actually there's a big advantage of withdrawals which I can touch on a bit later and on the other side there's DVT but DVT at this core just means I'm Distributing a private key so we're already non-custodial with funds like rocketpool for example but we're still custodial with validation do this so what DVT allows you to do is split that between different people but it does not provide the trust assumptions behind that model so you still need to trust those people or find a way to reputation with them collateral with them or create some create the right incentives right to reward good actors and punish bad actors and eject them so I feel like DVT at the moment is still at that first stage of the technology is interesting but it still hasn't reached the level of trust that we need so I think really there's three levels there's a level where you need to trust them by you Distributing there's a level where you can socialize rewards amongst the network and there's maybe even a third layer where you can attribute in a granular performance each actor of the network whether it should be rewarded or punishing we are working towards the third step of that and I think DVT and liquid staking is really a great combination peanut butter and chocolate because it allows you to have better risk management better up time for example Remodeling and if you look at a set of 16 different nodes and each one of them is down five percent of the time so they miss one out of 20 other stations the combined result of a two out of three threshold will have 99.99 uptime so there's really a lot of promise I think and there's already a DVT validator live on mainnet I think is something that opal announced recently um are you in direct competition with opal um I like to think that we're solving a different problem so oval is creating a technology like distributed validation technology that allows different players to adopt it and upgrade their infrastructure while what we're doing is a fully vertical integrated liquids taken solution that relies on DVT to to have better risk management and better uptime very interesting so DVT is like one thing that I think will be important to watch po Shanghai that's really going to take off and another thing that I am looking quite heavily into lately is continued changes to kind of validate our economics and Dynamics moving forward one of the interesting things that vitalik had written about in the past few months was about how developers might have to cap the validator size in order to in order to enable something called single slot finality these are all pretty technical terms but Alex can you talk about some of the upcoming protocol changes that will impact validator economics and what are the benefits of doing that what are some of the trade-offs you guys are thinking about as ethereum core Developers sure that's a bit of a big question but yeah so in case uh someone's not aware this this concept of single slot finality uh refers to this consensus feature that we have today but doing it in a much shorter time frame so what happens today is basically there's uh something we'll call the available Ledger this like sort of suffix of the chain that people can build and there's this trailing process on the same chain but it gives you this economic concept of finality and this is like where we get proof of stake security um yeah so with single slot finality what we want to do is say okay the process today says you know maybe it takes like some number of epochs some like number of you know this this distance behind the head of the chain that you're trailing to like finalize we want to say rather than take like you know some number of slots let's do it in one so like one sort of round of the protocol in order to do this um yeah there might be some changes it's looking like just from some like early benchmarks that uh to have to do this in a way that you know is you know supported by the technology G you can only have so many sort of validators at one time and then from there you have to imagine like uh yeah what are the trade-offs there right now there's sort of like uh anyone can join set there's no sort of like back pressure on that other than like the staking yield um and so sure what other trade-offs there what does that look like um there there could be some the thing is there's like different designs like Christina we were talking a bit before and like your your concern was like oh hey like what if uh if there's like a fixed cap like let's say there's only a million validators like that seems a bit weird because it's like why would you want to cap the side at all I was shocked and yeah I was like kind of surprised to even think about it that way because I think like a different framing is just more that it changes like the implementation of the incentives um so you end up with like the same outcome just like in a different perhaps way that also gives us additional thing of the single soft finality um there was something else to say here you could potentially have to pay to be a validator and that might be good for Mev well sure yeah so like a way to think about it is that if we move to like a fixed cap oh okay this is the point I was gonna make I'll get to that in a second if there's a fixed cap then yeah you would imagine that basically the staking yield is adjusted such that like you know as much stake as the community decides is like a right amount to pay for security that's all staked and then from there it could turn out that like there's demand beyond that for example to access like priority fees within the execution layer and maybe things like this and the way it would work in that setting is you then basically need to like pay to be a validator but it would be economically rational given like the fee revenue and all this um so that could be like an interesting change again it's like it's just a different way to like think about the economics the other point I was going to make is that there's different implementations of the like validator sort of Set uh handling so rather than just have a fixed cap you can imagine you still let the set be as big as you want but then you like randomly cycle people in and out every so often um so that's why to me it's like it's not that we're like limiting the side like you know limiting decentralization that's in fact the opposite of what we're trying to do the classic trepidation of we haven't committed to this not not yet at least but no so really interesting things to watch out for protocol wise dvt-wise um let's talk a little bit about how the liquid staking derivative landscape is going to change um you know post Shanghai long term liquid stick liquid staking token right the liquid staking token now you gotta watch my words now the SEC is watching um but yeah I mean so Lido has an incredible share of the total staking market and it seems to be growing they've got their biggest inflows um daily inflows I think a couple days or weeks ago um where do you see rocket pool fitting into that competitive landscape and how do you see the market share of these liquid sticking derivative protocols growing um post Shanghai okay so uh to compare a decentralized staking protocol with a centralized taking protocol is kind of not comparing apples and and apples you guys aren't in direct competition um no no we're in obviously there's definitely a competitive um kind of uh nature there um but the only real difference is that so A centralized mistaking protocol can um has in infinite resources no matter how much East you throw at them they can actually scale to take that on with the decentralized staking protocol we're relying on individual node operators signing up with us and spinning up um kind of mini pools of what we call Mini pools they're called validators and so they're they're very very different in the terms of how they can scale now um so the difference there though is that so we're making a few changes to our protocol to so that so currently our node operators need 16 East um to be a node operator uh we're actually changing it to actually eight eighth so we've done a lot of risk modeling and we've figured out that we can actually um still provide you know the greatest degree of safety for our Earth holders and with a teeth um but well we can we can kind of lower that barrier of entry for node operators from our perspective that means that we'll be out of 3x on day one our capacity but we'll actually more likely 5x over time so that puts us in a much more competitive position against elido but at the same time we have also kind of signaled that we all self-limit at some point because ultimately if we get too large which uh arguably uh Lido is uh we will put um ethereum at risk and that we don't want to do that even a decentralized staking protocol um we would put them at risk so we have decided that we all self-limited at some point and so that our kind of like 5x um Improvement in in capacity enables us to get to almost that point um and but still be able to kind of you know take on as much to confirm you said you've committed to self-limiting but haven't committed to what the self-limit is uh I think there's a there's a number there's some numbers that are kind of floating around um and so we you know it's basically that we don't want to be we've committed to the the prospect that we will self-limit rather than put ethereum at risk um now I don't think we put a hard limit on that yet but we we're to be fair we're at 2.4 so uh it's not a big issue for us at the moment um but at some point we've made it clear that as a core team and as a community that we will self-limit gotcha at some point that is going to be yeah but in terms of in terms of also the the competitive landscape with Lido so a lot of the people that we've spoken to because we are now you know a kind of a credible alternative to a Lido um the idea that they put all of their stake in one provider is not great for for many many people so they they want to diversify and they want to diversify across multiple providers and we we provide a you know a very solid alternative to that right yeah yeah I think with recent regulatory actions by the SEC against Kragen decentralized taking protocol Solutions are becoming more important than ever um and also it sounds to me like there will also be a greater push for say institutional activity in staking as opposed to retail staking activity as the validator Whisperer next though I'd love to get your thoughts on do you see a really big appetite from institutions to stake on ethereum once staked withdrawals are enabled and how do you think about institutional staking services like alluvial that coinbase says is helping to support fit into the to the picture of Major players in staking so I definitely think that East acre focuses on the solo Staker the home Staker the person running it in their home because that leads to the greatest amount of decentralization and so that is our primary target that's our audience that we that we talk to we don't have a lot of communication with the big institutional stickers but I do see a bigger appetite already coming for staking from institutions universities just bigger organizations that want to be part of this because it is um I'm calling it the estate wealth of the future uh so I want eatstaker to East acre in general wants to to focus on onboarding as many home stakers and keeping as many homesteakers as possible are the technical requirements to be a homesteaker progressively getting a hardware and however over time I think when the beacon chain launch it was like you only have to run a consensus layer node and then merge it's like execution layer and consensus layer and now with steak dates withdrawals like you've also I mean every Mev boost also adds a layer of complexity and that's been around since the merge but this is also something that they're gonna have to prepare for I mean do you see it as sustainable that like homesteakers will still be around due to the complexity of staking absolutely and I think actually Alex Stokes can probably answer a little bit more of the technicalities of this but the um the requirements have gone up especially post merge when the execution layer became necessary but there are a lot of things on the roadmap that will also bring that down um and at the moment right now it's a you can buy an Intel Nook for five 500 five six hundred dollars and it is four inches by four inches by two inches it's super quiet runs on two dollars a month so it's not actually that high right now Paolo would you agree that do you have anything to add about the the Dynamics of like homesteakers versus people who are staking through Services um decentralized or centralized anything to add there yeah so the way the way I think about the second ecosystems there's a lot of value and diversity so definitely sort of stickers have been the pillar but uh unfortunately they're also getting a deal which um has some disadvantages in some areas and I think that's where citizens like rocket pool and Diva were trying to fill that Gap and allow you to have the decentralization impact that you could have as a solo sticker without missing out on the economic advantages of of being coordinated and some of those are well if you are running a solo sticker you need 32 eth right and today that is a high amount but unfortunately that's only going to get worse and what we observed is there's a high correlation of GDP versus staking activity the us having 40 followed by Germany and other high GDP countries um so things that will help in that are lower bonds like rocketpool is doing now with eight eth what we're doing is thanks to DVT we're managing to lower that to one eth without actually compromising on the on the bond but we're managing to distribute them more evenly and ultimately their smoothing effects of joining a large team that means that you can transform infrequent rewards into something much smoother and much much more predictable so I think there's definitely value in joining a network and being coordinated and we don't necessarily need to miss out on the benefits of running notes from home which I think you can have your cake and eat it too in that sense to be clear yeah I don't see solos taking at odds with all these protocols that are building on top of the validator set yeah that so rocket pool has got over 2 000 note operators and the majority of those people are home stakers the reason why we know that is because most of them are in our Discord talking about how they're homestaking uh in our support channels and all this sort of thing so it's really exciting to kind of have you know homesteakers be able to do this and be able to participate so I think we're getting close to time although I don't see the exact countdown oh there it is aha we have five minutes excellent okay good um and I do want to leave some time for just like practical people in the room who are either staking through rocket pool or running their own at-home validators how they can prepare and how they can make sure that they're ready for this upgrade I know that the Gorly testnet is launching on March 14th you've got a potential launch of this for for April um any advice on what at home stakers should be doing now um maybe let's yeah yeah sorry yeah so um so actually there's a another test net called Xiang which is which is uh post Shanghai now so um you can actually spin up a validator on on XI Jang and in actual fact it's super simple because xujang is a very small chain so you can sync it in 30 minutes um so you can sync in 30 minutes you can practice depositing you can practice withdrawing you can see your rewards coming back from rockable's perspective we've released our Atlas release which is the East mini pool stuff and everything on to Xiang so you can play around with that as well so yeah that's probably the best way is is going to be released very soon I think it was the 14th of 14th yeah yeah so uh 14th of March and so that that is a big one because there are a lot of people already running validators on growly so it's for them to be able to practice that maneuver I guess is is the best way to do it yeah and Gorly eat I think is trading at a premium there was a panel about this but yeah squirrely it's having some test net faucet issues but anyways um Alex how risky and from your perspective as a core Dev is the Shanghai upgrade I mean the merge obviously there was a lot of of developers that were like on their edge of their seats like hoping that the upgrade to as well are you guys very nervous about Shanghai like what are the risks associated with the upgrade yeah uh I don't think anyone is really that nervous at this point I think it's been pretty de-risked um the merge was like a huge thing and like it's uh almost a miracle that it went as well as it did and this is actually something I was going to chime in with earlier it's like there's so much work that went into like essentially de-risking the merge that this is like testing infrastructure that we can now leverage moving forward for like the rest of ethereum's future and so we've seen like for example like we have these like very rich and like uh test net settings where we can like run through different hard Forks we have this concept of like mainnet Shadow forking where basically we almost like dry run the fork at a time on mainnet so to speak um yeah so there's techniques like this there's like very sophisticated tooling just like absolute uh amazing people doing this stuff uh yeah to make sure that it goes well and you know all in I think this is like one of the less technically like complex works so I'm not worried um sepolia happened it went well gorley is coming up hopefully it goes well and then I mean that's on the horizon knock everyone yeah for sure and it does really seem like developers are trying to hammer down the process of doing these protocol upgrades and like yeah just really get up to speed about doing them quicker and and more often so really interesting to see um I guess one last kind of big final topic for this panel there has been quite a lot of concern around the future decentralization of ethereum's validator set and how to prevent it from being so vulnerable to regulatory capture and I think it's really relevant to some of the things especially on this panel DVT at home stakers rocket pool ethereum Corp protocol developers so what do you guys think is one of the most important uh things to like in your mind like what is the the most important solution to this issue like how to how do we keep ethereum incredibly neutral um censorship resistant are there any blind spots that you think currently are not being addressed on ethereum um that you want to highlight um for the audience today let's start with you nexo yes we need more user design people we need more user experience people who don't cater to developers because a lot of what staking is right now is command line interface and that touches a like the intersection of people who have 32 eth and who are comfortable on command line interface is such a tiny subsection of a very homogeneous uh group of people and so we need user design people who are looking at that and saying if I was from this country grew up this way had this much money I wouldn't be able to stake and so I need that to be easier things like the launch pad like the validator launch the launch pad is great the Launchpad does a really good job of being more easily accessible but you still need command line interface to stake I totally agree I think the main aspect is we talk about decentralization where we should be talking about inclusivity inclusivity for me comes in economic flavor like how can we lower the cost of Entry as much as possible that's why we're all like about Darren and I are working on lowering the the minimum eth requirement a second one is technical how can we fit it in a box that works as much as possible like an Apple TV it should be something that you just connected to your Wi-Fi you sign in with your wallet you sign a transaction and you're validating in my in my future world and ethereum validator should work just like veteran instead of streaming files and movies you should be streaming validation duties and until we get to that point we're going to be missing out on people and we're not going to be as implicit as we could be great point uh yeah so from our part we've been trying to do our part in in that um basically trying to make it easier for stakers so we have a whole kind of like smart note stack which makes it super easy but it's still command line well it's kind of semi-comm online it has like a terminal interface which is quite cool but anyway so that um so um we certainly trying to make it easier we also have like a massive Community who have onboarded hundreds and hundreds of homesteakers and they it's like a community are fantastic so you know that onboarding of new node operators is really important but the yeah the ux is very important we're trying to lower the collateral limit and required to be a node operator to do it in a responsible kind of way we have to make sure that you know overall that we're we're doing right by our protocol and we're doing right for ethereum um but yeah but by lowering the barrier of entries that's how we kind of get more and more mistakers and and more diversity of sakers yeah I agree with all that uh I think I'll zoom out a bit for my answer to this question and just say like complexity broadly of like the protocol and all these pieces we touched on this some earlier with like yes now you have to like run all these different pieces of software and like it seems like the number keeps growing over time like even you know within the core Dev Community like that pain is Fault right and so like uh the people building this stuff aren't very aware of that and like if anything I think there's a huge like very intense debate internally around like how do we you know navigate this sort of trade-off between adding all these cool features to the protocol versus keeping it simple and streamlined over time um and yeah that's something that like will always be this like delicate dance as as the future unfolds um there's like this context of the broader world around ethereum the protocol but also the community and things are moving so fast and constantly changing it's uh yeah it's it's a it's an exciting time great well thank you so much to our amazing panelists for this and thank you so much to our audience for listening thank you