Ethereum's Issuance Wars | Ethereum Korea One: Genesis

Transcript

Good afternoon, everybody. Welcome to the Issuance War. It's not as spicy, but I want it to be much more informative than be as spicy as today. My name is Hyungkyu, but just call me HQ. Likewise, I also mentioned HQ. I will be the moderator of this session. I work for a company, DSRV. We.

We run Ethereum validators since the Merge. Practically, we ran validators before the Merge, and we migrated our validators to the Beacon Chain. So, we have been working with the Lido team for a long time. And so, as a co-founder of the company, I'm worried about the revenue of our company. However, as a producer, I joined the Ethereum ecosystem because I believe in the world computer, so I would like to see if Ethereum can survive forever.

Worrying about the issues of inflation. So that's why I'm here to moderate this session. And I'm in the middle between these two. So, the topic: actually, two months ago, the general proposal of EIP-8363, named, set off the most intense debates for the Ethereum community. It asks how much Ethereum should pay for its security, and also who gets to make this decision.

So the proposal went up. Unfortunately, the proposal didn't make it into the Hegotá hard fork. Unfortunately, but the question remains. So, we have two perfect bright stars here. One is Jerome, co-author of the EIP, and the other is Izzy, from the Lido team, a team that has been in the debate. So So, I will start with a simple question, because I think that there are many people here who are not aware of this topic.

So, oh sorry, I confused the question. So, let me get back to the first question. I'm a student of your teacher, so please introduce yourself and how did you end up on your side of this debate? I'm Jerome, and I'm usually Izzy's co-host on our monthly podcast. I'm joking. I'm so passionate about this topic, and we get to speak with Izzy so much about this topic.

So it's great that we are proposing a change that can have both passionate company on the opposite side of this debate, to have the opportunity to speak here together. And I'm kind of imagining that this issuance stuff is also, we are the issuance piece. What are we going to talk about? I'm from Europe, and I used to work in a leadership position at EY and so on.

And I created a company, Cometh, which is a regulated custodian and a DeFi access provider in Europe. I started my company, and after I saw it I was like, hey, what crazy thing should I be doing? Or maybe change the issuance, it's a change, this is a pretty hard thing to do. And I'm pretty happy to be here and answer all of your questions. Yeah, hey folks, so I'm Izzy, I'm Chief of Staking at Lido, so basically I run the staking division within the set of contributors to the foundations.

I've been with Lido for five years, so not since the beginning of the protocol, not since Genesis, and not as long as HQ and DSRV have actually been participating with Lido. I started on the node operator mechanism side, so we were helping the DAO and the decisions about which node operators to onboard, how to do diligence on something without legal contracts backing it. So my background comes from risk assessment and basically how to create sustainable staking systems. I'm approaching the idea of issuance and what is the right amount that you should pay for its security.

Thank you for the introduction, and let's start the first question for Jerome. Because many people in this room are not aware of this topic, let's start from the basics. Why does Ethereum pay stakers? And how does the payment work today? And what problem do you find in this design, and how do your EIP co-authors see this kind of problem? In 40 seconds or less.

So DSRV is making too much money. We stop the subsidy. I'm just kidding. Indeed, an issuance change can be scary if you look at staking as your bottom line. If you're getting 2.5% of something that's worth $2.7k, what about getting 1% of something that's $5k? That's the question behind this. Of course, you cannot come and say we should change issuance because it's going to be good for the price, because nobody is the Oracle of Omaha, the Warren Buffett, and be like,

okay, this is how the price is going to go. Nobody knows where the price is going to go. But the main thing a layman should think about is what are the fundamentals behind ETH. And if we are able to print less, having less dilution, and at the same time get as much security, or even to some extent better security, we should consider it, because it's better to be less dilutive if we can afford to. And now, in the debate that we are having, it's like, what kind of shape of our security spectrum do we want to have?

And that's really the jigsaw at the end of the day. And of course, there will be consequences, because when we do this kind of issuance change, people get affected, just like miners were affected when they switched from proof of work to proof of stake. Nevertheless, bear in mind that seven times across the history of Ethereum we have changed the issuance. So this is something that happened before, we've always changed the issuance downwards. Now, to answer your question about how does it work: who here doesn't know how Ethereum staking works?

Good, so we have a well-educated crowd. So very basically, if you want to stake, you need 32 ETH to start off. I don't consider it a proxy, it's also a form of staking. But if you have 32 ETH, you have a validator ticket, so you can become a validator for 32 ETH. The reason why we put this threshold of 32 ETH is that we didn't want to have too many validators. It's clear that more decentralization is better. But the thing is, if you have too many nodes that are validating, you have a lot of redundancy. It's good to have a lot of nodes.

But at the same time, you have a network that's so dense that when you want to propagate a block, it takes more time. So there is a sweet spot: you want many nodes, but not too many nodes. Let's make sure that we have a healthy validator set. So don't make it too easy to start. And when you are a validator, you will be doing two things. Sometimes you are winning the lottery and it's your turn to propose a block. So you are proposing blocks, that happens from time to time. And you have to attest. Every time there is a new block being posted on the network, you have to attest that this block is right from the geographical standpoint and is the right person to be in it.

And you're getting paid for that. You're getting paid some amount when you're proposing a block, and you're being paid constantly when you're attesting. And when you're not attesting and not proposing a block because you're missing the call, like, oh, it's sort of a little bit of a problem, you're not responding, or it's sort of to attesting and not attesting, then we are taking money out of you. So you're also getting penalized when you're not doing your work. So you have a duty, at least, attesting and proposing blocks, and you're getting paid for that, and when you are not in time for your duties, we are taking money out.

So what is the payout? Because at the end of the day, I am talking about duties and payment and so on and so forth, but what is the payout? Well, the payout from a validator perspective is usually the amount that you are getting every year, what is your yield, how much percentage of the stake you get, And right now the payout is 2.5%. I'm also a solo staker since the Genesis block, so I'm very much looking at this staking rate. But globally, at the stage of Ethereum globally, there is issuance that we mint, we print ETH, we dilute all the network, we dilute all the holders, you're getting a dilution of 0.85 in say 12 to

24 months, close to 1% of the emission. That's the new emission that goes directly to the stakers. That's how we pay the stakers. And then 1%, 0.85% of the supply. So this 0.85% of the supply, due to the fact that there is 75% of the ETH staked, becomes 2.5% yield. And as more ETH goes to stake, the yield goes lower and the dilution goes higher. And that's from the yellow paper, or the specs, that gives you the exact formula.

On a yearly basis it's 166 times the square root of the amount of ETH staked. And if you remember your math class, 166 times the square root of x, if x grows, the issuance grows as well. So it's unbounded; the bound is the total amount of ETH at the end of the day, and so at the maximum amount of ETH staked, 1% of the ETH staked, you will get 1.5% yield before the dilution, which will be also 1.5%.

So, you will get 0. So, what we are looking at with this EIP is doing a clean 2% clear 1 state. That's for the criticality, because there are two issues here, to separate the concerns and the problems. The thing is, if we want to address the main problem from our perspective, which is the ratio of ETH staked shouldn't cross 50%, we should have more than half the supply at stake. If we want to address this problem, probably one of the most natural tools we have at our disposal to do so is reducing issuance, making yield and staking less attractive after a certain threshold.

We should pay less when we are close to 50%. We went so deep into the less things that 50% is basically wrong. So we make sure that it never crosses. That costs probably a bit much. I tried to keep it short. Actually, you already answered my third question. I already gave you time for it. You can add anything if you want to add to the. And also please share what does Lido do, and what does Lido provide to the Ethereum ecosystem?

And how does liquid staking actually provide to the Ethereum ecosystem? And also, please share your thoughts about the current design. Sure. Okay, so Lido's role in the staking ecosystem is, we take Jerome's earlier explanation about how solo stakers, you work to take 32 ETH and you run your own node. Lido is essentially a stake delegation protocol. Capital is amassed from a variety of different users, whether they're small users who have less than 32 ETH, or even very large users like SharpLink, who recently just allocated $200 million to stake through Lido.

It pools this in, and it essentially algorithmically distributes it to a number of node operators across the world, currently numbering over 700 node operators, but the main bulk of the protocol is still run by the 34 professional node operators, who run around 90% of the stake, and it's done so in a way to have maximum or optimal, if you want to think about it this way, risk diversification in the way that the stake is run, both in terms of jurisdiction as well as geography, as well as infrastructure type, client types, etc. So, Lido has essentially been using this system to help push the decentralization of Ethereum by onboarding more node operators, by putting nodes in underserved geographies and making Ethereum more accessible in those geographies,

and by essentially spreading the risk of that much stake. At the same time, Lido also provides an option for users that want to use it, which is liquid staking, meaning that instead of your capital being locked on the Beacon Chain, which is how it is in traditional staking, you get a token that represents the staked amount plus any accrued rewards or penalties. And then this essentially can be used in DeFi, which is a way to supercharge or augment the on-chain finance system. So the same way that you can go to a bank and make an investment or buy a house or something like that, and then collateralize this position to further use it for finance, this essentially has transformed the act of staking, which is providing a service to the underlying network, into something that can

be collateralized and used in on-chain finance. And over the last four years, this has grown to be one of the largest subsectors within DeFi. stETH is used in all of the large lending protocols, both on Ethereum and outside Ethereum. On Ethereum, it has been an entry point for a lot of new users coming to Ethereum, a way for them to use their ETH and also a way for them to hedge their exposure to ETH. One of the most common strategies in lending is collateralizing your stETH and then borrowing stables against it. This allows you to hedge ETH exposure, but it also allows you to essentially pay for operating costs. This is what we do at Lido, for example, without disposing of your underlying ETH position.

So you can stay long without needing to sell your assets. And this is actually a criticism that is often floated by the proponents of issuance reduction, that stakers are usually net sellers of ETH. And I think if we look at DeFi, this isn't necessarily the case. And it has also created a feedback loop. Now we can have an argument, and I think we will, whether it's positive or negative. But I think it has generally been accretive to DeFi. It helped DeFi survive the big bear that we had, you know, in between like 2020 to 2022. And I think it has generally been useful for the new entrants that we see.

So those are institutions that are looking to capitalize on ETH, both in terms of its status as a new digital asset class, but also this, I don't want to call it intrinsic yield, but this yield that can be attached to ETH, which Bitcoin doesn't have. And I think it's a very important competitive advantage. And so when you look at what DATs are doing, like SharpLink, BitMine, etc., and what ETFs are doing, we should think about whether issuance reduction ends up being net harmful to the ETH financial economy, or net positive. And I think that it's, our main contention is not that it's necessarily one or the other, it's just that it's understudied.

But most importantly, the thing that I want to add to Jerome's kind of intro to issuance and what do validators do, is that validators secure the network. And I think both sides actually have the same concern. The more this system runs, the more it centralizes. And what happens in a system that is somewhat centralized at a high stake rate, versus a system that is highly centralized at a low stake rate? I think, and I might be making a concession here, let's say you have three node operators at 33% stake each, that's better than three node operators at 66% stake each.

But then you have to have another discussion, which is who are these node operators? How is stake run? So most stake right now is not run via solo staking, unfortunately, or like it's just the way that the world works. Most stake is run via delegated staking. So Lido is the largest staking platform on Ethereum, with approximately 22% of ETH running through it right now. But the largest benefactors of staking over the last three years have not actually been LSTs and have not been on-chain staking protocols; they have been centralized staking solutions. Now, both custodians, so Coinbase is custodying eight of the nine Ethereum ETFs that are available right now in the United States, and exchanges.

So if you look at the largest growers in terms of stake over the last two, two and a half years, they're centralized exchanges like Coinbase, Kraken, and Binance. And Lido was essentially created when the only way to stake, if you weren't a solo staker, was through exchanges. And so we're facing a very similar set of problems, I think, in the endgame of staking now, which is, how do you want the stake to be distributed, and which are the best participants, if you want to think about it this way, or most aligned participants in terms of safeguarding the security of the network. Thank you.

So, Jerome, I want to share the details, the key idea of EIP-8363. It's really short, because you already shared the details of the end. Please also share what happens to someone who is staking nowadays, and also what happens to someone who is not staking. Sure. I want to bring up something that I was listening to you say regarding the presentation. You're saying, right, that you prefer a situation where we have three nodes and 30% of the stake to a situation where we have three nodes and 70% of the stake.

So, like, this is a very legal answer, but it depends, because it depends who's running the nodes and how it's being run. So everything else being equal? I'm on the same boat. So I would largely prefer a situation where we have less than 50% of ETH staked and few nodes to a situation where we have a lot of ETH staked and also a few nodes. But it really depends. So to put a finer point on it, if we have 66% stake and one of the three is Lido, and we have 33% stake and one of the three is not Lido, I think Ethereum is way worse off in that situation.

Sometimes when we present the issuance reduction topics, we get attacked by, you just want to kill Lido. A better way to kill Lido is to create a delegated stake in. And actually we support that. Yeah, let's just go on with it. To be fair, I think if we reach 60-plus or 70-plus percent of ETH staked, the question would be definitely around here: should we do delegated staking? The dilution pressure is so high that you just want to stake. Let's say, and we've witnessed that before.

Wrappers like for Bitcoin used to be very popular, and now the biggest wrapper of Bitcoin is Coinbase. At some point, let's speak in more, Terra Luna. Terra Luna just said, hey, let's run on Lido. And Terra Luna, they were extremely popular, they pay more, they do whatever they want to grow their thing, and suddenly this becomes a de facto LST. Instead of speaking about LST, I prefer the term staking derivatives, like a new staking derivative itself, which is considered a staking derivative and so on and so forth.

LST is the purest form of a staking derivative, but if you have an ETF, you can get derivatives on your ETF, and thus you are mingling execution problems with stake problems and consensus problems. So one of the things that makes me worry about the growth of the ratio of ETH at stake beyond 50% and so on is also the fact that as we do grow the, we have more and more stake exposed in DeFi, which can be a lot of hacks, can be a lot of frauds, and so on.

We had the case with a restaking protocol recently on the lending protocol. Super big hack. What happens if this kind of thing also happens with bigger hacks? But yeah, that's part of the thing. From a risk perspective, I like the separation of concerns. having a good crowd and a very virtuous crowd, staking, validating and securing, and the rest of the ETH being unstaked and able to participate in the DeFi economy, maximum monetary premium for Ethereum, and also being able to go back and forth between those two, suggest different things and make the fees go up by circulating, rather than having the default situation and saying,

well, I'm getting, I'm getting diluted, so I'm going to go stake. We're just removing the ETH from the active money that's going to be circulated in the economy. So let's look at what happens for the users. Right now the dilution is 0.85, so you're like, hey, I'm getting diluted 0.85. There are thresholds in the human mind. Somebody's like, oh, it's more than 1%, I need to do something. Maybe the threshold is 0.9, maybe the threshold is 0.1, 1.1 1.1. The thing is, at some point, the pressure of the dilution becomes such that you just want to stake.

And the stake just goes. There's no more to it. The best thing I can do is just to stake. So what if you are not staking in the game? What we are proposing is the following. We should change the game so that the market can actually find a way to play with it. As you mentioned, there's a lot of interest for it during the last class. I wish there was more interest, the interest was super high in the price of this kind of thing. And for those owners of ETH, you risk a lot of ETH.

The best thing to do, the lowest-risk strategy for ETH, is to stake and likely to use. So why wouldn't you do so? And that's why we have an entry queue: the amount of ETH that goes to stake just, it's just been full for the past year. Since May 2025, it's full. Just ETH going to stake. But the dam... We also put a rate limiter on the entry queue of. If it wasn't there, it would have cleared. So let's be a little bit fair about that.

That's true. The dam that's preventing the flood, it's actually the fact that there's an entry queue limitation. But this entry queue limitation is also. No, no, but it used to be increasing the amount of stake, and then we capped it to a level that wasn't even before it was increasing. That's true. I don't know if we should remove this. So I think we are running out of time, so let's get to the core problem, So, do you think there is a problem, or do you agree with some part of Jerome's?

So, I mean, the word problem is very, very loaded. I think the staking ratio is at an all-time high. I don't think the current staking ratio is problematic at all. I mean, I don't know that there is a lot of people that think that the current staking ratio is problematic. I think the main concern is where it's trending towards. So if we look at where the staking ratio was before ETFs started staking, and that's right, because these are, if you look at the new entrants over the last one and a half years, it is 85% institutional ETH that was essentially sidelined over the last 10 years, and over the last five years when you could stake and couldn't stake.

So there's like a lot of pent-up demand, you know, the people who are sitting outside the bar and waiting to get in, and they go and slam like two shots to catch up with everybody else. And so I think once this initial wave of institutional ETH has kind of settled, we will not see the kind of growth that we're seeing now. We will see the kind of growth that we were seeing two years ago, when the staking ratio was almost flat. I like, and I think once we see that, we will also have much less urgency around this kind of embedded fear in this argument that the staking ratio is climbing to 70, 80, 90% or something like that.

I also don't necessarily share the concern that above 50, like 50 is a magic number, that after that there's something wrong. 50 is a magic number in Ethereum because it has to do with how much of the network do you control, and can you do a 50%-plus-one attack. But staking is well enough diversified and decentralized that I don't think that's really a risk. And if you look at other proof-of-stake systems, they tend to have equilibria that are somewhere between like 50 to 70 percent. And there's various reasons for that, and I think one of them actually has to do with market equilibrium.

So the more ETH gets staked, even if it is in a liquid form, the more the premium of having raw ETH actually rises, which means that you are more likely to see raw ETH in DeFi to counterbalance the premium that staked ETH essentially has to raw ETH in the system, so that we will find a natural equilibrium around there. Now, I don't think it's necessarily good for Ethereum to reach 50, 60, 70% stake, but I just don't think it's a catastrophic risk. It is a catastrophic risk if it's controlled by three parties and they're all in one jurisdiction, and then essentially, Ethereum gets, I don't know, regulatorily captured, if you wanna think about it this way, by a certain jurisdiction.

But I think that is more likely to happen at a low stake rate, when demand for staking cannot express all of its preferences, than it can at a high stake rate. So, for Jerome, I also want to ask a question for you. The sharpest technical objection to your proposal was that under lower issuance, MEV becomes a bigger share of the validator's revenue. And MEV favors sophisticated and well-connected operators instead of solo validators.

In that way, it means that it enforces more centralization and things, instead of decentralization. So, this proposal that claims to protect solo stakers by pushing them out first, I'm afraid of that. Yeah, one remark on this is linked to this very question. As the staking ratio grows, is it a centralizing force, or is it a decentralizing force?

It's quite a question if you assume that there will be a lot of ETH, or if you assume that you need to capture more, if you stick with the highest figure, also because you are too big to fail. Let's say if Lido is a big problem, it's likely that we will try to find a way that Lido shouldn't suffer this much from this. The too-big-to-fail approach is a problem for community, but more importantly, centralizing more on the mode of, as we grow the staking ratio, it would be centralizing for a lot of players. That means that from an MEV perspective, it's important.

You have to consider that if more ETH goes to stake, there's less ETH in the economy, so there's less activity in the. Or it's quite little. Maybe it's not that much. But at the same time, we already see actors in the staking space that are playing games to maximize MEV. So we know that the MEV sensibility is definitely there. So I'm not too afraid of the fact that once we proposal we reach a high staking ratio, the yield will be super low so that the only way to really get paid is from MEV.

Because I feel like in those regimes, there will be a need. They want to be in the MEV game at the end of the day. And also because the market will find it's a good deal before. And at the end of the day, it's not really of any concern, because at least we can see what's happening in the economy. Specifically on the solo stakers, it's a very complicated topic to address, because there are multiple dimensions to consider. One thing I want to mention is that the current regime, the current staking curve that we have today, is also not good for the solo stakers.

It's also pushing them out. So in the process of making a curve, I think the fair game would be to compare with what's going on, with the effect of the status quo. And the status quo is probably not good for the solo staker itself, either. As a solo staker, I get 2.5%, and I live in France, so I have to pay my taxes in France. In most of the countries, the G20 countries, you need to pay tax on the yield. So my yield is 2.5 times 70%, because I have to pay 70% of the yield.

Yeah, that's 20%. So I pay 70% of the tax. So at the end of the day, I'm left with 2.5 times 0.7. But I also have a dilution. I'm paying, from a dilution perspective, 0.85. So my dilution, I just said, is pretty low if you consider it this way. So there's this effect of the tax interest rate for the solo stakers that's very, very done already in terms of the effect on the solo stakers. The curve that's paying less is also allowing for having a lower effect on the taxation, to save the solo stakers.

But overall, the problem is also like, if we want a healthy validator set or something, we need to save the solo stakers. They need to have solo stakers. And the higher the ratio, the harder it is for solo stakers to actually come to terms with it. And that's also a reason that we think is justification for keeping the ratio low. Social capacity of the folks to correct the situation. situation.

Just to make sure. The lower ratio and more diversified ratio, as long as it's about 20% or so, the better it is. So time is running out. And let's get to the last question. I just want to thank you for being here. And I want to thank you for participating in KBW this year. So if you are here next year, what do you hope will change by then? Like, in general?

I feel like this is where I need to say world peace. And that would actually probably help the markets more than an issuance discussion. So I think, like, Jerome and I have discussed this, but hopefully some sort of clearly aligned path of what to do with issuance, or at least separating these problems that Jerome discussed earlier, right, the staking rate and issuance in and of itself, and also stopping to conflate issuance with inflation, because we also do that sometimes.

So we have the burn in Ethereum, and it worked pretty well for a few years. So we should try to think about how to get back to that state of the network. I hear a lot of potential rumors about good regulatory change in Korea. So I'm hoping some of that happens so that we can take more concrete, deliberate, active steps forward. I think that AI has sucked a lot of attention and capital outside of the crypto space, but I actually envision in the next three years that there's going to be a rebalancing here.

And it's very, very good timing for nations like Korea that have generally been pretty forward in terms of how they adapt to new technologies and future finance type stuff, to consider this an opportunity to put one step forward and be ready for when this new wave of capital will join. I think this year's KBW is a lot more optimistic than last year, and so I'm also hoping for that trend to continue. Thank you. Jerome? Well, I think by next year, DB Securities, which is the title sponsor of this conference, will likely launch a major ETF of ETH.

Staking-based ETF. So your take that the queue will come down, which will never happen. So I would be very cautious on that. We're going to try to propose a path where we can converge on a consensual issuance. What I'm hoping for for 2027 is that at the very least we can agree on a breakout mechanism that says, like, okay, fine, at this amount of ETH staked we need to drastically lower the yield, because that's too much, just to make sure that we avoid crazy situations, and also we make sure that just forward thinking and forward relevance. Specifically, we have the burn, that's

great to have the burn, but in order to have the burn we need to have activity, activity on the L1 and activities on the L2s as well, and we should keep in mind that all the L2s that are bringing activity on their own to make their own type of adoption, and burning ETH through blobs as well, they need to stay relevant as well. And when they are using ETH as their own currency, those are getting diluted, so it's a problem for L2 adoption as well in this case. So what I'm really hopeful is that, uh, well, Korea continues to drive the success of blockchain as it has before, and that I can come to Korea Blockchain Week next year with another hot topic to debate with Izzy.

Block pricing? But we will be just in line on this. Thank you for sharing the optimistic view, and I also hope that happens. Thank you, and thank everybody here in the session.