Why ETH is Permissionless Money | Mike Neuder

Welcome to the Bankless Summit, a series of talks from speakers all around the Ethereum ecosystem, which were presented at a one-day event hosted the day after DEF CON called the Bankless Summit. Mike Neuter spoke at the summit talking about how ETH is permissionless money. Mike is a researcher at the Ethereum Foundation, and this talk is a formalization of an idea that combines the strong property rights assurances that Ethereum's stage two roll-ups offers, along with the monetary networks that they create, along with the sound money properties that ETH Economics has all wrapped together into a single talk, along with some very apt quotes from Hayek, Graber, and Mises to make this one of the stronger articulations of what ETH and Ethereum are in 2024. I hope you enjoy, because I certainly did, this talk from Mike Noider. Give a round of applause to climbing buddy Mike Noider, who's gonna talk about ETH as permissionless money.

Mike.

Cool.

Hello, hello.

How's it going? Okay, raise your hand if you were at DevCon.

Okay, pretty much everyone. So yeah. Thanks for coming. Second half of the day of the conference after the conference. So

Thanks for sticking with us. Yeah, the title of my talk is The Ticker is ETH, which is this kind of meme that Vitalik tweeted out. But really, the true title is ETH is permissionless money. So that's kind of what we'll go into. And yeah, I'll divide the talk into two halves. The first half will be ETH is permissionless. And so we'll kind of talk about property rights, about what I think is the most important feature of Ether the asset and how it spreads property rights to the L2s and kind of beyond. And the second half we'll talk about money. And that'll be kind of my own personal take on how I think about money and how Ether spreads as a monetary asset as well. So yeah, let's jump right into it. So ETH is permissionless. And I'll start with a few quotes to kind of motivate why this matters. This one's from Hayek, and he says, What our generation has forgotten is that the system of private property is the most important guarantee of freedom. And a similar sentiment from Milton Friedman I think that nothing's so important for freedom as recognizing in the law each individual's natural right to property. So yeah, both of these kind of allude to the same point and really is the core of why I'm super excited about Ethereum and crypto generally, which is it gives strong property rights to digital assets. And that's fundamentally the value proposition of Ethereum and ETH asset to me.

So, how does this work? What are property rights of ETH? I think it's useful to start with ETH on the L1 and kind of remind ourselves from first principles what purpose it serves on the L1. And it's pretty amazing. We can summarize it in, I think, one sentence pretty succinctly by saying anyone can permissionlessly store, send, and program ETH the native asset. Right? And this is kind of remarkable in its own right.

And to kind of help illustrate this, I have a diagram.

We have Ethereum Mainnet and it has a few different accounts and a contract, right? So these are kind of like the two different ways that you can interact with ETH on Ethereum Mainnet. Obviously, you can transfer ETH from one account to another. And just this in itself is like pretty amazing, right? You can send money globally in 12 seconds, and it's like super secure and extremely easy to do so. You know, sending money over banking infrastructure is like very painful, and this is a remarkably simple way to do it.

But of course, beyond that, there's this programmability aspect, right? So we have an account here, number three, and it's interacting with Uniswap to swap for some stablecoin, die, or as someone pointed out, it should be rebranded to USDS, which is kind of the maker rebrand of their stablecoin. So yeah, this is kind of table stakes. Everyone here has probably heard this story a million times, but I think it's worth reiterating and kind of grounding what Ether does on the L1 before moving on to kind of Ether in the roll up centric roadmap more broadly, which is what we'll get to now. So

What about the property rights of ETH on rollups? Right? This is kind of the next stage of the scaling roadmap of how we expect people to interact with these decentralized applications, these blockchains. And the one sentence summary here is actually maybe something different than what you've heard before, which is anyone can permissionlessly bridge ETH in and out of rollups. So this isn't to say anything about being able to kind of continuously use any application on any rollup. That's kind of not possible. That's one of the sacrifices we make by moving to the rollup-centric roadmap. But the core thing that the rollups do preserve is the property rights of the ETH. And importantly, once you bridge your ETH in, you can always get it out, no matter what the sequencer says, even if the sequencer goes offline or is like explicitly trying to censor you to stop the transfer of your value. It's not able to do so because of this forced exit mechanism. And we'll kind of go through that now.

And I'll say before jumping into this that when I say rollup, I'm talking about the kind of canonical definition of roll-up, where there's a smart contract bridge on Ethereum mainnet. It has DA that it posts the transaction data onto Ethereum blobs themselves. And that's kind of important, and we'll see why in the next slide.

Right, so diagram number two, we have Ethereum mainnet, and now we also have this Ethereum rollup on the right. And the rollup is a different state. It's separate from the Ethereum mainnet state. But you can bridge from mainnet into the rollup by sending ether to this bridge. I'm kind of representing it with this bridge in arrow.

This gets reflected in the rollup state by crediting account one with some balance. So now this ether is locked in the contract on L1, but kind of from the view of the rollup, this is now able to be used by account one in that state.

Obviously you can transfer it around, you can use it in applications on the L2, but the element of

The ETH on the Ethereum rollup that I want to focus on is the ability to exit, the right to exit. And that comes through this force withdraw mechanism that takes place for these rollups. So the force withdrawal mechanism is a pretty cool feature where you include a transaction on Ethereum mainnet to unlock it from the bridge. And the kind of key distinction about property rights of ETH on rollups specifically is that this force withdraw function only depends on Ethereum DA. So Ethereum DA is the set of blobs. This was what was the focus of the previous EIP4844.

And it's kind of remarkable that you get this property that now Ethereum roll-ups have given the kind of exact same property rights to ETH on their chains because they only have to depend on Ethereum blobs to do that force withdrawal exit. And this will allow the bridge out to take place. And you know, you now have access to your ether back on Ethereum mainnet. So this is kind of what I mean when I say that Ethereum roll ups don't only scale Ethereum execution, but they also scale the property rights of Ether asset as it kind of moves around this L2 ecosystem.

But what about property rights of ETH on non-ethereum DA L2s? So this is a very common pattern that people are saying, oh, the Ethereum blobs are going to be too expensive. There's only three per block. So we're going to use non-ethereum DA like Celestia, like Eigen DA, to post our L2 transaction data. So I think this is fine. I don't think that type of construction is kind of categorically wrong or that no one should use it. But it's important to acknowledge that this changes the property rights of ETH on those systems. And in particular, you now no longer have the exact same guarantees and kind of strong censorship resistance that you had previously. And this is why.

The reason is this kind of first italicized part of the sentence, which says if the external DA layer is live, anyone can permissionlessly bridge ETH in and out. So now you have this kind of added trust assumption, right? You have not only Ethereum mainnet, but you have to depend on some other data source in order for this forced withdrawal to take place. And I kind of

Put the card ahead of the horse there. Let's just show this on the diagram, right? So same diagram as before, except we have this new thing, which is another blockchain. It doesn't necessarily need to be a different blockchain. This could be like an Amazon S3 bucket, but the point is the DA is coming from some external source. The bridge inflow is quite similar as before. You send ETH from an account on L1 to the bridge contract. That gets reflected in the state. Oh, and I should say the right-hand side is an Ethereum optimium. I use that word to kind of succinctly say an Ethereum optimistic rollup that posts its DA to a different DA layer, not the Ethereum blobs. So that it's kind of one of those words that sounds like a sci fi term, but whatever. It's succinct for this case. So yeah, as before, you can kind of interact with ether on this L2.

But again, the thing we care most about is this process of force withdrawal. How do I get access to my ETH if the sequencer that is sequencing this optimium goes offline or is actively censoring me? And the kind of key dependency that we inject here is that this check validity function now depends on some external DA source. And now your property rights kind of fundamentally have shifted from being only centered on the Ethereum mainnet and the Ethereum validator set to depending on this other blockchain as well.

And only after that check passes will you be able to bridge the ether out of the bridge contract into your account number one.

And then one quick aside. So we just talked about like all of this was just focused on ether the assets specifically on L1, on L2s, and then on optimiums. But people also think about when they're talking about programmable money, oftentimes they kind of say, what about USDC? What about USDT? Like these things are denominated in US dollars. This is something everyone's very familiar with. And it has the same kind of programmability. It still has that ERC20 native token interface. Like you can do most of the things that you do with ETH. But the reality of the situation is that you have no property rights with either of these because you are effectively banking with the stablecoin issuer who has the real dollars that back the stable coins. So these two Dune histograms or bar charts show the increase in the number of band addresses for USDC and USDT. And these numbers are only going up. And it's not actually that different from the existing traditional financial rails, where the government can effectively decide whether or not you have access to this value. And so it's kind of important to acknowledge that sure, USDC and USDT can be programmable, but they're certainly not permissionless money. It's a very permissioned money, in fact. Right. Bad.

So, to summarize this first half of the talk, we made the argument that ETH is permissionless. First, on the L1, ETH is permissionless and programmable. Rollups scale the property rights of ETH the asset.

Property rights on non Ethereum DA L2s depend on another chain, but it still kind of expands the effect of ETH the asset and the amount of surface area you have to interact with it, which we'll talk about in the following slides.

And USDC and USDT have no property rights.

Cool. So

That was the permissionless part. Now let's talk about ETH's money. And to motivate this section, I have two more quotes.

This one from John Maynard Keynes. By a continuing process of inflation, governments can confiscate secretly and unobserved an important part of the wealth of their citizens. So this kind of motivates the initial part of the Ethereus Money discussion where I'll talk about supply, I'll talk about inflation and kind of how inflation is a process by which kind of wealth is transferred from those who aren't protected against the inflation to those who are. And the second quote is from David Graber in his book: Money has no essence, it's not really anything, therefore its nature has always been, and presumably always will be, a matter of political contention. So I put this quote up there just to kind of caveat and tor that this is my opinion of what ETH is money is. And hopefully it's just like a useful additional framework for you to think about. But clearly, like there's no definitive thing that we can say what money is or that ETH is money specifically. So yeah, that's kind of your mileage may vary.

Right, so I thought it would be useful to just go through a brief history of the ETH supply. Sassel did a great job kind of running us through the whole history of Ethereum, so some of this may look familiar and I'll go relatively quickly. But on the left side we have 2016, this is like immediately post-genesis, and here we have 2024 on the right. And you can see that there's kind of a few very important changes, like regime shifts in the history of the Ethereum supply. So in October 2017, there was an EIP that changed the block rewards from five ETH to three ETH. In 2019, that reward was changed from three to two. So these are kind of like monetary policy changes. Every block now issues a smaller amount of ETH than it did previously.

In August 2021, this is when EIP 1559 was activated. This is probably the most famous EIP, and the burn started happening. So as you can see, prior to this, the supply was kind of increasing at a relatively linear rate. But at this moment, you have the burn kick in. And now not only do you have the inflationary pressure of block rewards coming at each block, but you also have a burn that is.

Resulting from transaction fee usage on mainnet. So

that kind of leads us to the most important day in Ethereum history, September 22,

where the merge happened and we no longer have proof of work rewards, right? And you can see that at this point, like for the past two years effectively, since then, the supply of Ether has completely flattened out. And it's right about 120 million Ether tokens in existence. And

Yeah, effectively the total growth of the supply has plateaued.

And one more date to point out March 2024, this is about six months ago now, which is when 4844 went live and blobs came out. So as you can see, like this didn't really impact the supply very meaningfully, even though a lot of the L1 activity did start to migrate to L2s. And that's because if you zoom all the way out, on the total supply change, you know, view of ether the asset, the amount of inflation is like effectively a rounding error at the current moment.

Cool. So just to kind of double click on that, let's look at the numerical values for ETH inflation because it's interesting to kind of see and think about compared to other assets.

So there's 120 million ETH supply. We saw that in the previous slide, and about 34 million of it is staked. So that's about 28%. And at that stake rate, we have 3.25% yield for each of the ether staked.

So you multiply that through and you get about 1 million new ETH per year, which is less than a 1% annual inflation rate. So, you know, if you think about compared to fiat currencies that have seen like much higher inflation rates recently in the US and you know the target of 2%, we're already like way below that in terms of what Ethereum supply is growing at currently.

On the Solana side, things look slightly different. The supply is 588 million, and 400 million of it is staked. And it's staked at a 7% yield. So not only is the stake rate more than double what the Ethereum stake rate is, but it also has more than 2x of the yield for each of those staked sold. Again, you do the math and you see that the overall annual inflation rate is 4.7% or five times higher than the ether inflation. So this is kind of just another point of comparison to another asset in the ecosystem and showing that the monetary properties of ETH are actually very good compared to this.

Bitcoin also has inflation currently, which is kind of this counterintuitive thing because when people think about Bitcoin, they think about the 21 million supply cap. They think, oh, it's zero inflation. But actually, we have not reached that 21 million supply cap. There's only 19.78 million Bitcoin in supply. And with the present 3.125 Bitcoin per block reward, that results in about the same inflation rate that we have in Ethereum, which is yeah, 0.8% annual inflation. So, you know, for the next four years until the next halfing, the rate of Bitcoin inflation is going to be about the same as the rate of Ether inflation, which is kind of an interesting.

Thought experiment. Of course, the monetary policy of Bitcoin is programmed so that that inflation rate will get cut in half in 2028 and 2032 and so on. But obviously, the thing that they compromise on in that situation is that's your security budget. Like that's what you're paying people to secure the network. You keep cutting that in half. And the kind of in my mind, the the multi trillion dollar question for Bitcoin security is how does that economic security play out when the block rewards are actually zero? And there's kind of a lot of work questioning that. I don't have time to go into it too much here, but that's why Bitcoin inflation.

Kind of in the current form is still necessary to compensate miners for producing blocks and doing the proof of work. Cool. So let's talk briefly about the burn. You know, we're talking about Ether's money. We've made it, you know, 15 minutes into the talk and haven't even talked about the burn. So

currently on the L1 transaction fee level, over the past 30 days, we're kind of amortizing it out to about 470,000 ether burned per year, which is pretty remarkable given that a lot of the L1 activity has actually already migrated to L2s. Like we're still burning a lot of ETH on the L1, and there's still clearly demand for L1 block space.

And I think a very important point about the burn is that it's highly dependent on market conditions, right? So currently we're in a period of high volatility, prices rose after the election, stuff is happening. So people are very eager to consume block space and very willing to pay high transaction fees to interact with the Ethereum state. However, there's gonna be periods where there's not as much demand to use Ethereum block space, the burn will go down. And so I view the burn as kind of this natural absorbent of business cycles, right? There's kind of periods of inflation, periods of deflation.

But overall, you still have the kind of underlying block rewards that come from participating in proof of stake that serve as like the honesty budget. This is how much we're willing to pay people to run proof of stake. And this is critical to keeping the chain online and safe.

Yeah, just as kind of a reminder, we're only producing about 1 million new ETH per year. So even after 4844, after a lot of L2 activity migrates, we're still burning 50% of the total inflation per year just through L1 base fees.

Which kind of leads directly into the second question or the next slide, which is what about burning from L2 usage? And the way that L2 usage gets reflected into the burn is through blob fees. So blobs are these data types that are written to the Ethereum blockchain. L2s pay to post this data to the Ethereum network.

And it's kind of worth going through the napkin math to see how much block space can be exported to L2s and how cheap transactions can still be while continuing to burn a lot of ether on the L1. And this is kind of like alluding to Justin's point earlier, which is in his vision, there can still be very cheap transactions on L2s, but those transactions will still in aggregate end up paying a large amount for Ethereum DA and for that like very strong property right guarantee that is provided to the L2s.

The kind of full denk sharding DA layer vision of Ethereum has 128 blobs per slot, and you can kind of multiply that out to get to two gigagas per second. So that's two billion gas per second. In terms of transfers and swaps, that turns into 95,000 transfers or 6,000 swaps. So these two things have different gas costs, but this is just like two reference points to think about how much scale can be absorbed by these L2s given 128 blobs per slot. To fully offset the issuance, which is just kind of like a reference point for how much each of these transactions would have to pay to allow the burn from blob fees themselves to offset the issuance completely, is one tenth of a cent per transfer and 1.5 cents per swap. So the point here is that in aggregate, these transactions are still effectively zero for all intents and purposes for the users, but in aggregate, they can burn a lot of ETH. So the TLDR is L2 transaction fees stay low and can burn a lot of ETH. Cool. So the first part of this moneyness arc was really talking about ETH as a store of value, right? We were paying attention to the supply, we were thinking about inflation, but there are other elements of moneyness that are probably worth calling out and worth digging into when we think about ETH of the asset and how it is interacted with across the ecosystem. So this is the last quote. This one's from Adam Smith. He says that wealth consists of money or in gold and silver is a popular notion which naturally arises from the double function of money as the instrument of commerce and as the measure of value. So if you Google

Kind of like what is money, you almost always see kind of a version of this statement, which is that it's a store of value,

it's a medium of exchange, and it's a unit of account. And this quote kind of focuses on the second two, which I'll now like highlight specifically about ether the asset across the L2s.

ETH as a medium of exchange. So the one sentence summary here is that ETH is the natural instrument of commerce, using Adam Smith's words, for paying gas and for use in applications. So back to our diagrams. We have Ethereum mainnet here. And when we're talking about instrument of commerce, there's kind of like a few things that this kind of represents in my mind for how ETH is used on the L1. First, it's both used as the gas token and it's useful in DeFi on L1 itself. So you might send it from an account to Ave to land against it,

or you might use it in Maker to mint USDS.

But the point is in both of these cases, Ether is kind of both being used to facilitate this transaction. It's kind of the lubricant that allows this transaction to take place. And it's all also used in each of those applications as the medium by which the value is stored. Of course, you can also bridge, as we've been talking about repeatedly. And if you bridge into an Ethereum rollup,

you can do many more things. And ETH is still being used in the medium of exchange way throughout the rollups ecosystem too, right? So you might swap with different pools on Uniswap to buy other tokens, or you might use EAS, the Ethereum attestation service, to make attestations. But the point is in each of these cases, you're still using Ether as the medium of exchange in the L2.

Also, you can bridge to Ethereum Optimiums.

And in my mind, Optimiums serve the best role as kind of the lowest security transactions. You might bridge a smaller amount of Ether because you don't have the same property rights that you do on mainnet or on a roll-up, but you're still willing to kind of do the long tail of activity, the high volume activity that

might come up in SociFi, so you might send a cast to Farcaster, you might battle your Axis and Axie Infinity, but the point is you're still using Ether as the medium of exchange to facilitate this on-chain activity.