125 - Ethereum's Hidden Power Structures | Matt Cutler

Welcome to Bankless, where we explore the frontier of internet money and internet finance. This is how to get started, how to get better, how to front run the opportunity. This is Ryan Sean Adams. I'm here with David Hoffman, and we're here to help you become more bankless. Excellent episode for you today. We're talking about Ethereum's hidden power structures. Some of the power structures you didn't know existed but that do, and how those power structures are going to change post-merge. This is coming down the pipe really soon, and you need to know about it. Some things to look out for in this episode with Matt Cutler, who guides us through and educates us through these new power structures. Number one, what the power structures in Ethereum are today. Number two, how they're going to change post-merge. Number three, what this future world will look like. Is there a future world, as Matt says, where users might actually be paid to use Ethereum? No gas fees. Instead, Ethereum pays you. And number four, why the value of this new power structure, this new economy, flows back to ETH the asset. Yep, that's right. Back to ETH the asset. We talk about all of those things with one of the people who is most educated and most articulate in expressing it. David, what were your thoughts on this episode going into it?

In post-merge Ethereum, ETH pays you for gas. I thought that was like one of my favorite parts in the show. And we really go through how the supply chain of block creation is what that looks like in post-merge Ethereum. You might think it's simple. People add blocks to the blockchain. It's not that simple. It's not. It's not. There's way more moving parts. And as we go further into Ethereum's roadmap, there become even more moving parts. And Matt does a really good job illustrating the supply chain, illustrating this how this works, but also just talking about this at first principles perspective, where modularity creates flexibility. And as soon as we have flexibility in how Ethereum is designed, we can start to tinker with how we think that it should be designed. And ultimately it's in the Ethereum ethos to make sure that the user has the most amount of power and sovereignty and optionality. And Matt walks us through how this concept of proposer-builder separation and MEV boost and MEV searchers and block proposers and validators, how all of this modularity ultimately flows back into enabling the individual ETH staker and ETH the asset. And so Matt is just a gigabrain of what we call the pre-chain layer, aka the mempool. And the mempool is this very dark and disorderly and chaotic place where it's not very friendly to individuals, but it's really a scary place. This is where the monsters lie, this is where the MEV lies. But what Matt is building at Block Native is helping illuminate the mempool and helping us reason about it and turning the mempool in a place of chaos into a place of order that we can reason about and have a dependable ecosystem. And so he just walks us through all of these details. And I think it's going to be a really fascinating episode for people that care deeply about the protocol layer or just want to understand how everything flows back to ETH at the end of the day. Either one.

Look, these blockchains are economic computers, and Ethereum is an economic computer. And the best way to understand an economic computer is by going through and understanding the motivations of all of the economic agents. And there is this new economic agent, brand new agent being born post-merge called the builder. The builder. And I absolutely had to understand more about the builder class of economic agent going into this. And so Matt certainly scratched my itch there. I feel like I have a far deeper understanding of all of the economic players in this economic computer and the builder class specifically. So stay tuned if you want to listen to all of that. We're going to get right into our conversation with Matt Cutler.

Hey Bankless Nation, super excited to get into this next topic. It's kind of geeky, and that's why I love it. And we are bringing to you Matt Cutler for this episode. We're going to talk about power structures in Ethereum. Matt is the CEO and co-founder of Block Native. Block Native is a team looking at the pre-chain layer of Ethereum, this thing called the mempool. I don't know if you've ever heard about it. But what he does with Block Native is listens to every single transaction that's ever broadcasted for chains like Ethereum uses that data to predict the future. So it's right down deep into Ethereum's computer at the firmware layer. But what we're going to talk about in this podcast is something that I think affects all of us. This is a post-merge topic, an Ethereum 2.0 topic, if I can say that, if that's still allowed. And we're going to predict the future of Ethereum based on the new power structures that will arise in the post merge, post proof of stake world. And this is going to be super important for you to tune into so that we can predict the future of Ethereum and the future of the crypto economy. Matt Cutler, how are you doing, man?

I'm doing great. It's wonderful to be here. Thanks for having me.

Can you help us with this topic? Because there's a lot of pieces here. There's a lot of like terminology. Some of it sounds a little geeky. Can you break it down for us in layman speak? Is that what you can help with?

Yeah, I would love to. There are definitely a lot of moving parts. There are a lot of abstract concepts, but they all come together in some pretty compelling ways, and it's a super fun topic to explore.

Okay, can we start with this? Because uh we are merging for a reason. We're getting rid of proof of work and going to proof of stake for a reason. But let's kind of start from the beginning because there are some reasons we're doing this. There are some flaws in the existing Ethereum system that we're trying to correct. And by transitioning to proof of stake, I feel like maybe some new flaws that we introduce. I'm not sure. Take us through this, Matt.

Sure. So I like to frame this up as: hey, look, I've been building companies for a long time. And in the 90s, the whole world went from being offline to being online, the world that we live in today. And that required a huge amount of infrastructure buildup because the internet in 1995 was no way ready for the thing type of things we enjoy today, like streaming video and social media and all that sort of jazz, right? Similarly, right now in 2022, we're in the beginnings of this transition from going from an off-chain world to an on-chain world. And as we look into the future, there's just massive demand for scalability, throughput, data storage, and blockchain systems. And the existing proof-of-work chain, you just can't get there from here while preserving decentralization. And so the merge and the transition of proof of stake is the necessary foundation for Ethereum to really start to stretch its legs and scale to the on-chain future that we're all anticipating. And so it's very much sort of foundational and core frameworks. It has a bunch of really interesting consequences and non obvious sort of elements to it, but it does set us up for the next generation of Web3 growth.

And Matt, just to really double down on this, let's talk about why proof of stake, why the merge preserves some of the values that we want it to preserve? Before we get into the conversation of how blocks come to be, because that's ultimately where this conversation will lead. I just want to really get into when we live on chain lives, and eventually the metaverse will come if we're using the metaverse as a way to describe this on chain future. When we live our on chain lives, why is proof of stake and why is the merge an important component in preserving the values that we want to preserve in the crypto ecosystem?

So, you know, fundamentally, you know, the basis of Ethereum is a decentralized smart contract platform. And so this core theme of decentralization, where basically anyone can participate in securing the network, is really quite critical. However, as you embrace smart contracts, all sorts of additional overhead comes into play. It's harder to reach consensus, it's harder to do execution, there's a lot of computation requirements, there's a lot of data storage requirements. And sort of the shortcut here is just say, well, you need a larger and larger computers to be part of the network. This has a fundamentally centralizing force because it pretty quickly gets out of the domain of your average consumer with an average laptop. And so what the transition of proof of stake does, and in particular, the very aggressive move towards modularity in this transition, is its preservation of decentralization through modularity. And what this does is get a lot of the benefits of, hey, we can break the pieces apart, you can decentralize them individually, folks can participate in many of these layers with different capabilities. And the net result is a much more capable network with much fewer externalities or external dependencies that still has a very low threshold for participation, where basically any consumer with any consumer grade laptop can participate equally. And that's really core if we're going to have a truly equitable Web3 and metaverse future. So I think this is critical stuff, and it's super exciting to be a part of it.

Right. And uh just to remind people, the we have this thing called the Ethereum virtual machine. And what that means is there is a virtual computer in the cloud that is created out of the shared resources of all of the computers of the Ethereum network. But in order to keep this thing decentralized, those shared resources have to be balanced and have to be inclusive. But we also want this super high-powered, decentralized, finance, metaverse future that requires all of this infrastructure. And people that have been listening to the Bankless Podcast, they know they probably haven't heard this modular design structure before. Ryan and I have done a few podcasts on it. And our first modular podcast was really all about the separation of the execution layer from the consensus layer. This is putting all of people's transactions and their speed, all their DeFi activities to have very fast transactions with low transaction fees by putting that on the L2, thinking of L2s as modules, where you have the consensus layer as the beacon chain, the proof of stake layer. But I think people that might have just listened to that uh podcast might not be aware of how further the modularity of Ethereum goes, because there's so many other components of modularity that are about specific parts of Ethereum. And this is again what this podcast is ultimately going to be about. But Matt, can you walk us through the parts of the Ethereum ecosystem that Block Native really has expertise in, where it is now? And that is in uh the process of block building, uh, because we ultimately want to go to how block building becomes modular. But let's talk about the current state of block building. If we can also as well define block building, talk about the current state of how a block is built and added to the blockchain. Can you just walk us through that process for us?

Sure. So the existing model is proof of work. And it's a fairly monolithic model, meaning everything is sort of all tied together in a fairly small and compact set of software and a fairly small and concentrated set of actors who are called miners, but are actually more likely mining pools. Okay. So basic proof of work consensus algorithm is you build what's called the block template, which is basically a candidate block with the transactions. And then you try to solve for the golden nonce, where you try to do a bunch of computation. To do the computation, you have to pour energy into computers. And so you have this externality of you need energy that you pay for in fiat, you need computers that you pay for in fiat, and you solve these problems. And the first one to solve the problem basically says, hey, I have a block, and they broadcast it to the network. And the other miners in the pool or the other nodes basically inspect that, validate it, and say it's okay. That's the top level.

Underneath the covers, it's a little bit more nuanced, which is you typically have what are known as mining pools. And so you have uh folks who contribute computing resources to a sort of a collective set called the mining pool. And the mining pool operator, the one who's sort of in charge of that pool, has a very critical function, which is they specify the block template. So the mining pool operators, of which they're not that many, there's just a handful really, are the ones who are basically responsible for block construction. They pick block templates, which they think are sort of the most valuable block possible based on the current contents of the mempool and other external factors sometimes. And they say, This is the block I want to mine and confirm, and then they send that out to their pool, and all the various computers in the pool churn on that to try to solve for the golden knots. Now, what that means is block building.

And consensus are kind of combined in this world. And that this idea that you have a fairly small number of people who determine what goes in a block and what the sequence of those transactions in a block is perhaps a bit more centralized than we would like over the long term. And so that's some of the design objectives and sort of design consequences of what's happening around the merge and some of the other external factors there to address and to create additional layers of decentralization at all levels of this process.

Hey Matt, can you uh just walk us through? I think a lot of the people listening are probably fairly distant from the actual process of mining and like don't even know when you say mining pools, I don't know that they have a vision for what that is in their head. What are these mining pools? Who operates them? Like what kind of entity are we actually talking about here? If these are the the folks doing the block building and also like part of the consensus, and you're saying they're all one, what kind of entity are we looking at? This isn't an exchange. Are these the mining operators that are out there that are kind of underneath the protocol?

Yeah, so these are the mining operators. So one of the major ones is called Ethermine, and you could read about them. And what these do, rather than Ethermine itself buying and controlling all of the GPUs necessary to do the mining, they basically operate software that lets anybody with a GPU participate in their pool. And so imagine I have a bunch of GPUs, which I don't, but they're under my desk. And I can say, well, on my own, I could mine, but I don't have a lot of hash power. And so I don't win very much. So I don't get a lot of income. It's not great. What I can do is I can take my GPUs and I can join them to a mining pool. And then I participate in all of the victories of that mining pool, all the blocks that they win, as a share of sort of how much hash power I devote to that. So what it does is it makes my life a lot easier in terms of my income because I spread it across everything that's that's happening in the pool. And what it also does is it I have less things to think about because I don't have to do the hard work of figuring out how to build a block. I just get a block from Ethermine. They tell me what to do, and then I spin my GPUs up behind that. So this is the mining operators underneath the network who are aggregating uh compute power, who are aggregating hash, and have this sort of interesting role. There's a bunch of these, but they're really less than you might expect. I'm not sure how many mining pools there are in operation today, but there's not thousands, that's for sure.

Yeah, five takes up the vast majority of all hash rate.

Yes. And so the five, you have five block builders, five entities, which specify these transactions are going to go in a block. And critically, this is the order of those transactions. And so, hey, one of the things that's critical about any public blockchain network is transaction sequencing matters a lot. And modifications to that sequencing can have pretty significant economic effects. And so you're in a situation where we would like to see more decentralization of block building, and we'd like to see perhaps more diversity of block building strategies than just the few that a couple of private actors can deploy today.

Okay, so we got these mining pools. These are doing the block building. And by the way, um, Bitcoin works the same way, right? They have mining pools, they have block building works the same way. In Ethereum, we have these five entities plus, one is, you know, of which is uh Ethermine. I bring my GPUs to a mining pool and it kind of smooths out my revenue. And you know, so I'm happier than doing it on my own. There are benefits from economies of skill in the block building function. Just refresh us really quickly on like the incentives here, the reasons for doing this. How am I getting paid, Matt? So I guess as a mining pool, or I guess contributing GPUs, a miner in that mining pool, I think I have three sources of revenue. I got the block fees, the block rewards, and then I have transaction fees, what remains, anything that's not burnt by EIP 1559. And then I think I have this other thing from all the block ordering you're talking about, and we call that MEV, minor extractable value or maximum extractable value. And that is basically some sort of fee that I get because I am ordering the blocks in a specific way, and someone has kind of paid me, incented me to order it in a specific way. There's some kind of arbitrage there. Can you explain those three categories in a bit more detail for us?

Sure. So the core of proof of work is when you mine a block, with each block comes a reward, which is called network issuance, where basically new ether gets created, and whoever wins the block receives that ether. Okay. And uh the same is true of Bitcoin and many others. Now, notably post-merge, that issuance rate goes way, way down, and we we can talk about that later, right? Two, the block is consisting of a series of transactions. Those transactions have network fees associated with them. And under EIP 1559, the regime that we're in today, the fee is split into two parts. There's the base fee, which is set by the network, and then there's what's called the priority fee, which is up to the person creating or entity creating the transaction. It's also referred to as the minor tip. So you take all the fees that are in the block.

And the base fee gets burnt, it literally goes away. And so that basically is has a deflationary effect on the circulating supply of ether. But those priority fees, all of them go to whoever won the block. So the mining pool operator who won the block receives those as well. Now, typically these are pretty small. I mean, it's enough to sort of incent them, but it's not a big thing. But under specific circumstances, these priority fees can get quite lucrative. And there can be a lot of competition to make sure these transactions are included in your block and that you win those blocks because there's some juicy blocks that are out there. Now, as a result of the nature of the financial systems and that and games that are played on these Web3 systems, sequencing matters a lot. And so, for instance, there's arbitrage opportunities where a user conducts a trade and on a decentralized exchange like Uniswap, and that puts a pool out of balance, right? That there's slightly more and slightly less of a certain asset. And that, hey, if I could get transactions surrounding that or just behind that, I can capture that little imbalance. Or, hey, this transaction over here is going to move the price of this asset on this exchange, but it's not going to move on the other exchange. So if I get right behind that, I can basically buy over here and sell over here and take a little bit of arbitrage. There's many of these types of strategies. Now

The challenge is only one or two people might participate in that, and they need to be in exactly the right spot in the slot. Okay. So exactly the right sequencing in order to capture that opportunity. Well, in the past, what this resulted in was like spam behavior. Folks would sort of just hammer the network with transactions to try to sort of crowd in and get there and crowd other people out, which was really bad for the network and really bad for like bloat and bad for fees. Okay. And so what started to happen was the miners would themselves say, Hey, if you really care about being in a specific order in a block, you tell me separately, offline.

Don't broadcast your transaction, just whisper it in my ear, and I'll process the rest.

Whisper it to me, but tell me specifically where you want it, and then tell me what it's worth to you. Okay?

Now, this is this notion of MEV or minor extractable value is by expressing your preference for ordering.

You can get beneficial outcomes for you versus somebody else. Now, this was starting to happen privately and darkly. There wasn't any transparency or visibility into it. And there's all sorts of negative problems here. Like the miner themselves could theoretically say, you know what? That's a great idea. I'm just going to do it myself and censor your transaction and sort of steal the opportunity. And so a research collective and company emerged called Flashbots, which is very well known in the ecosystem. And they basically built an open marketplace for this MEV. And so where you have actors called searchers who identify these opportunities, searchers submit what are called bundles and they provide a fee and say, hey, if you include my bundle with certain conditions, I'll pay you, hey, the miner, this fee for doing so. And it turns out those fees can be quite significant. And so this becomes a third source of revenue for miners under proof of work. You have the issuance for winning the block, you have the priority fees of the transactions included in the block. And then you have this sort of sidecar MEV for actors who specify specific ordering. And when you can deliver that ordering, they pay you an additional amount.

And real quick, just to be clear, when you were saying the miners just then, are you talking about the mining pools themselves? Are they the primary beneficiaries of the MEV right now? Are they like in the secret whisper chats?

Yeah, well, so what's interesting is most of the value associated with the sequencing winds up aggregating to the miner. And so, or the mining pool. And this is because they're the ones ultimately who have control over sequencing. And it turns out that the privileged actor in the network, he who can control or she who can control the sequencing of the block, stands in this really powerful position. And so, therefore, if you want to take advantage of that, you got to give them almost all of the opportunity that's there. And so this value and this power accrues. By the way, this MEV is specific to block building itself, not to mining. And so we said before there's a small number of actors who actually build blocks. There's five of them who control the vast majority of hash power. So those are the five that receive are the beneficiaries of this MEV. And in a pretty significant percentage, there are certain estimates that some classes of MEV, 97% of the value of what's there winds up going back to the miner.

The miner or the minor pool?

the mining pool operator, who then probably distributes that to the pool itself, but yes.

I just want to say this because this was kind of a breakthrough in my understanding of things, and it's part of the big theme of this episode is that there is this hidden force we've talked about called MEV, but there's also this hidden actor. You just called it the one with the power to sequence the blocks. You've also used the term block builder. And I think we're going to use those terms synonymously. But right now, the block builder is the mining pool. And so often, like, we don't really talk about the mining pools or the block builders in Ethereum, right? Everyone knows that there's miners, everyone knows, you know, that we're going to have stakers in uh post-merge. Everyone knows the other parties and the other actors, people that run nodes and this sort of thing. But we don't talk about this economic agent that is very much at play. And these are the block builders, the sequencers, the mining pools for now. But in a post merge economy, they're not going to be the mining pools anymore. And we'll get to that later. But I think, David, you wanted to follow up on this.

Yeah, just a fun quick story. I got into Ethereum via GPU mining. So this part of this story I actually have very close to my heart. And the way that you described this earlier, Matt, is like I had like something like 24 GPUs. And so if I mine solo for the about 12 to 18 months that I was mining with my 24 GPUs, I might not, probability-wise, it was like 50-50 where there where I was going to like mine a block in like a three-month time span. It was a total gamble. Like I might have mined a block. And if I did mine a block, I'd win two ether. It was actually three ether at the time. But in order to smooth out for my returns, I would just donate my hash power to a mining pool. Like all the small fishes in the big, big sea would all come together and we'd pool our hash rate into the mining pool so that we could share our dividends.

Amongst all of each other, and it would smooth out this very bumpy, inconsistent rewards of the hash power. And so it's like if all of your friends all bought a hundred lottery tickets rather than a hundred friends buying one lottery ticket, you're more likely to win. And then you just spread out the returns. And over time, this like smooths out their ROI. But what we are accidentally doing, and this was really before MEV was really a thing, but what we do know now is MEV is huge. And so when we give up the right of our own ability to be our own miners, we're giving that power up to a mining pool. And so mining pools, maybe we don't really talk about them a lot on banklets because we're more focused on the future of Ethereum. But mining pools and staking pools are largely the same thing. They're the same type of players, same kind of flavors. And so, in one part, I was giving up my hash power to join a larger pool, but that was also giving up my ability to construct the transactions inside of the block. I was giving up that power to somebody that had more power than I have. And the same thing you could say is also true of something like Lido, where you're putting your ETH into the Lido staking as a service app, but you're giving Lido the power to order those transactions, the Lido validator. So this conversation is the same, whether or not we're talking about proof of work hash power or proof of stake ether supply. And so just wanted to go through this story. And so as a result of me giving up my hash power or me giving up my ether, I'm delegating this power to somebody else who is able to extract MEV from the DeFi ecosystem and perhaps not pass it along to me. Uh, just wanted to compare and contrast the the proof of stake and proof of work side of things. And so, Matt, like can you kind of just like give us the bad scenario as to what happens if we don't solve this problem of decentralization of power with technology? Like, what happens if we were to let the mining pools just go rampant with MEB or the future staking pools or large stakers just go rampant with MEV if we gave them too much power? Why do we need to decentralize this power?

Well, again, determining which transactions go in a block and the sequencing is a privileged position, and it has huge economic value associated with it. And so the issue here is the level of reward, the level of value grows over time with these blockchain networks, with the opportunity to determine which transactions go in and go out, and those sequencing. And so there's ever greater incentives to have corruption at that layer. And corruption can be I favor one party over another party, I favor my own transactions versus the market's transactions. And you wind up with a system which is supposed to be, you know, fair and balanced and equal and is actually not that you have haves and have nots. And this is something we think a lot about at Block Native is sort of information asymmetries that exist in this ecosystem and sort of how everybody having access to the same set of capabilities is actually really beneficial. And so what the focus is as part of a certain dimension of the merge is to move to modular decentralization of block building and to create new actors in the core of the network. And therefore, there's probably going to be some pretty interesting power dynamics that emerge from there.

We've often on banklist called uh if you are the person who gets to construct the next block, and right now in proof of work, it's random. You don't know when the next block comes, but you still are constructing every single block as if it is going to be your block. And in proof of stake, you actually will get to know when it is your turn to construct the block. And so in that moment where you, as a validator, you as the miner who's proposing the block, you have what we call God mode. You have God mode over that one block, you know, blocks added to the blockchain in a serial order. And if it's your turn to produce a block, you get to determine the future state of Ethereum. You, for that one instance of time, you have God mode. And that's powerful. God mode is powerful. You get to actually determine the state of Ethereum. You get to bend the transactions. You could accept those transactions, but not those transactions. You get to embed your friends' transactions ahead of your enemies' transactions. And so it's nice that we get to distribute God mode across like an array of proof of stake validators from all across the world. And those can get to be individuals who are staking their ether at home. But if we accidentally have a part of our system that creates higher returns on capital faster than the individuals, then that God mode will accrue to one single entity faster than we would otherwise like. And so this is the conversation that we're having today is like the technology. Don't worry, Bankless Nation. We have the technology to fix it. And that's what this podcast is about. And that's what the second half of the show is. And so, Matt, can we get into the future state of Ethereum in the post merge world? How do we solve this problem of preventing one central entity from having God mode too much? Where does that conversation start?

Sure. So the transition from proof of work to proof of stake is rather than proving that you've done some work, i.e., burn energy through computers to solve a block, you stake your ether. So you need 32 ETH to be an independent staker. And what you basically do is you commit to tell the truth. And if you tell the truth, you get a little bit back. And if you don't tell the truth, in fact, if you promote alternate truths, then you get slashed. And the economic properties of this are such that it's highly secure and it requires vastly less energy. And it creates sort of room for a whole bunch more scalability at other levels of it. Now, this proof of stake is the consensus mechanism. It's how the network agrees on what reality is. But what's happening is we're beginning to split apart the block building aspects, the execution layer. And so by default, today, a validator will run a validation client. There's a bunch of them out there. There's like six major ones. And they'll also run an execution client like Geth. But what they'll do is they'll say, hey, Geth, give me a block. And Geth will give them the block. And it's pretty straightforward. They can manipulate that block if they want, they'll validate that, pass it along, and that's kind of the end of it. It's a fairly easy thing to do, but it as you mentioned, when you combine execution and consensus into the same actor at the same time, it creates room for corruption. It creates room for weird things to start to happen. And so what's occurring at the merge is this idea of block building, where a new set of actors will be enabled in the ecosystem who all they do is build blocks and propose them to validators. And those block builders

I'm sorry, Matt. When you say build blocks, I just want to double down on this definition. Building blocks means I'm a block builder. I listen to the mempool. I listen to all the available transactions. I order them up to the size of all the transactions that can fit into a block. And I have, with that order, I have built that block, and then I pass that block on to the validator who's going to validate this block. Is this correct?

Uh yes. So the block builder's job is to assemble the block. And what they do, which is quite interesting, is they bid to win the block. They say, Hey, I have this block. Hey, validator, if you use my block template, i.e., the you know the block that you've created, I'll pay you 0.2 ETH. Okay.

And then, David, you're running a block builder as well, and you build something and you say, I'll pay you 0.3 ETH.

Because I've built a better block.

Well, there's various theories here, right? One could be there's more value in your block, and so you can bid higher. The other is you'll take smaller margins. There's all sorts of interesting mechanics. But from the perspective of the validator, they just get a bunch of people who are saying, hey, pick my block that I've created for you, and here's the bid that I will give to you. And that value accrues, you know, pretty cleanly for them. And so it simplifies the job of being a validator. Uh, the validator now can run on very low end hardware. There's no sort of advanced sort of capabilities that they need or access to special compute power that the network sort of serves to them.

What the optimum block template should be, and they accept it and they get rewarded for doing so. And so what it does is it splits out the act of validation and the act of block building between two separate parties, which is a form of modular decentralization for the network.

But it requires this new class of actors called block builders to do this job because, as we said before, there's only a handful of people who do block building today.

And so the cool thing about this is that it is creating a process of block building that creates a competitive market. And I don't expect there to be a few block builders or maybe a medium amount of block builders. I kind of think in the future there's going to be a very, very large array of block builders that are all highly competitive. And for the listener who has listened to Bankless and has got really, really excited about if staking Ether and participating in the network validation, and you should be, I'll have to ask you, listener, do you also know how to extract MEV? Because I'm going to go ahead and guess the answer is no. But the cool thing is, all of a sudden, you don't have to worry about that because through the process of separating block building and forcing block builders to pay bribe you to accept their block, all of a sudden you do know how to accept MEV because it's built into the protocol. Am I tracking on this, Matt?

Yes, it democratizes access to MEV. Again, what's interesting about all of this is the value still accrues to the validator, just like today value accrues to the miner, but it can do so in a decentralized fashion and it creates all sorts of new possibilities for value to move in new and different ways for members of the ecosystem who don't participate in this value, potentially to participate in this value. And also there's some pretty interesting sort of ethical decisions to make as well as a validator. And so we're about to see something brand new the network's never seen, new actors, new behaviors, new possibilities with new outcomes. And so that's a pretty exciting consequence of the merge.

So I want to go back to just kind of establish this, like in a post-merge world, what the rules will be, right? And so we've talked about validators a little bit and we've talked about block builders. I do want to get back and just hammer home what those economic entities are actually like, right? And then I also want to talk about this third group, which we really haven't talked about too much so far, which is kind of the MEV searchers, right? These are the market makers, these are the people that um actually want the blocks to be built in a certain way, are exposing the arbitrage top type of opportunities from one Uniswap pool to another, for instance. Can we go through each of these roles on the new Ethereum post merge world and explain each of them? Maybe starting with the MEV searchers and then working our way towards a block being built and a validator approving it and saying, Yes, this is the one. Thank you for the bribe. I select this one. Let's walk through those from an entity perspective. So tell us first about the MEV searchers. Who are these people?

Searchers are a class of actors who evaluate pending transactions and determine are there mechanisms to extract additional value from them, right? So this can be front running, this can be back running, this can be sandwich attacks, this can be liquidations, this can be arbitrage, and others. There's sort of many, many forms of MEV. But it turns out this is not the sort of thing that you just sort of write a script and it does for you. It's a fairly specialized skill set that requires fairly deep expertise. And so you have.

Searchers. There used to be individual searchers, but it's so competitive now that generally there's not a lot of individual searchers. They're more small teams, all the way up to quite sophisticated organizations. And they do things like look at the transactions in the public mempool where Block Native specializes and say, every single transaction, is there a way to extract value out of this? And if there is, they say, yep, this one is amenable. There's some MEV. So by the way, not all transactions have MEV associated with it. So if I have a simple ETH transfer, I'm going to send 0.1 ETH to David. There's no MEV associated with that. And so that transaction, in fact, the majority of transactions don't have this effect. But if I'm doing certain things in DeFi, if I'm trading on a decentralized exchange, you know, if I hold a collateralized debt position that falls below a certain threshold, if I'm even trying to participate in a major NFT drop, that there's games that can be played with the sequencing of transactions and with constructing specific transactions in specific ways to extract value from that. And what the searcher does is a few things. They first evaluate as MEV there, they then create companion transactions that have to either go before or after or before and after. That's called the sandwich. And then what they do is they bundle these together and they say, hey, they pass it along to a third party. If you include this bundle, which includes the original transaction, which they didn't create at all, some user and maybe in their MetaMask or Ledger sets a transaction. The bundle includes that and the searcher transactions in a specific sequence. They wrap that together and they say, if you include this in a block, I'll pay you this price. Okay. And what happens is there are many searchers who compete for many of these opportunities. So sort of a classic, you know, typical ARB opportunity, which is pretty obvious, you'll have many searchers compete for that. And they'll bid up the price until almost all of the value of that ARB is in the bribe to the miner. And that's why it becomes minor extractable value, and the miner gets paid what the ARB is basically worth, with a small amount left over for the searcher.

And who are these people? Are they like high frequency traders? Is there a traditional finance analog to who these people are? Or do they have capital? Are they like funds? Are they just, you know, super geeky people at home, just like typing stuff in on their computers?

All of the above, right? They're shadowy super coders. They are large existing actors, you know, the HFT funds that are moving into the space. They are independent developers too, folks who really know how the infrastructure works, you know, at a pretty low level. Mainly they're building bots. So I always think about searcher as bot builder, and these bots are quite sophisticated. They require a lot of compute to sort of calculate various outcomes. And what happens is searchers tend to specialize in specific strategies. So you'll have a searcher and all they do is arbitrage, right? That's all they focus on all day long. And they're looking at all sorts of unique and unusual things in arbitrage. You'll have a different searcher that focuses on sandwich tax, right? And there's something that's worth saying here about searching an MEV is a lot of the MEV is benign, meaning there's value that gets created and they extract that value and sort of no transactions are harmed as a result. But there are plenty of forms of MEV which are not benign, which, as a result of the MEV, a user might have more slippage, or might have to pay a higher fee, or might get a different exchange rate. And so there are certainly consequences to this that many people who participate in these networks may not be aware of or exposed to and realize that their transactions being sandwiched and therefore they have less favorable settlement for what they're trying to do.

Right. MEV can certainly span the spectrum between just an actor that is making DeFi more and more efficient on the good side of things to perhaps what some people have called actual theft on the other side of things and the full range of everything in between, good or bad. Matt, there's something you said that I want to clarify on. You said searchers will create these transactions, they're specialized, and then they will bid up to the value of the MEV that they will create, and then you will send that over to the miner or validator. But that also sounds like the block builder that we talked about earlier. Can you differentiate between an MEV searcher and a block builder and overall how a block is constructed?

Uh sure. So before I do that, it's worth noting MEV is a function of all transaction systems, not just blockchain systems. So MEV exists in the stock market today. So it's just it's an unavoidable consequence of an ordered transaction system, and all transaction systems have to be ordered, and the ordering counts.

Any ledger has MEV.

Exactly. MEV is a fact of life. Okay. So it's not do you avoid it? It's more like what can you do about it? Okay. To your question, David. So the searcher

Basically, it specifies a sequence of transactions for inclusion, right? They basically communicate that preference to a block builder, i.e., a mining pool operator and proof of work. And they do so via a variety of mechanisms. One of them is the Flashbots has a marketplace for this. There's a group called Eden Network. There's a Blockstroud has some of this stuff. I think Ethermine, one of them has their own private mechanisms for these things. And so you have multiple marketplaces for searchers to submit their bundles to specify what they would pay for inclusion in that way. And there are certain trust assumptions that are built into that because each actor in that value chain can sort of see what's going on. And so you got to make sure you're everybody is doing what they say they're going to do and not playing further games. But the searcher is submitting transactions. The block builder, the mining pool operator in proof of work, is ordering transactions into a block. And basically, what this allows is the searcher to say, put these set of transactions in this sequence, maybe in this specific slot, specific order for this price. And then what the block builder or mining pool operator chooses whether or not to accept that. And again, there's many factors that go into that, how many bundles go into a given block? Because remember, there's only so much block space. And all of this MEV requires unique transactions to do it. So the MEV begins to crowd out regular transactions that can have some negative externalities. And you may have multiple bids for the same thing. So it's a dynamic, real time marketplace. People need to make decisions very quickly. And it's a pretty complicated domain today.

Okay, we have three parties here. We have MEV searchers, we have block builders, and we have stakers and validators. And this happens in a serial order. I think the best way to view this is kind of like as a funnel or as a pyramid, depending on what your orientation is. There's only one validator for every single block. So there is only one of those entities per block. And so that is the conclusion of all of these efforts of these other two parties. So they're at the top, they're at their very end state. They're the last thing that happens before a block is added to the blockchain. Searchers are on the other end of the spectrum where there are so many searchers. And imagine for a second, let's fast forward into 10, 20 years in the future where the Ethereum DeFi economy has taken over the world. At least that's where I expect the world to go. We know we don't have just like AMMs and like, you know, money markets. We have the full entire world economy being built on a blockchain technology. And so all of the world's financial ecosystem, all of the world's finance is built on blockchain. And in this world, imagine there is not going to be one type of entity that is an expert in ordering the world's transactions. And so what searchers are doing, what I'm hearing, Matt, is that searchers are specializing in one specific domain of the economy to make that economy hyper-efficient. Let's make sure all the AMMs are perfectly balanced, and searchers will arbitrage between sushi swap, uniswap, balancer, all the other AMMs, and they will balance those things out and they will extract that MEV and they will be hyper specialized at doing that. And then there's going to be different searchers who are good at liquidating money markets to make sure those are super efficient. So we got Ave, Compound, Merari Fuse Pools, Euler. We have searchers that are optimizing for liquidations on those, and they're highly competitive to optimize to make that happen. And so pick your flavor, pick your part of the economy, and searchers will find a way to make that part of the economy hyper efficient. And so if searchers will bundle up those micro parts of the economy, and then those bundles, I'm sure we've all seen a transaction on Etherscan that has like 15 different ins and outs. And this is kind of like what a searcher would look like, at least if I'm imagining it correctly. And then the searchers bundle up all these transactions, and there's a ton of searchers, but then they send them to the block builders. And so the builders take all of the bundles from the searchers, and then they are the ones that construct it into a block. And there's fewer block builders, I'm guessing, than there would be block searchers or MEV searchers. And so we're starting to consolidate and like funnel down all of the transactions in the mempool down to a fewer parties. So the searchers bid to block builders to accept all of their bundles, and then the block builders compete with other block builders to bundle up all of the searchers' transactions, and then to finally submit it to the person that is their turn to have God mode, which is the validator. Matt, did I get all that correct?

Yes, largely. So, first off, MEV bundles are multiple transactions. So you have transaction A submitted from the user, you'll put transaction B right behind it from the searcher. So instead of one transaction, you have two, that's your bundle. Okay. And so you can actually look in Etherscan. You'll see there's a little flag. I think it says non traditional ordering. I think it's something like that. And those are examples of MEV transactions where they appear in an unexpected sequence in the block. And that's the clearest indicator that there's some MEV there. So yes, searchers look at sources of transactions, they create responses.

Group those up into bundles, they submit them to various locations for inclusion in block building. Now, this is this new thing under proof of stake, and in particular, some new sort of sidecar capabilities through what's called MEV Boost, which is also from Flashbots, enabling sort of a new class of actor to be a standalone and specialized block builder. And this new role, which doesn't exist in the network today, is that anyone can build blocks. Anyone can look at the mempool, can look at other sources of transactions, can receive bundles from searchers, and can try to build an optimum block. And an optimum block is a block which occupies the maximum uh amount of gas that's available and that has the maximum value associated with all the various transactions, and then can say, well, this juicy block I want to propose up to the validator whose turn it is to have God mode. By the way, significantly, when the validator is in that position, they're known as what's called the proposer. Okay. So this will matter in a second. So the block builder creates a template based on all these transactions, including bundles from searchers. They then push that through MEV boost to the proposer who's a validator. And the validator says, this block looks great to me, and I'm going to suggest that to the network. There are some interesting nuances in how the various pieces work, but sort of the punchline to all of this is standalone block builders have not been a thing in Ethereum to date. And immediately post-merge, it's going to become a thing, and that's going to be a pretty interesting set of activities and actors. And now that you have independent block builders, they can start to do things that haven't been possible before. They can start to move value around in ways that haven't been possible before. And they can start to express sort of values in ways that haven't been possible before either. And I think there's going to be a lot of interesting possibilities that emerge from that.

So we want to talk about the block builder new character class that's like sort of entered the list of economic agents, right? And I guess before in the existing Ethereum today, uh block builders and kind of uh validators have been one in the same, been these kind of these mining pools. You know, like as David was talking about the kind of the efficiency and the specialization of these setups, particularly around the MEV uh search. It's so interesting to me because it's almost it's starting to resemble a very organic system. Do you know, like in the ocean, like cleaner fish, where you have these like fish that go clean other larger animals, like clean their gills, or they'll pick out the bacteria of a whale's teeth, you know, and they're just chomping away for all of the energy, right? They go take the energy. Well, this is how we sort of extract all of the economic energy from the Ethereum ecosystem, right? It's kind of a symbiosis, it's like a mutualism between different organisms. And that's what these MEV like bots to me are. You describe them as bots. It's like these little nanobots, these little cleaner fish going and finding all of the economic energy and just like biting it and cleaning things up. But this new character class, the block builder, okay? Can you talk about what this entity might be? And I wasn't entirely clear, Matt, on why this new entity has entered the picture. You mentioned something called MEV boost from Flashbots, but I don't think that's part of the core Ethereum protocol. That is something else, that is something third party, and yet.