019

/

August 10, 2022

Maximal Extractable Value (MEV) Explained

with

Matt Cutler of Blocknative

Apple podcast logoSpotify logoSpotify logo

Today, we're breaking down the basics of Maximal Extractable Value (MEV) with Matt Cutler of Blocknative.

Why? Because every blockchain's inherent MEV has important implications for its on-chain efficiency and security, but  many people aren't aware of why this is the case. MEV the maximum value that can be extracted from block production in excess of the standard block reward and gas fees. Given that there are often hundreds of thousands of pending transactions vying for limited spots on each block being validated, there is intense competition for block space and potential payouts for participants who find an edge. How does this actually work? How does it affect the gas prices, which taxes every trade? What does MEV mean? And how does it affect traders? Today, we're going to find out.

Matt Cutler is the founder of Blocknative, which is building embeddable widgets that translate complex blockchain technology into clear, relatable, human terms.

Mark Lurie:

Welcome to WTF Crypto, where we peel back the layers of the onion of the crypto universe to understand what's really going on and how it affects you and your portfolio. I'm your host, Mark Lurie. And as a caveat, nothing in this podcast is legal for investing advice. Today, we're talking about the Mempool and MEE with Matt Cutler of Blocknative. Welcome, Matt.

Matt Cutler:

Great to be here, Mark. I'm happy to be on and pleased to dive into the conversation.

Mark Lurie:

Awesome. You told me that any given block has 150 to 250 transactions and for context, block times range from a few tens of seconds on Ethereum to 10 minutes on Bitcoin. And you told me that at any time, there's actually 80,000 to 250,000 pending transactions to get into that block with only about 200. That is insane competition for block space. How does this actually work? How does it affect the gas prices, which taxes every trade? What does MEV mean? And how does it affect traders? Today, we're going to find out. So, Matt, can you have some context for what makes you such a credible expert in guide on the subject?

Matt Cutler:

It's a big topic obviously. And we, Blocknative have been building core Mempool and pre-chain data infrastructure for four years now. So, we're an infrastructure provider in web three. We support many networks, including Ethereum, Polygon, Gnosis chain, Phantom B and BCHAIN, Bitcoin, and we're engaged in a whole bunch of others. And this is what we do all day every day is work with wallets and protocols and dApps and L1s and L2s and exchanges and marketplaces to help them make sense of Mempool data and incorporate in their user experience and to make sure that they're not subject to hacks and exploits and to basically make the whole ecosystem better.

Matt Cutler:

And this whole area, I think, as you alluded to is often hidden and mysterious, but it affects everything about transaction settlement. And so, it's hyper relevant to anyone who's active in the category and particularly relevant to traders.

Mark Lurie:

And just for context, do you have a long background in infrastructure entrepreneurship, right?

Matt Cutler:

Yeah. So, depending on how you've count, I've done seven or eight startups in various industries, but mostly in technology. My very first startup was a core infrastructure company and the Web 1.0 days, I founded that while I was still an undergrad way back in 1994. So, long before it was fashionable to do internet infrastructure, long before it was fashionable to do as an undergrad. I was building and that one was called NetGenesis. It was the first ever web analytics business. And I like to say it was a nine-year overnight success, zero to IPO. We went public at the height of the Web 1.0 boom. And strangely, at the time we went public, we were one of the top 35 IPOs in history, which is crazy. And my most recent startup was a mobile collaboration platform called collaborate.com that wound up getting bought by Cisco.

Matt Cutler:

And so, I worked in Cisco's core collaboration business unit and actually across their entire portfolio. So, deep core networking infrastructure for about four years before I started Blocknative, but I've done a bunch of different things and a bunch of different categories in technology. And actually, that's a lot of what attracted to me to Web 3 is, once I started to tune in and pay attention, I sort of felt like, "Gosh, I'd seen this movie before." And it felt eerily familiar to the formative stages of Internet 1.0. I think that this has played out quite a bit. And I think we're still pretty early in that whole transition.

Mark Lurie:

It makes sense. Okay. You have been around the block. So just out of curiosity, given all your experience, you said you've seen this movie before. How much of a nail biter is Web 3 compared to the other bubbles and environments that you've built in?

Matt Cutler:

Oh, I mean, each one has their own unique characteristics So, I was an undergrad at MIT and I got to MIT in 1991 and all over campus were computers and clusters called Athena clusters. And they were all connected to each other and you could message anywhere and your schoolwork was there and they're connected to other schools. And I happen to live in a fraternity across the river in Boston. And we had our own Athena cluster inside our fraternity. We could do schoolwork there and that was just going to school. And then, we later learned that's the internet. And it turned out ...

Mark Lurie:

Well, I've heard of that.

Matt Cutler:

... that we just had some of the best, well, no one had at that point. I mean, it was just tiny research network that was just way below the radar.

Matt Cutler:

And I tell that story, because if you were at MIT at that time and you had this access, and by the way, Stanford was the same way. It was so obvious that this was the future. This was how things were going to be. It was so much better than going home and using your computer in isolation or being on dialup or using any of this stuff. And then you realize, "Oh, we just have advanced access to the future at school." And it seems pretty obvious that this is how things are going to play out. So, I tell the story, because if you were inside the internet in the mid 90s, it was really obvious what the future was going to be. It was going to be a whole lot more internet.

Matt Cutler:

But if you were outside the internet in the mid 90s, it was not obvious at all. There was many, many potential outcomes, right? And of course, there were many twists and turns and it took a long time to get there, but we all know how the story played out. And I say it's the same in Web 3 because I really started to pay attention in about 2017. And once I started to tune in, I had exactly the same reaction like, "Oh, this is way better than all the alternatives. This is superior."

Matt Cutler:

By the way, it's hard, it's weird to access. You got to dig into these weird forums. There's a bunch of junk out there. This was all true to the early internet too. And so, the thesis is just like the whole world went from being offline in the early 90s to being online today, right now, the whole world is going from being off chain to on chain. And to me, that seems like a foregone conclusion. There's no question that this is the future. No. Which chain? How's this going to happen? How much time? What twists and turns? There are still a lot to be played out. But we're going to live in a non-chain future. That much is clear, right?

Matt Cutler:

And it is merely now, what set of activities do you choose to embark on to participate in that? This is really obvious to those of us who are inside Web 3 and not at all obvious to folks who are outside Web 3. And again, this is why I say, I've seen this movie before. It doesn't mean I can predict what the future's going to be, I know all this stuff. It's just on a long enough time cycle. I think the destination is just as clear.

Mark Lurie:

Yeah. I'm sold. What a coincidence that we're here.

Matt Cutler:

Yeah, here we are.

Mark Lurie:

I'm sold and we're talking about this. But I think a lot of listeners to this podcast listen and some of them are already deep in crypto, but a lot are just getting into crypto and want a common sense first principles-based conversation that gets into like, "Okay, what's actually going on?" And so, one of our goals is try to make that clear to people who it may not be obvious to yet.

Matt Cutler:

Sure.

Mark Lurie:

So, with that said, let's dive in. What is the Mempool and why is it even relevant?

Matt Cutler:

So, we like to say, "An informed user is an empowered user." So, we think it's in everyone's best interest to really understand how this stuff works, because you'll be able to navigate it with a lot more confidence. So, okay. Public blockchain networks. By the way, there's Ethereums of public blockchain network, Bitcoins of public blockchain network. Solana, Avalanche, there are many. What happens is users submit transactions. And then the network needs to evaluate is the transaction valid and as it gets included in a block, it is written down to the chain. Okay.

Matt Cutler:

And so, by definition, there's got to be some place for transactions to go to be candidates for inclusion. Because if you could just write directly to the blockchain, you wouldn't have a blockchain at all. You'd have a database, right? And so, this place where transactions go to be candidates for inclusion is commonly known as the Mempool. Now, Bitcoin has a Mempool. Ethereum has a Mempool, but not all chains have Mempool. Some use slightly different techniques and terminology. So, we often will generalize to the term of pre chain, which is sort of all the stuff that happens between a transaction being an idea and getting on chain.

Mark Lurie:

So, this is the emotional purgatory when I click confirm on a transaction. And then I hold my breath, hoping that I didn't mess anything up and then before it actually gets completed and finalized.

Matt Cutler:

Exactly. And so, I often will ask that question, when you press submit and then you wait for a while and then you get a response if it's confirmed or failed? Have you ever thought about what happens while you're waiting for a while? That's the Mempool. And it turns out a whole bunch of really fascinating, really important things are happening, but you have no visibility into that and no understanding of that. And by the way, Blocknative provides infrastructure to correct that, to basically make it. So, you do have visibility, you do have transparency, and you can use this to your advantage because there's certainly actors out there who are, right? And so, we think everybody needs to be aware of the pre-chain layer and Mempool and understand what's going on and basically be aware that there may be opportunities to profit from that.

Matt Cutler:

And there are certainly situations where you may be harmed in that situation. And by harmed, I mean, get less favorable settlement or no settlement depending on the circumstance that you're in. And so, it's often referred to as the dark forest. The dark forest meaning, it's dark, you can't really see along and it's full of predators that don't have your best interest in mind. It's a fascinating space. It's super fun. It's rapidly evolving and it's challenging to operate in, which is why we do what we do and we've been at it for four years. It's not the sort of thing you spin up in a weekend, that's for sure.

Mark Lurie:

Okay. So, that makes sense. I'm following you so far and I am embarrassed. But willing to state that while I should know a lot about how that works, I actually have very little conception of it. So, who decides what transactions make it into a block? How does the money work? Because usually that explains a lot and what are the tricks and treats, fun games that people tend to play?

Matt Cutler:

So, okay. It's sort of intricate and complex. So, give me a moment. You use a wallet to submit a transaction to the Mempool. Now, how you submit that is you actually need an entry point into the network. So, you might run your own node. So, your wallet is configured to talk to your node and then your node has peers. So, what happens is your wallet signs a transaction.

Mark Lurie:

And a node is an instance of blockchain database.

Matt Cutler:

Exactly.

Mark Lurie:

It's kept in sync with everyone else's.

Matt Cutler:

Exactly.

Mark Lurie:

Part of an actually running part of the chain.

Matt Cutler:

Yeah. It's participating in the chain. And what happens is the nodes have peers in the network where they communicate with each other over the peer-to-peer layer. And so, basically, your wallet signs a transaction, shares it with your node, your node inspects to make sure it's valid, i.e., it's a proper signature, you have the balances. All these sorts of things. And if it accepts it, inserts it into its Mempool and then propagates it out to its peers. And in Ethereum, by default, there's about 50 peers. And all those peers do all the same work. Again, they receive the transaction, they inspect the transaction. They make sure it's valid. If it is, they insert it in their Mempool and they spread it along.

Matt Cutler:

Now, there's a couple of really important nuances to that, which is how is it priced i.e., one of the fees associated with it. Because if the transaction is underpriced, it's not a candidate for inclusion. And so, on Ethereum, you have what's known as the base fee. And if your transaction is below the base fee, you're out of luck. It's not marketable until the base fee comes down and it becomes marketable. Or if the network is very congested, there's something going on. There's a lot of people that are trying to get transactions. Your transaction may not be one of the top few thousand. And so, many nodes will say, "I don't have room for your transactions." And so, it gets ignored. So, when you say, "I submit it to the Mempool," there's no such thing as the Mempool.

Matt Cutler:

Every node of the network of which there are thousands has its own unique copy with its own unique data set. It makes its own unique set of decisions. And so, it's this sort of interesting probabilistic environment. And so, there's all sorts of interesting characteristics that emerge from that. But because there's no consensus, there's no truth. You can overwrite stuff in the Mempool. You can replace your transaction, you can speed them up, you can cancel them and other stuff. So, in many ways, the rules of the blockchain are the opposite in the Mempool.

Matt Cutler:

Now, what happens is your transaction propagates through the network and eventually it finds its way to miners. So, miners are the entities who, today at Ethereum, and by the way, this is going to change in probably just a few months do what's called proof of work. And so, what they do, these minors is they basically try to assemble transactions into a block. And then, they try to solve a difficult computational problem to try to win the block. And if they do so, they get the opportunity to write the block to the network, have it become the truth and they get a reward from the network for doing so.

Matt Cutler:

Now, it's actually a little bit more subtle than that, which is the vast majority of hash power on the network is controlled by what are known as mining pools. So, you'll pool their computing resources together and there's a whole bunch of advantages as to why. But there's a handful of mining pool operators, less than a dozen. What the mining pool operator, the folks who orchestrate the pools, they're the ones who set what's called the block template. They're the ones that decide which transactions go in and critically in what sequence. It turns out in an ordered transaction system...

Mark Lurie:

So, we talk a lot about decentralization, but really there's 12 people essentially or companies that everyone's delegated to. And these 12 people actually control a lot of what gets put in the [inaudible 00:14:27].

Matt Cutler:

They control what goes in and in what order, which determines a lot of value. I often challenge people. So, can you name a big mining pool? They're like, "No." Do you know where they are headquartered? Do you know who operates them? And it's just this incredibly opaque ecosystem. And that, by the way, for the record, they're generally good actors right there. There hasn't been a lot of problems.

Mark Lurie:

So, this isn't like a smoke-filled room where people decide the future of the universe?

Matt Cutler:

It's an interesting conversation. But first off, people are sort of surprised by this. By the way, fast forward. The Ethereum network is going through this notion of the merge. And as part of the merge, this problem of there's a small set of actors who decide what goes into a block is going to get changed in a big way. So, it's a big upgrade. One of the many hidden upgrades of the merge, it's not just proof of work to proof stake, but a bunch of this other stuff.

Mark Lurie:

Okay. All right. So, we won't harp too much.

Matt Cutler:

Well, it's a known issue in the ecosystem that is being addressed proactively, which is one of the big benefits of Ethereum is that it's very technologically progressive and it's self-critical and says, "Hey, here's something that we think could be better." And they do the hard work to try to address it. Now, interestingly, so as you might imagine in an ordered transaction system, which are all transaction systems because they don't have sequencing, you can't really sell balance. I give the example of, I get $5, I pay $20, I get $35, I pay $100. So, what happens is, well, it depends on the sequence, right? Because if I get the $35 first, then this one's going to clear and that one's not.

Matt Cutler:

So, the sequence really matters. And it turns out that in many ways, sequence matters a lot. So, imagine you, Mark, conduct a trade and that trade moves the price of an asset on a certain decentralized exchange. And now, there's a price differential between that asset on Exchange A and the same asset on Exchange B. And that's called arbitrage. And there are actors in the ecosystem. They're generally called searchers, but they're really just automated systems bots that detect these. And then they say, "Well, if I can get my transaction immediately behind marks, I can capture the arbitrage that results and make a profit."

Matt Cutler:

And in the past, there were all of these techniques that would be used to try to get specific sequencing in order to capture these sorts of opportunities. But they played out in public and they tended to have a lot of negative externalities or consequences, like bots going to war that would bid up the gas price. So, two bots or a few bots are trying to make something happen and everybody on the network pays higher fees as a result. Not great.

Mark Lurie:

Yes. Okay. So just so I make sure I track you.

Matt Cutler:

Yep.

Mark Lurie:

So, I'm making a trade in Dex. Let's say, I trade $1,000. Okay. That, because of the way Dex is work is going to shift the market up. Let's say, I buy something. It's going to make it a little bit more expensive. So, what happens is a bot comes in, sees that I'm trying to get a transaction and pays higher gas, which is what incentivizes a transaction to be prioritized. And they buy that $1,000 first. Then, I buy the $1,000. It moves the market up a bit even more. And then they sell $1,000 and they make a little spread. But this is probably like pennies, right? I mean, do I care that much?

Matt Cutler:

So, okay. We're actually talking about two separate things. So, what you talked about is, it's called the sandwich. And a sandwich is, "Hey, because I can see that Mark's going to do a transaction, I can bid up the price of that asset and give him a less favorable exchange rate. And then, as soon as this transaction happens, I can sell behind it and I can capture that Delta." And the consequences, if I do that, I get a profit, but Mark actually has less favorable settlement. He gets a less favorable exchange rate. And this is typically materialized in the form of slippage, where you always go, "Oh, I don't have so much slippage." Just because it has happened to you, if you've never traded on Uniswap with any significant value, this is the sort of stuff that happens all the time.

Mark Lurie:

But order of magnitude, how much does [inaudible 00:18:36]?

Matt Cutler:

MEV is conservatively estimated to be over a billion-dollar market per year today on the major networks and growing. It really depends. So, MEV takes many formats, right? Some of it is arbitrage, exploiting price differential. And by the way, arbitrage is constructive for the network because it helps equalize prices. There's arbitrage in the form of liquidations. So, an Oracle is going to update its price. And as a result, certain collateralized positions are now undercollateralized and become eligible for liquidation.

Matt Cutler:

So, by then, what you want to do is as someone who cares is you want to put your transaction right behind the Oracle update because that will ensure that you capture the liquidation. This is good for the protocol because it ensures that the protocol is operating efficiently and is properly collateralized. It's not great if you're subject to liquidation, but that's going to happen anyway. So, there's that. There are malicious forms of MEV, like sandwich attacks, like we talked about. There's front running, "Hey, there's an NFT drop that's happening. You want to get in? It's really hot." People just jump in front of you and they get into the NFT drop and you don't. You're out of luck.

Matt Cutler:

And so, there's many different forms of MEV. And by the way, some of them are quite exotic. So, they are cross chain MEV where you're spanning chains. There's multiblock MEV. So, there's all sorts of degrees of sophistication, each of which come with different forms of risk, but all of which get expressed in the form of specifying ordering in the block. And so, what was happening was this is just a fact of life. I mean, preference on ordering is valuable. All transaction systems have it, whether you're a stock system or whether you're a blockchain.

Matt Cutler:

And this was all happening in the open, which was impacting gas prices and making it more expensive for everyone and happening in smoke filled groups. Meaning, each one of these mining pools were cutting side deals and it was really opaque. And a research project got spun up known as Flashbots, which is based on a core piece of research that one of the founders of Flashbots created called Flash Boys 2.0, which was all about the new games and tricks that can happen inside the network.

Mark Lurie:

And Flash Boys is the name of the book about high-frequency trading.

Matt Cutler:

Yes. In the world of stock markets, which by the way, then Flash Boys 2.0 or many of these same ideas got applied to public blockchain networks and Flashbots was formed first as a research collective. Now, it seems to be more of a company to basically create a marketplace for this stuff, to bring this stuff, at least somewhat out into the open. So, what it has done, they provide tooling for this, but also, they've accelerated the market a whole bunch. And so, as a result, there's a lot more MEV. It's a lot more efficient to extract. There's a lot more competition. And so that the economics of it happen.

Matt Cutler:

But probably, the most important thing is if you're not paying attention to the Mempool, it's like playing soccer against a team that can see five seconds into the future. That you're operating on past data and they're operating on future data and that's going to have bad outcomes for you. And so, for any trader or any actor in the ecosystem, we believe it's critical that they are aware of Mempool data. They're aware of these techniques. They're aware of what's happening their transaction as it is pending and they take adequate measures to either participate or to protect themselves. And it's amazing how many times we engage with folks and they go, "Oh, I've always wondered about this. I can see things happening." And I'm like, "What does everybody else seem to know that I don't know?" They're operating in the Mempool. They're using private networks.

Mark Lurie:

Yeah. Exactly right now.

Matt Cutler:

And so it does have this Alice in Wonderland feeling where you're like falling down the rabbit hole and you get to see how things really work. And perhaps, the better one is like a Wizard of Oz, and hey, let me pull back the curtain and show you what's really going on in the network. And so anyways, this wild and interesting world of Mempool data, of MEV and of what we think is starting to change around it. And there's tons of resources out there. So, we're [inaudible 00:22:50]. We have a blog that talks all about this. But you could certainly do research on these topics and find thousands of reasonable and relevant articles, podcasts, and videos.

Mark Lurie:

Okay, cool. So, I have a few questions and then I want to pull out and maybe actually talk about the future, which is at least, Ethereum, proof of stake. But just some questions on this first. I mean, you mentioned it's about a billion-dollar market a year. There's about, let's say a hundred billion a month index volume, right? So, it's almost like 0.1% of trading volume a month is extracted by MEV. And on the one hand, I'm like, "Well, that actually is a lot." On the other hand, I'm like, "Well, day to day, most people probably don't care that much." But it sounds like it impacts, it's not like it's a consistent tax. It impacts you much more sometimes and much less others. So, that makes sense.

Mark Lurie:

And then, the second thing is that it actually raises gas prices for everyone. And that's perhaps an even bigger tax on the system that inhibits the ecosystem. Am I thinking about that right?

Matt Cutler:

Not all transactions have MEV associated with them. So, for instance, a simple ETH transfer has no MEV associated with it. So, you don't need to worry about it there. Generally, the more complex the transaction, the more likely there is to be MEV. And then, there's plenty of transactions which have MEV, but the MEV is so small. It's not worth the marginal costs to extract it, so it just gets ignored. So, there maybe 35 cents of MEV, but you're going to eat that up in transaction fees anyways, so why bother?

Matt Cutler:

But if you're a trader who's moving, doing real trades, this is absolutely something that you need to be aware of. And it's like, "Hey, do you care about a few percentage points of slippage in your trades?" You should, because that's often the difference between success and failure. And this is where this stuff can come from. And so, it is absolute material for every actor on the network. And it is highly asymmetric. Meaning, there's certain circumstances where this matters a lot and there are many circumstances where it probably doesn't matter that much, but it's a fact of life that everybody should be aware of.

Matt Cutler:

So, by the way, that estimate is a pretty conservative estimate. The MEV market may be significantly larger than that and it's growing. And so, it's growing overall. Some of this, it's often referred to as a tax. Some of this is benign. Meaning, the arbitrage prices have to equalize. That's how this happens. It's no big deal. Some of it is malicious where you literally get less favorable settlement than that is an actual tax. It impacts gas prices sometimes, but not all the time. And one of the big impacts of Flashbots is it moved a lot of the activity that was driving up gas prices to a private marketplace where that doesn't happen.

Matt Cutler:

And so, in the past, yes, this was a huge driver of fees on the network and of all sorts of forms of network congestion. Today on Ethereum and most other chains, it's not that big of a deal, but on certain chains, you'll hear about, "Oh, we have really low transaction fees." That's bad. Why? Because if the transaction fees are very low, bot operators can spam the network with tens of thousands of low fee transactions in pursuit of some profit opportunity. And then, guess what? The network gets congested. And if it's not very decentralized, the nodes get bogged down and then it stops producing blocks, and the chain stops and then the Oracles can update and then...

Mark Lurie:

Low fees can become a victim of its own success.

Matt Cutler:

This is interesting. Any chain which winds up getting reasonable economic value will have to increase transaction fees. Otherwise, they get swamped with junk transactions. It's just a fact of life, okay?

Mark Lurie:

That's a big part of why, at least in DeFi, I don't think order books will ever work on any chain because the faster the chain, the more viable it is to have an order book or the more viable it is to have an order book, the more bots and the more congestion you'll get and the more trades you'll get. And then, it'll raise transaction costs, slow it down. And it just becomes a big [inaudible 00:27:12].

Matt Cutler:

And we've seen that play out with many of the L1s. And it's just really interesting, you create economic incentives and they get exploited, right? And you can't really count on people not doing harmful things to the network. There ample precedent for exploiting network anomalies for profit, particularly in an anonymous fashion. So yeah, there's an equilibrium that forms and that equilibrium's in existence on chains like Ethereum already. And it's starting to form elsewhere. We saw this happen on Polygon where they had to raise fees and then it stabilized a whole bunch. There's been a lot of network instability on Solana driven by exactly these consequences.

Matt Cutler:

And by the way, the more decentralized the network gets, the more equitable the access to this information becomes on certain chains where they are highly concentrated, where there's only a handful of node operators. To be a node operator, you have to stake a whole bunch of resource. You're basically entrenching information asymmetry where there's haves and have nots. You have a small number of well-resourced actors who have view into the future, and you have everybody else who doesn't.

Matt Cutler:

And guess what happens? It just creates all the conditions for inequity. And so, we think it's really critical that L1s and L2s provide equitable access to this data. So that at least, it's a level playing field. And then, what we do at Blocknative is we provide tooling to ensure that everybody can get access to it, so you don't have to build a whole bunch of infrastructures on your own. So anyways, there's a lot of things happening all at once in this, and then it's I want to be a part of.

Mark Lurie:

Yeah. Okay. So, two follow-on questions from this. The first is, let's take as decentralized and as transparent a network as possible, okay? It seems like if everyone has access to this information and you are providing infrastructure to make it easy to access the Mempool and Flash Boys or Flashbots are providing tooling to write these bots easily, it strikes me that if there's ever a time when the efficient market hypothesis applies, it's here. How is it possible to make money doing this when it's all public, anyone can do it, and you just have to spin up a bot? I get that Joe Schmo on the street can't do that, but there's a lot of developers, plenty that it should really compete a way-out profit. And I guess I don't understand why this is for the profit.

Matt Cutler:

Well, so the truth is, is that, it's not profitable for common or obvious forms of MEV. And so, here's what happens. You identify a sequencing opportunity. So, I want to capture this arbitrage. I want this transaction to go behind that transaction. You then have to bid what you're willing to share of that to get that sequencing. And it turns out that today, about 95% of all MEV by most estimates goes to the miners, goes to the mining pool operators. So, you're operating this software infrastructure to identify a hundred dollars opportunity. And what you do is you give $95 of it away to keep five, okay? Because if you don't, someone else is going to bid $95 and you're not going to get the opportunity at all.

Matt Cutler:

So, first off, many forms of MEV are hypercompetitive. And as a result, they're very low margin. What this does is force folks to find more exotic or esoteric forms of MEV, where there's not as much competition or they go to other chains where that marketplace isn't as well developed. But the fact of life is today under proof of work on Ethereum, the mining pool operators sit in this privileged position of specifying sequencing. And they control a lot of power as a result. And certainly, they deserve to participate in all of this because they provide a function, but it doesn't seem particularly equitable that they take 95%, the searcher index 5%.

Matt Cutler:

And by the way, the whole reason why this happens is because Mark did a transaction and Mark's not involved at all. This is completely invisible to mark. And in many cases, Mark is actually harmed. So, look, if Mark doesn't do his transaction, there's no MEV. And so, what's emerging, what we see is coming as part of the merge and some of the upgrades associated with it are the conditions where that won't necessarily be the case. Where in the future, any old user using a sufficiently intelligent wallet can participate in this, in ways that are quite meaningful and we think are a lot fairer.

Matt Cutler:

If you think about where the value originates from and where it should go to moving forward and that ultimately is in the best interest of the ecosystem and the growth of Web 3. Remember, we started with the whole world is transitioning from off chain to on chain. And we think of that transition is going to happen, we have to be fair in how all these systems work and how the money flows. It's not very fair today, but it might be a whole lot fair tomorrow.

Mark Lurie:

Okay. So, we should definitely go into the merge then, because it sounds like that's the key to understanding the future. I want to flag that there's one issue that I didn't quite understand, which is, you said it solved, part of this is solved. The gas problem is solved because a lot of this MEV is actually happening off chain in a smoke-filled room. And I was wondering if before we go into the merge, you could just explain that comment, tie that loose thread.

Matt Cutler:

Sure. So, much of MEV today runs through the Flashbots marketplace and they provide a fork of Geth. So, Geth is one of the major nodes on the network called MEV Geth. And most of the mining pool operators now operate MEV Geth. And what MEV Geth includes is a separate side Mempool. And what happens is searchers, bots, can submit transactions, bundles of transactions into the side Mempool. And what they basically say is, "Look, here's my target transaction, Mark's Trade, and here's my transaction behind it. And I want them to go in the block in exactly the sequence. So, if Mark's transaction goes in slot three, I need to be slot four. Because it's the only way I can assure that I get the arbitrage opportunity and if my bundle gets successfully included, here's what I will pay for that to happen. So, I'll pay a hundred dollars for this to happen."

Matt Cutler:

Now, critically, if my bundle does not get included, I pay zero. And so, what it does is it prevents all these gas war from happening because there's this private marketplace where people bid and there is, it's you either win or nothing, right? There's no risk where you're going to sacrifice your gas or things like that, which is what was happening before. And so, by having this private off chain marketplace for transaction sequencing preferences to be specified it allow, and by the way, the price used to be specified in the form of gas, now there's a completely separate mechanism, independent of gas.

Matt Cutler:

And so, as a consequence, gas fees are an actual reflection of demand for block space, not a reflection of MEV games between automated systems, which is often what happened before. That certainly means there's volatility and gas prices. It certainly means there's variable demand for block space, but these automated systems are much less likely to create these sorts of conditions. And for those of you who've been around in the network for a long time. In 2018, 2019, this was pretty common. And all of a sudden, out of nowhere, gas would go crazy. And then it would go back down right away, right?

Mark Lurie:

Oh, yeah. It was impossible to actually build it into [inaudible 00:35:04].

Matt Cutler:

It created all sorts of problems, right. By the way, EIP-1559 got implemented in part to address some of this. It doesn't always work. So, there are many conditions where we do have priority gas auctions today, but it all goes to the priority fee for NFT launches and things like that. So, it's not perfect, but it's a whole heck of a lot better than it was. And again, this is the story of Ethereum, it's technologically progressive.

Mark Lurie:

Okay. So, that makes sense. And normally, I would go down this rabbit hole because it's so interesting that you're disaggregating gas price from this kind of gaming. And on the one hand, I'm thankful because it makes my day to day usage better. On the other hand, it does feel weird that there's this off chain, private B2B market where that decides what goes in into the block. And so, I won't unpack the implications of that for our ecosystem because it sounds like we're going somewhere else. So, let's go there. And I hate to have you do this, but I think it's important so that we can all follow along with the story.

Matt Cutler:

Sure.

Mark Lurie:

Can you explain the merge? And then, how it actually affects proof of stake? And then we can...

Matt Cutler:

So, the Ethereum network goes through upgrades from time to time where the rules change and everyone agrees on that. Well, the biggest upgrade in its history since its inception is coming and it's known as the merge, okay? It was previously called the Ethereum 2.0. So, if you've heard that phrase, that phrase was retired and it's the merge and the merge is the transition of Ethereum from proof of work, consensus algorithm, how it works today to proof of stake, consensus algorithm, how it's moving forward. And this is sort of hard to understate how big of a deal this is.

Matt Cutler:

Now, significantly, it's called the merge because right now there are two networks operating in parallel. There's the existing Ethereum 1.0 chain proof of work with all the balances chumming along and there's the Beacon chain, which is the proof of stake network that doesn't have any activity that's been running now for over a year. And what the merge is, the Beacon chain and the Ethereum 1.0 chain are going to merge into a single network. All the balances are going to get carried over. So, your balances that you have today are going to be the same moving forward, but how the network agrees on what is truth is going to transition from solving hard computational problems, which by the way, the only way to prove that you're doing that work is to expend energy. And this is actually the proof of work.

Matt Cutler:

I'm proving I'm doing the work because I can't prove it by voting because I could send up a million accounts pretty easily and they can all vote one way, but with proof of work, there's no way to fake having done the work, which is why these networks are very secure, right? The good news about...

Mark Lurie:

And this is what causes the energy usage for Bitcoin more than the energy use of the [inaudible 00:38:08].

Matt Cutler:

So, as these networks grow, they wind up consuming a lot of energy, right? And by the way, that may be a perfectly valid use of that energy. I'm not here to debate that, but the truth is, it's proof of work is energy intensive by design. Proof of stake is an alternate consensus mechanism where rather than...

Mark Lurie:

It's also, incidentally, a hard cost for running the network.

Matt Cutler:

Yeah.

Mark Lurie:

Which has to be paid by users and makes things more expensive.

Matt Cutler:

Yeah. And by the way, there are some interesting things, which is like, how come gold mines don't just take all the gold out of the ground. And the reason is, it's expensive to mine gold. And it turns out that when the price of gold goes up, more gold gets mined because gold that's hard to access becomes profitable to mine. But basically, the margins are very small because you have to have equipment and fuel and permits and people to mine gold. And so, by the way, you don't pay for all that stuff in gold, you pay for that in dollars. So, most of the gold that the mines get out of the ground, they sell for dollars to fund their operations.

Matt Cutler:

The same is true of mining and proof of work, right? That even though you're doing all this Bitcoin or Ethereum, you actually need computers. You got to pay for those in dollars. So, you got to sell your Ethereum or your Bitcoin to get the money to buy the computers. And then by the way, you got to power them with power that's also denominated in a hard currency, typically dollars. And so, there's this weird, the network is secured by this compute power, which is purchased with this other asset. So, you have these external dependencies and that's viewed as not great because the price of the dollar fluctuates, the price of computing power shifts. And it creates a lot of forces on the network that are sort of hard to predict and control.

Matt Cutler:

So, what happens under proof of stake is quite a bit different. Rather than basically solving a hard, cryptographic problem, you stake resources. You basically put assets up and you say, "If I tell the truth, I get a small reward. And if I don't tell the truth or I play games, I get slashed." And the way the economics work is you're always incented to tell the truth because lying is expensive and it gets exponentially more expensive to maintain that lie over time. Now, there's a bunch of really nice consequences to proof of stake is, one consequence is, it hardly consumes any power at all. Energy consumption by some estimates will drop more than 99.9%. So, suddenly, this massive energy footprint that Ethereum occupies will go away. Second is...

Mark Lurie:

And just to be clear, so basically, what this is doing is saying instead of one vote, it's one token, one vote. One state token, one vote in what truth will be and what transactions will go on the blockchain.

Matt Cutler:

So, to be precise, okay. On Ethereum, you're going to operate what's known as a validator. And to have a validator, your own validator, you have to stake 32 ETH, which is a lot of money.

Mark Lurie:

Let's say a thousand bucks. I mean, it's above a thousand, but it was a thousand. I mean, that's 30 grand. That's a lot.

Matt Cutler:

Exactly. So, by the way, the original specification was 1,500 ETH. So, it's a lot better than it was, right? There are, by the way, staking, just like there's mining pools, there's staking pools, there's rocket pool, there's Lido. There are others where you can contribute fractional share and they'll do all the work. They'll operate it on your behalf. And basically, what happens is, each validator controls 32 ETH of stake and there's a lottery process whereby a validator gets picked to be the head of the chain. It's called the proposer. So, how many validators are there on the network? How many blocks are being produced will determine how often your validator gets to be the head of the chain. And obviously, the more validators that you're operating, the more often you're going to be the head of the chain on a probabilistic basis.

Matt Cutler:

Then there are other roles where you're called attester. So, a validator says, "Here's the block I think should go on the network." And the attesters look at that and say, "Yes, this is thumbs up, thumbs down." And there's sort of various mechanisms in there to ensure that a block gets generated. It's a valid block. Everybody agrees, it's valid. And it goes on the chain and gets appended as it. So, it's not exactly one to one. It's more nuanced than that. But essentially, the more ether you have, the more validators you can operate, the more often you get to be part of this selection process and the more you get a reward for it.

Matt Cutler:

But there's a couple of really significant consequences to this. First off, the staked asset on the Ethereum network is ether. And you don't need anything else. You don't need any dollars. So, now, you don't need to sell your ether to get computers and bipower, right? And so, you have much less of this sort of externalities that affect the network. That's great. Two is, it requires virtually no compute to really do a very little compute. And so, any consumer can do this on any consumer grade laptop without bogging down their machine. And so, it facilitates decentralization, meaning more people can participate in securing the network.

Matt Cutler:

Three, it relies on game theory as opposed to cryptographic. So, there's some controversy over that. But for the most part, proof of stake seems to have been validated in the real world, but there's those who would agree that there's some significant differences there. But more than anything for Ethereum is it sets the stage, it sets the foundation for a whole bunch of future upgrades that will increase transaction throughput, improve block size and space, data availability, all of which translates to a more capable, higher scale network.

Matt Cutler:

Now, significantly people say, "Hey, at the merge, will Ethereum go faster?" No, not really. "Will Ethereum get cheaper?" No, not really. You go, "Why are we doing this, you say? Because it creates all the conditions necessary for it to get faster and cheaper.

Mark Lurie:

That's definitely a [inaudible 00:44:26].

Matt Cutler:

It's a major step. So, this is the merge, transition from proof of work algorithm, proof of stake algorithm. There's a whole bunch of really interesting nuances to exactly how this cutover occurs and why and what. But the current estimates are the most recent is the merger occurred during the week of September 19th. I think that's the date. And so, it may adjust one way or the other, but the assumption right now among the ecosystem is that the merge is quite close. So, that's the merge. Are you with me?

Mark Lurie:

Okay. Yeah. [inaudible 00:45:03]

Matt Cutler:

The merge is an upgrade to the network, but it contains many upgrades, not just the consensus algorithm upgrade. So, this is something that Ethereum has a pretty good track record on. This is one headline thing that's happening, but there's other stuff happening as well. And one of the big upgrades that's going to occur is the first step of what's called Proposer Builder Separation, PBS. And by the way, Flashbots is playing a significant role in all of this too. But effectively, what they're doing is they're taking this job of block construction away from mining pool operators and creating a separate marketplace for independent actors to build blocks and to propose them to the network. And this has never existed before. There's never been a separate role called the block builder.

Matt Cutler:

And this is, we think, very constructive. In part, because it separates out power. So, today, the miners have all the power. Tomorrow, the validators could have all the power. This sort of breaks that apart. It also makes the act of being a validator easier because block building can be computationally intensive. Sorry, the sun is coming into the sort of funny way. Let me see if I can move over here.

Mark Lurie:

Like proof of work, the sun is coming out.

Matt Cutler:

Yes. Exactly. Exactly. You get to see a different part of my background over here.

Mark Lurie:

Like the blockchain ecosystem.

Matt Cutler:

So, what this does is create this new class of actor in the network called the block builder. And it creates all sorts of new possibilities for how blocks get constructed. And so, this is, by some estimate, several hundred million and maybe billion-dollar new economic opportunity that's going to blink into existence immediately after the merge and is going to potentially really change the rules of the game for how power, and by power, I mean, money, value flows in the network.

Matt Cutler:

And so, this is a very interesting sort of additional component to the Ethereum merge that we think is quite constructive for the network. And we may see some very different dynamics start to happen as money starts to be able to move in new and different and more programmable ways that could benefit end users and traders significantly.

Mark Lurie:

Okay. So as a trader, are there any tricks that I should be keeping an eye out for in particular, which you think will come up?

Matt Cutler:

So, okay. So, one of the consequences of the move to proof of stake is because you don't have to worry about burning a whole lot of power. You don't have to worry about having specialized computing. The rewards for validating a block go way down. So, the rate of issuance of new ether being created goes way down. But there'll be folks who still want to make money out of all of this. And so, the prevailing wisdom is MEV will make up for a lot of those differences. And so, the network will become a lot more efficient at extracting MEV. So, if you think MEV is a fact of life today, it's going to become a bigger fact of life, a more ever present fact of life, post merge, to help make up for some of the gaps in network issuance.

Matt Cutler:

Two, because of block building, and because you're going to have specialized actors, there will be new marketplaces for this to happen. And we believe potentially, new ways that this value could float, as in wallets will have the opportunity to alert users when something they're doing has MEV associated with it and to invite them to participate in that. And in particular, the wallet could pay you back. So, imagine using a wallet that when you're trading says, "Oh, you're good. No MEV here." Or "There's MEV, but it's a little bit." Or, "Hey, there might be a thousand dollars ARB that emerges from your trade, would you like half of that?" Whoa, I didn't even know this was happening.

Matt Cutler:

And now, for basically doing, for just opting in, I get $500 back potentially. And the consequence by the way is you might have to wait a little bit longer for your transaction to confirm. There's more work that needs to be done. And by the way, imagine you, Mark use your meta mask wallet to do a Uniswap transaction on Zapper. Now, the MEV that is a consequence of that only exists because you did that transaction. So, we would consider you a transaction originate.

Mark Lurie:

I only use Clipper for the record, but Uniswap [inaudible 00:49:52].

Matt Cutler:

Oh, right. Yeah, of course, of course, Clipper, sorry. You use Clipper, right? Who is the transaction originator on this? Well, there's the user, there's the wallet, there's the protocol and there's the dApp. And all of them played a role in that transaction happening. So, all of them played a part in the MEV emerging and all of them we believe should be compensated. Appropriate, right? And so, this new world that we think is dawning is the power has the opportunity to get redistributed through the network. And instead of consolidating all the way at the miners like it does today, the future version is the validators that it may get more equitably distributed through the ecosystem. And that's really quite constructive.

Mark Lurie:

That's interesting because yeah, quite constructive. It also creates a business model for a lot of these intermediate layers, right? That today may not be able to have a business model because they have no way to charge or get income. But now, you you get a little slice of the transaction pie or the MEV pie because you participated in it. And all of a sudden, you can get a lot more interesting models. And that's a lot more interesting innovation because money fuels innovation elsewhere in the ecosystem. Very exciting.

Matt Cutler:

Exactly. And this, of course, will require additional infrastructure. So, people will need, and by the way, this is the type of work that Blockdata does. And so, as we think about this, this is we think there'll be a new class of infrastructure providers that emerge, that have many of the skills and capabilities that we have today to help people participate in this. Because quite frankly, there's a lot going on. There's a lot of connective tissue that's required and most operators are not going to have the skill set to build on their own and they're going to want to maintain it over time. It's going to be quite expensive.

Matt Cutler:

And so, you're going to have infrastructure providers which are set up exclusively to do this. And so, for most of the participants in the network, it will be relatively little work to get access to a whole new revenue stream. And we again think that this is highly constructive for the network. Meaning, more users will come in. They'll conduct more transactions of greater value because there's more transparency and there's greater sort of rewards flowing through it overall.

Matt Cutler:

And so, the end state is not something dystopic, where users are getting abused by these faceless robots and the Mempool that are preying on their transactions. Instead, there's infrastructure, which has been constructed, which ensures everybody's best interests are looked out for.

Mark Lurie:

Got it. My mind is turning. I'm going to have to rethink a bunch of stuff with how Clipper and the protocol Shipyard develops, how their business models work. But I know the guy come to now when this comes up. You're the guy.

Matt Cutler:

Please do so. I'm @mcutler on Twitter, M-C-U-T-L-E-R. And you can find Blocknative at @blocknative on Twitter. DMs are open. You can always reach out. We have a pretty vibrant Discord community and we think there's a whole lot that's super interesting here. And to your point though, you can either ignore this or you can be proactive about it. And our sense is those who are proactive will profit. And those who are passive will suffer, as a result, will basically lose market share and they'll have to get proactive. So, our sense is that there's not going to be a lot of discretionary that everybody's going to need to formulate how they're going to be participating in this ecosystem and get proactive.

Matt Cutler:

And let's be honest, those who are first movers here have probably the most to gain, but there's a lot to figure out. It's going to take some time to sort this all out. So, we're entering into what I like to think of as a new Genesis block. There's a whole new network that's coming, whole new games that can be played, whole new flows of economic value. And we're just excited to help facilitate that and allow folks to explore the possible here. So, we think there's a whole bunch of really cool stuff that's coming and it's an interesting time to be building on in these networks.

Mark Lurie:

Super interesting. And the first step to being proactive is becoming informed. So, thank you so much for joining us today and informing us. Matt, we really appreciate it. Thank you.

Matt Cutler:

My pleasure. And again, tons of resources out there to look into, tools you can use for free on our website at Blocknative. And I really invite everybody to check it out and looking forward to hearing feedback and to getting questions on Twitter and elsewhere, okay?

Mark Lurie:

Awesome. Thanks, Matt.

Matt Cutler:

Thanks, Mark.

Listen to WTF, Crypto on your favourite podcast platform.

Apple Podcasts logoSpotify logo
See more platforms