30 May 2024
What’s MEV?
For the uninitiated, Maximal Extractable Value (MEV) involves various techniques used by market actors to capture additional value by exploiting price inefficiencies in blockchain transactions.
One naive form of MEV is called “transaction sniping.”
Most recently, manifestations of this phenomenon have begun populating your local Bitcoin memepools because of ordinals trading. Without going into too much detail, the way on-chain trading currently works is through the use of pre-signed Bitcoin transactions (PSBTs).
The concept is simple: some users list pictures on a marketplace by crafting a transaction with the details and the price they would like to sell it at. You, a cat aficionado, can purchase this feline grail by completing the transaction, adding to your address, the transaction fees, and your signature. The transaction is then broadcast and eventually gets settled on the Bitcoin network.
Easy, right?
Not so.
Turns out cats are a popular commodity these days and other fellow cat enjoyoooors are eyeing your prize. The open nature of the offers allows any of them to interfere with your purchase. This is because PSBT listings are auctions, not exclusive sales. Every transaction in the mempool associated with a cat is up for bid. Bitcoin’s 10-minute block interval opens up a window for cat opportunists to “snipe” each other’s transactions for the most valuable pieces. Nothing is settled until a transaction makes it into a block.
We learn from this that on-chain auctions are vulnerable to the settlement time of the blockchain they settle on. This creates an especially thorny problem for anyone with a little more ambition than trading cat pictures.
What’s causing MEV?
Now what’s the big deal with MEV? Surely everyone is not up in arms about a handful of degens outbidding each other for cat pictures?
Answering this question requires opening up an entirely new can of worms. This is going to be a bit of a trip in shitcoin land but bear with me, I promise it’s worth it.
You see, MEV is big business. To give a rough idea, MEV-Boost, the software responsible for coordinating the extraction of MEV on Ethereum has distributed, in less than 2 years, close to 500,000 ETH in rewards. That’s nearly 2 billion dollars!
What drives this entire market is, of course, trading. (read: degens)
On-chain AMMs (Automated Market Makers) use a popular concept called liquidity pools to allow users to trade assets without relying on centralized order books. Pools are typically composed of a pair of assets, like Ethereum and USDC.
When users want to trade one asset for another, they interact with these liquidity pools. Each trade adjusts the ratio of the two assets in the pool, affecting their relative prices. By removing centralized order books where a buyer needs to be matched with a seller for every trade, AMMs can be deployed as a decentralized on-chain contract.
Once a trading pair is created, any user can trade against the pool or contribute liquidity to it. Contributing liquidity involves supplying one or both assets to the pool, ensuring it maintains the appropriate ratio. Any imbalance creates an opportunity for market makers to engage in arbitrage by buying the cheaper asset from the pool and selling it at a higher price in other venues, such as centralized exchanges. This activity, along with the fees accrued by liquidity providers (LPs) from trading activity, incentivizes people to keep these markets liquid.
If you’ve heard of DeFi or Uniswap but never looked into them, this is the secret sauce. Liquidity pools can be deployed to trade any asset, and their permissionless nature has made them wildly popular. Upon closer inspection, we can observe that these trades are not fundamentally different from the cat market we discussed earlier — they’re just on-chain auctions.
As you might imagine, Ethereum’s architecture and its additional programmability create a particularly fertile environment for those auctions to be abused and manipulated.
Perhaps the most popular and intuitive one is front running. Remember that trades on AMMs are not instantly settled. In the same way that cat snipers can monitor the Bitcoin mempool for juicy deals, Ethereum also has an army of financial mercenaries diligently watching over every trade opportunity.
Except that the ETH boys aren’t playing with this stuff. It’s no Private Ryan, it’s SEAL Team Six over there. They employ several elaborate techniques to exploit the time gap between when a transaction is broadcast and when it is confirmed on the blockchain. As a result, front-runners can place their own transactions ahead of the original trade, profiting from the price changes their actions cause. This generally causes regular users to receive worse prices than expected. One of the worst manifestations of this is the practice of sandwiching users, where a buy order is placed just before a user’s trade and a sell order right after, capturing the price difference at the expense of the original trader.
While those dynamics have been controversial because of the effect they have on user experience, they only represent a portion of the MEV economy. The biggest source of MEV, by a rather large margin, comes from something commonly known as “Loss-Versus-Rebalancing.” Put simply, it is an adverse form of the pool arbitrage described above affecting liquidity providers.
When the price of the assets in the liquidity pool diverges from the price at which they were originally deposited, arbitrage traders step in to rebalance the pool to reflect the global market prices. This rebalancing process leaves liquidity providers vulnerable, as they’re unable to adjust to market volatility between blocks. Due to their exposure to outdated prices, they become an easy target for traders with access to centralized order books. These traders exploit the price discrepancies, often leaving liquidity providers with a less favorable asset mix and reduced overall value.
The situation is so dire that those were the conclusions of a recent research paper on the matter:
Our main result is that earnings from fees are smaller than losses to arbitrageurs in the majority of the largest Uniswap pools, currently holding hundreds of millions of USD. This result raises the question of why LPs nevertheless contribute their capital to these pools.
Other factors such as hedging have helped larger operations mitigate those issues but the extreme conditions described have likely resulted in the consolidation of liquidity provisioning into a smaller number of actors.
Why do we care about Ethereum?
Good question, anon! The reason I bring this up is because a lot of the recent conversation around MEV on Bitcoin completely misses the fact that those systems don’t exist in a technical vacuum. I understand that the new and unknown concepts can provoke skepticism but a lot of the dynamics involved are well understood by now. Strictly looking at it under a technical lens does us all a disservice.
We understand that different new proposals floating around to improve Bitcoin’s scripting capabilities may introduce more expressivity to the protocol. It’s not impossible that a combination of those features may allow someone to build the equivalent of an AMM on-chain. It’s clear something on the scale of Ethereum would have adverse effects on the decentralization of Bitcoin. We know that MEV tends to incentivize high levels of specialization at the mining level. If you want a better understanding of the risks at hand, Spiral developer Matt Corallo put together a decent primer on the topic.
Unfortunately, the most important aspect of this subject has remained more or less completely ignored by everyone at the table. MEV and every associated system are driven by economic incentives. Different parameters can have a significant impact on the viability of this activity.
Our story about cats illustrates how the interval between blocks plays a crucial role in the game theory of on-chain auctions. This theory is now supported by documented evidence. Researchers generally agree that longer block times exacerbate issues around MEV. This poses a significant challenge for anyone considering building AMM systems on the Bitcoin blockchain.
Is Bitcoin in danger?
Comparing Bitcoin’s 10-minute block interval to Ethereum’s 12 seconds, it’s fair to wonder whether or not the settlement times required by Proof-of-Work security are compatible at all large-scale on-chain auctions.
The prolonged interval between Bitcoin blocks means that liquidity providers (LPs) would be exposed to stale prices for extended periods, making it impractical, if not irresponsible, to commit substantial capital. This latency increases the risk of front-running and other forms of MEV exploitation. It’s an arbitrage dream!
Those observations suggest that on-chain trading on Bitcoin may not be viable even if it becomes technically possible. Applications targeting this use case are increasingly optimizing for speed and efficiency, leaving little room for Bitcoin to become a competitive option. Capital allocators are likely to shy away from the risk associated with this architecture, and users will simply prefer platforms better aligned with their interests.
This highlights the critical importance of economic considerations when evaluating the risks and rewards of technical changes to the Bitcoin protocol. Don’t get it twisted, the noise surrounding this conversation is driven by economic interests eager to replicate the financial flywheel of MEV on top of Bitcoin. Now that the attitude with regard to innovation at the protocol level has changed, they look at this situation as a billion-dollar opportunity to recycle proven business models.
What many fail to appreciate is that Bitcoin’s slow but steady settlement process acts as a natural deterrent to predatory MEV activities. This isn’t a comprehensive study and further evaluation is necessary to assess MEV risks on layers built atop the protocol. On the other hand, it’s a pretty compelling reason to believe that the fear of the MEV monster on Bitcoin may be grossly overestimated. The inherent delay in Bitcoin’s transaction finality offers a unique form of protection, making it less susceptible to the same level of MEV exploitation seen on faster chains like Ethereum.