MEV and Its Impact on the Future of DeFi

In the Ethereum network, in both proof-of-work and proof-of-stake consensus mechanisms, pending transactions will wait in the Mempool (publicly visible waiting area of the network) until a miner or validator selects them, orders them, and creates a block out of the transactions. That block is then validated by nodes in the network and added to the blockchain. However, when pending transactions sit in a Mempool, miners and validators have found ways to profit from them by including, excluding, or reordering transactions at block creation.

However, MEV is not an Ethereum-specific issue, although it is a common topic within Ethereum Virtual Machine (EVM) compatible systems. Smart contracts are now the primary opportunity for MEV on Ethereum and similar blockchains. MEV strategies are less lucrative on Bitcoin than Ethereum because of its lack of smart contracts.^1,2 The smart contracts allow complex transactions within the EVM-compatible systems, including lending, borrowing, and trading. This complexity enables miners to create more MEV opportunities. Moreover, the accounts-based model for transactions used in the EVM-compatible systems is more MEV-friendly than the Unspent Transaction Output (UTXO) model used in Bitcoin.³ The accounts-based model encourages decentralised applications and smart contracts, while the UTXO model focuses strictly on facilitating peer-to-peer payments.

In theory, miners or validators should get the full MEV amount because they are the only party who can guarantee MEV execution. However, Searchers (independent network participants) extract a large portion of MEV. Searchers run complex algorithms to detect profitable MEV opportunities, use bots to automate the process, and pay high gas fees (which go to the miner/validator) to include their profitable transactions in a block. The Flashbots project is one of the leading projects in this space where searchers perform MEV extraction. The MEV-Boost is another project from Flashbots that enables validators to sell block space to an open market of block builders.⁴ The block builders collect, sequence transactions, build a block with the maximum MEV, and submit it to validators for validation and inclusion. For example, the data from mevboost.org shows that over 80% of Ethereum relay blocks were built by Flashbots.⁴ However, this approach from Flashbots is imperfect because Flashbots are under centralised control and not censorship resistant.

MEV should be addressed within the Decentralised Finance (DeFi) domain. MEV has both positive and negative impacts. While arbitrage and liquidation settlements can result in users getting the fairest prices across exchanges, some methods - including front-running and sandwiching transactions - negatively impact the DeFi ecosystem. MEV is now a $1 billion industry, and most of MEV extraction comes from searchers.⁴ Moreover, based on the MEV explorer statistics, the searchers mainly seek arbitrage and liquidation opportunities based on the different prices of tokens at various exchanges. However, the centralized nature of searchers goes against the inherent decentralized philosophy of DeFi.

The financial industry is transforming its financial applications into DeFi applications - stablecoins, digital assets, and tokenization platforms are emerging on public blockchains seeking regulatory compliance. Moreover, companies need to protect their customers from external MEV extraction. Therefore, it is vital to manage MEV within the Ethereum ecosystems; otherwise, MEV could negatively impact customer transactions and regulatory compliance.

The Most Common MEV Strategies

MEV comes from both the execution layer and the application layer. For example, block producers can generate MEV at the execution layer. Trading mechanisms and on-chain vs off- chain latency arbitrage are examples of application layer MEV. Some commonly used MEV extraction methods found so far within the EVM ecosystems include:

• Arbitrage – When two decentralised exchanges (DEXs) offer a token at two different prices, someone can buy the token from the lower-priced DEX and sell it on the higher- priced DEX. This transaction creates an arbitrage opportunity. However, it is essential to note that this price difference may only last for a brief period of time. Therefore, bots must ensure faster transaction execution to earn a profit.
• Liquidation – Searchers perform MEV extraction by liquidating over-collateralized loans from on-chain protocols like Compound, Maker, and AAVE. Like arbitrage, liquidation is a net positive activity to the DeFi ecosystem in which buyers buy locked collaterals automatically issued by the lending protocols and quickly resell these collaterals at market value - this transaction ensures that liquidity continues to move between applications and prices of assets normalize rapidly across the ecosystem. For example, Searcher 1 finds a loan of 1000 DAI liquidation from Lender 1, purchases the liquidated asset from the auction by Lender 1 at a 10% discount, sells at market value at DEX1, and profits 100 DAI.³
• Sandwich trading – In sandwich trading, a searcher will watch the Mempool for a large, decentralized exchange (DEX) trade because a large trade significantly increases the price of the token being bought. A searcher calculates the price effect of a large trade, executes a buy order immediately before the large trade, executes a sell order after the large trade, and collects the profit.⁵ However, sandwich trading is risky because transactions are not atomic.
• Front-Running – Front-Running is also called Priority Gas Auction (PGA). Front-running enables selecting a transaction ahead of a pending transaction from the Mempool by allocating a higher gas price. For example, searchers compete against each other by bidding on transaction fees to obtain priority ordering for their transactions. However, the negative side is these drive-up fees for other users. Block producers (miners) can apply front-running to win arbitrage and liquidation opportunities.

MEV, DeFi, and Regulations

As explained in the section on MEV strategies, DeFi transactions are likely to be impacted by MEV. The Bank for International Settlements in Europe has already warned about front- running and sandwich trades for MEV, which are illegal in the traditional markets. They further emphasised that MEV is an intrinsic shortcoming of pseudo-anonymous blockchains. Addressing these market manipulations may call for new regulatory approaches to this new class of intermediaries.⁶ Therefore, finding technical solutions to control or mitigate illegal MEV extractions is essential.

The Office of Foreign Asset Control (OFAC) of the U.S. Treasury Department sanctioned Tornado Cash on 8th August 2022 and added it to its specially designated nationals list. The list includes thousands of people, companies, and organisations identified as posing a threat to U.S. national security.⁷ It is illegal for U.S. persons and entities to interact with members of this list. This news made shockwaves across the crypto industry, as major DeFi protocols, including AAVE, Uniswap, dYdX, and Balancer, started blocking the sanctioned addresses. Centre consortium blacklisted 38 wallets holding 75,000 USDC in response. Flashbots confirmed they would comply with the sanctions on 17th, August 2022.⁸ According to MEV Watch, 33% of Ethereum post-merge blocks are now OFAC compliant (as of March 2023).⁹

How to Address MEV in Longer Term?

The centralization of block content creation is the main reason for MEV extraction. Vitalik Buterin aims to address MEV and censorship from his new Ethereum road map using In- protocol Proposer Builder Separation (PBS) and application layer MEV minimization.¹⁰ In PBS, Block builders bundle up transactions into blocks and submit them to the proposers for selection. After the proposer selects a block successfully, the block builder will process the block. Block proposers are the miners before the merge and validators after the merge.

The road map starts from the sidecar architecture using MEV-Boost until finding the in- protocol PBS (PBS is a part of protocol-native in Ethereum).¹¹ PBS architecture suggests separating the functions of block-building and block-proposing on the Ethereum core protocol. The philosophy behind this concept is isolating MEV activity and minimising centralising forces on the Ethereum base later. MEV-Boost and Searchers approach from Flashbots allows validators to engage in an experimental version of PBS, as illustrated in Figure 1.^{4, 11}

In the initial implementation of PBS(MEV-Boost), the design envisions each party trusting the parties to whom they connect.

• Searchers should trust the builders to whom they send bundles – when builders see the bundles that searchers submit to them, they can copy the transactions and replace the searchers’ addresses with their own.
• Builders should trust the relays to whom they send blocks. Relays won’t leak transaction data or use it to extract additional MEV, and relays should respond to the validator’s requests quickly so that the validator does not miss the chance to propose a builder’s block.
• Validators should trust the relay(s) they receive blocks from. If the relay does not respond in time, a validator may miss their window to include a block in the next slot and will not earn any rewards for their participation. These trust dependencies are significant challenges in this approach.

If PBS is built natively into the Ethereum protocol, relays will be eliminated, and these trust assumptions will be reduced.¹² However, it still needs to be determined how to integrate PBS into the Ethereum protocol and eliminate trust dependencies. PBS sidecar architecture implementation is still in the experimentation stage.

Experimenting with auction mechanisms to minimize centralization concerns is discussed in several projects.

• The SUAVE project (Single Unifying Auction for Value Expression) from Flashbots is one of the efforts to build decentralized builders by decentralizing and democratizing MEV-Boost.¹²
• Frequent Batch Auction (FBA) to reduce MEV on the application layer is another direction of MEV management proposed by the Coincidence of Wants (COWs) protocol.¹³ The COWs protocol uses parties called solvers to provide settlement solutions to batch auctions. Solvers compete against each other to submit the most optimal batch settlement solution, and the winner will get rewards from the protocol.
• Partial block auctions are proposed in, aiming to minimize builder centralization concerns.¹⁴

Moreover, several improvements to PBS are discussed in references 15-17. However, it is yet to explore a promising solution to deal with fair MEV extraction and the lack of customer-focused best execution requirements or mechanisms

Conclusion

Within the current EVM ecosystem, miners (validators) control block content building, a centralised process that leads to the MEV problem. The MEV problem directly impacts customers, and it has drawn serious attention from regulators. PBS, PBS improvements, the SUAVE project, and the COWs protocol may provide potential solutions in this domain. The future of the Ethereum ecosystem and DeFi clearly needs resolutions to the MEV problem.