This article is compiled and organized by BlockWeeks
MEV, namely "Miner Extractable Value" or "Maximal Extractable Value" (Miner/Maximal Extractable Value), refers to the value that miners or validators obtain by leveraging their power to order transactions within a block. With the explosion of DeFi applications, MEV has become increasingly profitable on Ethereum; and the emergence of the Flashbots Auction, a market specifically matching the most profitable MEV opportunities, has made it easier to access.
Data can illustrate the scale of this dark river: in 2021, miners and operators profited $730 million from Ethereum MEV; from January 1, 2020 to December 31, 2021, the cumulative MEV extracted on Ethereum exceeded $773 million. At the current rate, Ethereum miners alone are expected to earn over $750 million in additional profits from MEV each year, most of which comes from pure arbitrage.
The "Frontrunning" Problem in the New Financial Era
The new finance is being built on Ethereum. Despite various differences between DeFi and traditional markets, both industries are stuck on the same old problem: frontrunning. In traditional finance, frontrunning usually refers to trading based on material buy or sell information that has not yet been made public. On paper it is clear, but in reality there is great controversy over how widespread frontrunning actually is. IEX founder Brad Katsuyama and Michael Lewis, author of the book Flashboys, are widely credited with bringing the operations of high-frequency traders to public attention—traders who execute buys and sells based on information obtained a few milliseconds earlier than others in the market.
On Ethereum, similar arbitrage, stop-loss hunting, and high-frequency dominant trading are taking root in DeFi markets. As DeFi trading volume grows, the value of "executing faster than others" also rises. This new type of high-frequency trading allows technically skilled participants to arbitrage between decentralized exchanges (DEXs), or to immediately force liquidations and suppress positions after sharp asset price fluctuations on trading and lending platforms, thereby obtaining substantial profits. This is MEV.
The motivations of MEV are no different from the opportunities in traditional finance; the root cause lies in certain participants having privileges over submitting and reordering transactions. But the consequences differ: such opportunities in traditional finance often raise the threshold for market participation, whereas MEV on Ethereum may instead bring higher participation, transparency, and efficiency. At best, MEV makes DeFi markets more efficient by providing economic incentives to correct price deviations; at worst, MEV disrupts network consensus, damages users' trust in the Ethereum protocol, and exposes user transactions to unexpected slippage or attacks.
The Three Main Types of MEV
The most common types of MEV on Ethereum are three:
Arbitrage
Profiting from price differences between different DEXs or trading pairs is the largest source of MEV profits and also the most "benign" type—it objectively promotes price discovery.
Liquidations
In lending protocols, when collateral prices fall sharply and positions approach the liquidation line, submitting liquidation transactions before others to earn liquidation rewards.
Sandwiching
Inserting one's own transaction before and after a user's large transaction, artificially pushing prices up or down, and profiting from the user's slippage. This is the type of MEV that most directly harms ordinary users.
It should be emphasized that MEV is not limited to these three types. As searchers continue to explore the frontiers of DeFi innovation, new types of MEV are being continuously created.
From Miner Income to Validator Income
Technically, MEV mainly comes from miners' ability to reorder transactions within a block, and it constitutes a third category of miner income beyond block subsidies (new issuance) and transaction fees. Once Ethereum abandons proof-of-work (PoW) and shifts to proof-of-stake (PoS)—a transition then expected to occur in 2022—the ability to reorder transactions will shift from miners to validators. In other words, MEV will not disappear with the consensus mechanism switch; it will only change who collects it.
The scale of MEV is highly synchronized with DeFi activity. Since May 2021, MEV has declined significantly, due to the release of Uniswap V3 and the migration of DEX activity from Uniswap V2 to V3. It is worth noting that the data provider Flashbots, although tracking MEV activity on Uniswap V2 and seven other DeFi protocols, has not yet included Uniswap V3, which ranks first by 24-hour trading volume. This means that public MEV statistics are likely systematically underestimated.
"Extreme Rent-Seeking" or a Necessary Evil?
MEV is often criticized as the ultimate manifestation of rent-seeking behavior, but upon deeper analysis, the issue is not so black and white. In addition to some MEV being beneficial to market price discovery, MEV is also an inevitable byproduct of the mechanisms that safeguard Ethereum's operation: miners' discretionary power over transaction ordering is itself an important defense against network spam transactions.
But this important power (transaction ordering) has spawned an industry extremely similar to high-frequency traders in traditional finance. Both MEV and HFT rely on executing transactions "in a specific order, usually ahead of others' transactions" to capture profits; both may negatively affect market participants directly or through externalities, and erode traders' trust in the market, prompting people to turn to private channels to complete transactions.
Unlike traditional markets, there is no centralized regulatory agency on Ethereum to supervise and enforce rules regarding MEV. As a decentralized, permissionless system, the only "law" governing MEV is the rules written into the consensus mechanism. This places greater responsibility on protocol developers, dapp users, and the broader Ethereum community—they need to drive code changes and establish norms for which MEV is tolerable and which is not.
A Three-Pronged Approach: Three Paths to Govern MEV
There are roughly three paths to counter the negative externalities of MEV:
- Consensus protocol layer updates: Change the way transactions are ordered and blocks are built at the protocol level;
- Decentralized application design changes: Compress the attack surface by improving dapp matching and slippage mechanisms;
- Education: Help users understand the MEV strategies being executed on-chain and their costs.
However, every effort comes with trade-offs. The most typical example is the birth of the Flashbots Auction: it created unprecedented transparency in the types and scale of on-chain MEV, but also significantly lowered the threshold for miners to rely on MEV for additional profits. In addition, by moving most of the bidding war among searchers competing for MEV profits off-chain, Flashbots helped alleviate congestion in Ethereum's mempool, but at the same time created a centralized "gatekeeper"—most Ethereum miners now rely on it to obtain MEV revenue.
Uniswap V3 and JIT Liquidity Attacks
Beyond sandwich attacks, another noteworthy form of MEV is the Just-in-Time (JIT) liquidity attack. This type of MEV can only be executed on Uniswap V3, the third generation of Ethereum's most popular DEX.
JIT attacks exploit Uniswap V3's concentrated liquidity design—liquidity providers (LPs) can allocate assets within custom price ranges rather than spreading them evenly across the entire price range. LPs can thus concentrate funds near the most actively traded mid-price, creating deeper local depth and earning more fees. Conversely, this also makes traditional sandwich attacks harder to succeed on V3 than on V2: the deeper liquidity supporting a specific price range makes it difficult for searchers to artificially push prices with a single large transaction.
Faced with this reality, MEV searchers on V3 instead rebalance their portfolios into more profitable structures by providing and quickly withdrawing liquidity. For example, suppose a user wants to exchange 1,000 ETH for 4.5 million USDC. A searcher executing a JIT liquidity attack would first provide liquidity for this order and complete the match. As a "temporary LP," the searcher earns the transaction fee from this trade.
More critically, by withdrawing liquidity immediately after the user's transaction is executed, the searcher simultaneously takes away a rebalanced portfolio—containing both USDC and ETH. Essentially, the searcher provides liquidity for the user's large transaction with USDC, thereby obtaining ETH. It not only takes the fees that should have belonged to Uniswap's passive LPs, but also saves the trading fees that the DEX usually charges when swapping the two assets. After withdrawing liquidity, the searcher can then perform further operations on their new portfolio.
Conclusion: MEV and DeFi Must Coexist
MEV is an "innovation" that exploits Ethereum miners' (soon to be validators') discretion over the ordering of transactions within a block. This power exists for an important purpose—to resist spam transactions and protect the network. But it is precisely this power that has given rise to an industry highly similar to high-frequency traders in traditional finance, and has brought efficiency improvements and depth enhancements to the market.
The relevant upgrades to Flashbots are expected to improve the aforementioned trade-offs, making Flashbots Auction a system with fewer trust assumptions; but no single upgrade can create a network entirely free of transaction ordering manipulation. In a future where MEV and DeFi must prosper together, the truly unresolved question is: in the long run, can people's trust in and the resilience of the Ethereum dapp ecosystem remain intact.
