How I understand "MEV"?

  • Modified: 
  • Published: 
  • MEV DeFi

MEV exists because blockspace is scarce and valuable. This might sound abstract, but it's actually quite intuitive once you think about it the right way. Here, in a simplified version, is my mental model for thinking about MEV.

MEV as economic rent for blockspace

In my work with Alex Sarch and Natasha Vasan, we've argued that one helpful way to understand MEV is through an analogy to ground rent.[1] Just as a landowner can charge rent because they control a scarce piece of land that others need to use, validators (or miners, block builders, sequencers) on blockchains control scarce "blockspace" that users need to execute their transactions.

Here's the key insight: because validators control their piece of blockspace (a block), and because users have rivalrous, valuable uses for that scarce space, validators can charge users proportionately to the value they attach to what can be done in the space.[2]

Why blockspace creates MEV opportunities

The crucial difference between blockchain systems and traditional finance is that blockchains don't use a first-come, first-served model for transaction ordering.[3] In traditional continuous limit order books, transactions are processed in the order they arrive. But public blockchains like Ethereum deliberately chose a different approach.

Why? Because blockchains are meant to operate globally in a decentralized fashion without privileging particular geographic locations. A time-based system would advantage traders physically closer to key infrastructure—one of the things that such blockchains are designed to avoid.[4]

Instead, validators have discretion over which transactions to include in their blocks and in what order to execute them. This discretion, combined with the scarcity of blockspace, creates the conditions for MEV as we know it.

Taking away this discretion from validators would affect the kind of MEV we see today, but it would not address the issue of competition over the uses of blockspace. Hence, we would likely see competition move upstream—for example, taking the form of bringing trading infrastructure as close as possible to the computers that run validator software (collocation in the same datacenters and so on). It's not clear that the outcome would be overall better.

Competition makes MEV systematic

What creates, or at least increases, MEV is competition. Take arbitrage as an example: when prices differ across decentralized exchanges, multiple sophisticated actors will spot the opportunity simultaneously. Only those who can get their transactions executed first will capture the profit.[5]

In a competitive environment, the rent that validators can extract tends toward the total value of opportunities available in the blockspace they control.[6] It's similar to how ground rent in a competitive market tends toward the maximum productive value of the land.

This competition happens at multiple levels:

  • searchers compete to find and execute profitable opportunities;
  • block builders compete to construct the most valuable blocks;
  • validators effectively auction their blockspace to the highest bidder.

Discretionary transaction ordering is not inherently unfair

It's tempting to think that discretionary transaction ordering is somehow "unfair" compared to first-come, first-served. But as we argued in our paper, there is no natural or pre-existing transaction order that MEV extractors are disrupting.[7] Before transactions are included in a block, they exist in an unordered state ("the mempool").[8]

When a validator orders transactions in a way that creates a sandwich trade or enables arbitrage, they haven't "reordered" anything—they've simply ordered transactions according to economic incentives. This is fundamentally different from queue-jumping in traditional finance, where there actually is a temporal queue.[9]

Beyond Ethereum

While much MEV discussion focuses on Ethereum, these dynamics exist wherever you have:

  1. scarce blockspace
  2. valuable uses for that space
  3. discretion over transaction ordering
  4. competition among users

This means MEV exists not just on Ethereum, but on Solana, Polygon, and other blockchains where DeFi activity occurs. The specific mechanisms might differ—Solana's continuous block production creates different dynamics than Ethereum's discrete slots—but the fundamental economic logic remains the same.

Implications

Understanding MEV as rent for scarce blockspace helps clarify several important points:

First, MEV isn't inherently good or bad—it's a natural consequence of blockchain architecture. What matters is how that rent is extracted and distributed.

Second, attempts to "eliminate" MEV entirely are likely futile. As long as blockspace is scarce and valuable, those who control it (or control the relevant network connections) will be able to extract rents. The question is how to align these incentives with the health of the overall system.

Third, different MEV extraction strategies have different implications. Some, like arbitrage, may improve market efficiency by aligning prices across venues. Others may involve breaches of trust or manipulation that harm users and markets.

The key is to evaluate specific MEV practices in their specific contexts, rather than treating MEV as a monolithic phenomenon to be eliminated or embraced wholesale.

More nuance: state vs blockspace contention (or contention vs congestion)

Congestion vs. contention—­or block-space contention versus state contention[10]—frames two different ways to define MEV.

The block-space (congestion) view treats MEV as the value a proposer (sequencer) or external block builder earns because it controls which transactions squeeze into a finite-size block. Generally speaking this is what some call "in-protocol" revenue. This is easier to measure with basic on-chain data. Its weakness is that it doesn't capture MEV that is not reflected in congestion (see below). It can give a “just raise the gas limit” illusion about reducing MEV—scaling may lower congestion fees while state-contention remains.

The state-contention view instead defines MEV as value unlocked when two or more actors race to touch the same storage slot—an NFT mint counter, a Uniswap pool, a liquidation slot—so it matches how searchers actually bid around specific conflicts. This is "out-of-protocol" revenue. However, it is harder to measure, risks expanding until “all economic activity is MEV,” and understates proposer/sequencer-level powers such as re-orgs.

In other words, block-space congestion happens when too many users want transactions included in one block, irrespective of the transaction ordering in that block. For example, multiple users bidding high gas fees to enter a crowded NFT mint. State contention occurs, for example, when two traders race to liquidate the same lending position, each raising fees to be executed first, even if the block itself is not full.

Blockworks’ REV (Real Economic Value) metric[11] covers both. It sums base fees (congestion), blob fees (data-bandwidth congestion), priority fees (a mix of congestion and contention) and additional builder bids (like MEV-Boost tips; likely to reflect contention). So REV is meant to capture both "in-protocol" and "out-of-protocol" revenue.

Even more nuance

What makes things even more complicated is that the way congestion and contention are usually defined and measured doesn't take into account that searchers (traders), block builders, and validators can strike side-channel deals. In those deals, the the real payment for transaction ordering may be routed off-chain (e.g., a wire transfer, a crypto transfer on a centralized exchange) or at least through a more complex revenue-sharing agreement with on-chain payments which are not easily identifiable as such. In other words, real payments don't appear on-chain as a tip or builder bid. Hence, any lens that focuses only on observable "standard" measures of validator/block-builder revenue understates MEV gains.

  1. Mikołaj Barczentewicz, Alex Sarch and Natasha Vasan, 'Blockchain Transaction Ordering as Market Manipulation' (2023) 20 Ohio State Technology Law Journal 1. ↩︎

  2. ibid. ↩︎

  3. ibid 7. ↩︎

  4. ibid. ↩︎

  5. ibid 17. ↩︎

  6. ibid 55. ↩︎

  7. ibid 82. ↩︎

  8. The mempool (or "transaction pool") is where pending transactions wait before being included in a block. Each node maintains its own view of pending transactions. For technical details, see ibid 14 n 39. ↩︎

  9. ibid 57 n 228. ↩︎

  10. For the distinction between contention and congestion, see eg Justin Drake, 'MEV Burn - a simple design' (ethresear.ch, May 2023). ↩︎

  11. Blockworks, 'Ethereum: Network REV (Real Economic Value)'. ↩︎