Order in ‘the dark forest’: private resolution of a market failure

Transcript

All right. My talk is on ordering the dark forest, which is of course MEV, but before that, let's take a step back. 45 billion, that's the current TVL on Ethereum mainnet held in permissionless contracts. Data valid as of last night on DeFi Llama. TVL being of course our industry standard industry jargon for investor deposits in smart contracts. And when I mentioned this figure to people outside the industry, they often ask but is they they ask is it even regulated? How do people trust it? Why is there so much money?

Because the assumption in scholarship, but also in the wider public is that markets thrive only with state-backed legal guarantees, right? So the obvious question is how order is maintained when the usual institutions like financial regulation and courts with clear jurisdiction are pretty much absent. And broadly, there are two accounts that try to answer this. The first one is what I call techno-determinist. It's the idealized view of DeFi as governed entirely by immutable self-executing smart contracts where whatever execution permits is legitimate.

The sentiment that captured is that captures this is famously Code is Law. And you'll probably remember that defense that the execution was valid within the protocol logic and therefore it was legitimate. It was used by Index Finance and Mango Markets exploiters. The second account is what I call legal centralist. It's the classic idea that law and order depend on the state, right? There's no law and order outside the state and without it, there is just chaos. Now that's difficult to in the face of a rich legal pluralist scholarship, which I won't go into details of now, but I'll I'll I'll mention a little bit in the next slides.

But it's of course even harder to square with how DeFi actually operates. Now there's the 45 billion TVL on Ethereum mainnet. And then we also know in October 2025 for instance, during the biggest on-chain liquidation cascade, you know, billions of dollars were processed completely orderly with no, you know, negative spillovers or domino effects, which that is part of my Ethereum Foundation research because it touches on a fundamental question in law and economics about so-called incomplete contracts. But the overall PhD research on is on DeFi is private order funded by Claros that's taken up the

tech legal and governance hub at Devcon today if you want to go check out. All right, back to the accounts. The third account I propose is private ordering which is the production and enforcement of rules by non-state actors. The canonical cases are well known. I mean some of you know there was a famous film I I keep forgetting the name. New York's Diamond Traders is one. And you got the early European stock exchanges that emerged as private clubs in in New York, London, and Amsterdam.

And prison gangs of course. The Fishers of Marseille is by a professor at my department. Fishers in Marseille, France apparently for centuries have had this institution where they didn't have to go to state courts to resolve their disputes and they just like relied on community generated norms to prevent conflicts from arising in the first place, but also to resolve them. Then in Afghanistan, Kabul, money market traders in the main bazaar of the city. Um they they lend to each other, they they exchange. So there's a lively commercial life happening there again without the state.

And I propose that DeFi is probably a candidate for this lineage of scholarship as well. I I believe it's order or it seems to me it's order is produced by three things together, the protocol rules, right? So the smart contract code is low idea without the reductionist take. And then decentralized human discretion, coordination plus the shadow of the state which is this idea in the scholarship that eventually all this private order still depend on the state in some form or another. So think about like DAO legal wrappers and you know all these interactions vis-a-vis state law.

Um and of course the interesting thing about DeFi is that unlike these canonical cases that are you know based in physically proximate areas you know tight knit communities with repeat interactions where you know reputation matters. On DeFi it's it's mostly largely synonymous. We are globally dispersed and of course it's all digital mediated which is an unusual fit for the scholarship of course if it is if it is the case.

Okay, so how is is so what's a good example for private ordering? For this talk I'll just focus on one which is MEV, maximal extractable value. You might know that when you submit a transaction on the Ethereum blockchain for a few seconds your transaction will sit in this public waiting room called mempool. And anyone can see it for those few seconds and people found ways to exploit that information asymmetry. They will deploy bots, it's called searcher bots and they will launch what called a sandwich attack.

So sandwiching is when, you know, they'll submit a transaction before your um token swap, let's say, and then they'll submit another after your token swap essentially um selling it uh selling the token to you at a higher price as much as your slippage allows, of course. Now sandwiching is permitted by the code, um but the legal status ambiguous. So the so-called codex law allows it, um and the state law is silent. Yet the community came to describe it as exploitative and unfair.

Um in other words, the people closest to the problem diagnosed it as a market failure. And the classic antidote to market failures in economic sense uh in in law is state intervention. Um but that's not quite what followed. Through community research efforts alliances you know, experimentation of different solutions all invented within DeFi, um MEV problem began to lessen or or invented on the application layer or the block protocol layer is a better word of using it. But of course MEV is a DeFi problem because it's on on DEXs and AMMs.

Um in fact, I remember myself a couple of years ago losing a chunk of money on Uniswap during a token swap to um to an MEV bot, uh only then to learn to adjust my RPC settings on on Metamask to flash bots, um which was also no longer necessary at some point because then I would use MEV protected protected uh DEX aggregators like CowSwap, LamaSwap, uh which is now becoming the industry standards. Um I had saved this flash bots data on from June last year and I don't believe that dashboard is active anymore, but it shows you the trend. You can see the share of daily Ethereum transactions sent through private channels increasing, and this is of course the amount of transactions, right? It's not

uh the volume. Um in fact around that time, October 25th last year in New York, I was talking to this MEV researcher who said it's about 90% of volume on Ethereum that now goes through private relays, which I believe is still the same now. So we have pretty much eliminated the malicious MEV issue by and large, and the community didn't need financial regulation to, first of all, qualify the problem or give its give it directions. But we've been there before. Go back to the Chicago Board of Trade from the 19th century right up until the mid 2000s when the electronic

trading took over. Proximity to to the trading pit, the place in the middle was an edge for traders. You stand closest to the pit where trades took place. You hear orders first, trade ahead ahead of them. And on Ethereum, the equivalent is your position in the block and the order or the sequence of it. So a sandwich is trading ahead of a customer's order down through sequencing rather than shouting. So what if Ethereum DeFi is a private order? What if it joins this other cases in the scholarship? Other than this colorly contribution, of course, there's a practical

implication. The state never simply ignores a private order. Interesting paper by legal scholar Ralph Michaels. He says that the state responds in one of four ways. It can reject it, refusing to treat it as law, incorporate it, defer to it, and delegate to it. And there's case law from different countries showing this. But the state courts need compelling theory and empirical demonstration for them to be able to defer to rather than superimposing rules from another industry like TradFi onto DeFi or just inventing new rules without paying attention to what the community has produced.

With that, I'll leave it here and happy to talk to anyone about it later during today. You can also find me on