Ethereum Builder Concentration Hits 93% as Research Flags DeFi Recentralization Risks

Decentralized finance (DeFi) was built to remove intermediaries, but new research argues it has instead created a fresh layer of middlemen—one that is increasingly concentrated. A report from CEPR and a University of Turin-linked research group warns that the market structure emerging around Ethereum (ETH) and on-chain trading is pulling DeFi back toward ‘recentralization’, undermining one of the sector’s core narratives.

The paper, titled Can Blockchain Decentralize Money, Contracts, and Finance? and published in February 2026 by a team including Prof. Bruno Biais, extends earlier concerns about stablecoins’ tendency toward centralized control. Its sharper critique, however, focuses on DeFi: the study argues that while the technology removes familiar gatekeepers, it recreates familiar frictions—information asymmetry and order manipulation—under new names and with crypto-native tooling.

Old market problems, new labels

The report highlights two recurring patterns.

First is the transfer of value from slower participants to faster ones—a dynamic the authors frame as a classic information advantage problem. Automated market makers (AMMs), popularized by Uniswap’s 2018 design, use algorithms to quote prices and execute swaps against pooled liquidity supplied by users known as ‘liquidity providers’. While AMMs eliminate the need for a traditional dealer, the report argues liquidity providers still bear losses when better-informed or faster traders trade against them.

In crypto markets this effect is widely discussed as ‘impermanent loss’, a term that suggests a peculiarly blockchain-era tradeoff. The report contends the economic substance is older: liquidity providers are effectively underwriting trading opportunities for informed counterparties, and the costs resemble traditional losses driven by information gaps rather than an entirely novel phenomenon.

Second is transaction reordering—more openly adversarial in nature. Because many blockchains expose pending transactions in a public queue (the mempool) before they are finalized, bots can detect a large trade and insert their own transactions around it. In a typical ‘sandwich attack’, a bot buys ahead of a victim’s trade, pushes the execution price against them, then sells immediately after—capturing a spread while the victim receives worse pricing.

The paper groups these opportunities under ‘maximal extractable value’ (MEV), stressing that MEV represents a transfer rather than a net gain for the market. In the authors’ framing, the extracted ‘value’ is profit for the extractor but a cost for the victim, generating no broader social surplus. If such behavior becomes pervasive, they warn, it could discourage ordinary users from trading on-chain, reducing participation and harming market quality in venues that depend on continuous order flow and liquidity.

Ethereum’s new chokepoint: block builders

The study’s most striking evidence focuses on Ethereum’s post-merge transaction supply chain. After Ethereum’s shift in 2022 from proof-of-work mining to proof-of-stake validation, block production became more predictable in one key respect: the right to propose the next block rotates according to protocol rules rather than being won through open-ended mining competition. That predictability, the report argues, creates a momentary ‘exclusive’ position for the next proposer—an opening that has supported the rise of specialized firms assembling blocks to maximize fee revenue.

These entities, commonly referred to as ‘block builders’, aggregate transactions and construct candidate blocks, then compete in auctions to have those blocks accepted. The report emphasizes that the builder market contains strong scale dynamics: if participants believe one builder is most likely to win, it becomes rational to route more order flow to that builder, reinforcing its advantage and creating a feedback loop where ‘concentration begets concentration.’

Citing data from analytics firm Rated, the authors report that as of Nov. 10, 2025, the top three Ethereum block builders accounted for 93.19% of block construction. For a network whose brand rests heavily on ‘censorship resistance’ and decentralization, the finding raises uncomfortable questions about how much practical control over transaction ordering and inclusion can end up in the hands of a small number of high-throughput operators—even if the base protocol remains open.

Tokenizing real-world assets still requires trust anchors

The report extends its critique beyond DeFi to the tokenization of real-world assets (RWA), including bonds, equities, real estate, art, wine, gold, and commodities. While tokenization can improve transferability and potentially broaden access, the authors argue it cannot eliminate centralized dependencies when off-chain rights must be recognized, enforced, and updated in the real world.

Smart contracts are powerful, the paper notes, when they operate entirely on-chain. But when a token is meant to represent a legal claim—such as ownership of a house or a parcel of land—critical questions reappear: Does the token transfer automatically change legal title? Will courts recognize the on-chain record as dispositive? Is notarization required, and who holds authority to correct errors or resolve disputes?

Bridging these gaps typically requires ‘oracles’—connectors that feed external information into blockchain systems. The report warns that if oracle control is effectively centralized, the system reintroduces the very ‘trusted intermediary’ blockchain architectures were designed to avoid. As an example, the authors point to a land registry initiative in India’s state of Andhra Pradesh, where records may be stored on a blockchain but transfers still require an official to submit updates using government-controlled keys—leaving ultimate authority with a centralized institution.

Implications for Korea: STOs, fractional investing, and narrative risk

The findings carry direct relevance for South Korea’s ongoing push into security token offerings (STOs) and ‘fractional’ investing models tied to assets such as specialty goods, gold, music royalties, and real estate. The report’s message is that the hardest problem is not issuing tokens; it is ensuring ‘rights consistency’—that the token’s on-chain status reliably matches the off-chain legal reality, with clear disclosure standards, enforceable claims, and credible mechanisms to prevent double-selling or resolve conflicts.

More broadly, the authors argue that decentralization should be treated as an empirical claim, not a marketing assumption. The 93.19% figure in Ethereum’s builder market underscores that economic forces can drive centralization even on open networks, complicating arguments that systems are inherently safer or fairer simply because they are labeled ‘decentralized.’

The report also notes that concentration is not limited to on-chain infrastructure, pointing to centralized exchange market share in which Binance plays an outsized role. In that context, the paper suggests the industry’s real test is whether blockchain-based markets can reduce ‘excess rents’ taken by intermediaries without sacrificing the efficiency benefits of scale.

Jean-Charles Rochet of Toulouse School of Economics, cited in the discussion, distills the skepticism more bluntly: “A ledger is inherently centralized. Who really wants a ‘scattered ledger’?” The question, the report implies, is not whether intermediaries can be eliminated altogether, but whether the new intermediaries forming around DeFi—MEV actors, builders, and oracle providers—can be constrained in ways that preserve open access and limit extractive behavior.

For an industry that has long sold decentralization as both innovation and protection, the paper reads as a warning that market structure, not ideology, ultimately determines who captures value—and whether the next generation of ‘middlemen’ looks any different from the last.