Flawed Incentives Threaten Ethereum's Layer-2 Security, Study Finds

A study by researchers from zkSecurity, Prooflab, and Imperial College London has identified critical flaws in Ethereum’s Layer-2 rollup fee models, warning that mispriced small transactions could expose the ecosystem to systemic risks. The paper, titled “Unaligned Incentives: Pricing Attacks Against Blockchain Rollups,” highlights that current fee mechanisms for computation, data availability, and settlement are overly simplified, leading to potential over- or underpricing of transactions. This mispricing creates vulnerabilities, including inflated user costs and opportunities for denial-of-service (DoS) attacks, as attackers exploit underpriced transactions to spam networks at low cost Decrypt.co^[1].

Rollups, which bundle transactions to reduce costs on Ethereum, rely on three distinct resource costs: computation (executing transactions), data availability (posting data to the blockchain), and settlement (posting proofs to layer-1). However, the study found that most rollups collapse these into a single formula or fixed rules, failing to account for their independent variability. For example, Polygon zkEVM, zkSync Era, Scroll, Optimism, and Arbitrum employ divergent fee structures, with some networks fixing charges at transaction submission, others waiting until batches are sealed, and some offering refunds if actual costs fall below estimates. These mechanisms, while seemingly technical, create exploitable seams. A refund system, for instance, allows attackers to submit mass transactions and reclaim partial fees while consuming network resources Decrypt.co^[1].

The risks extend beyond user frustration. The paper argues that mispricing could enable attackers to subsidize small transactions, overwhelming networks and degrading performance for legitimate users. These vulnerabilities stem not from faulty code but from economic design choices that misalign incentives. With Ethereum’s rollup ecosystem securing tens of billions in assets, the study emphasizes the urgency of addressing these gaps. “Ignoring these incentive gaps is no longer safe,” the authors wrote, noting that flat or static fee curves are particularly susceptible to exploitation Decrypt.co^[1].

Benchmarking five major rollups, the researchers observed wide disparities in fee modeling. For instance, some networks dynamically adjust fees based on demand, while others use static rates. The arXiv paper “Optimistic MEV in Ethereum Layer 2s” further underscores the issue, revealing that speculative transactions—such as cyclic arbitrage—account for over 50% of on-chain gas on Base and Optimism. These optimistic MEV transactions, which rely on on-chain smart contracts, contribute to persistent spam-like activity, consuming half of on-chain gas while paying less than a quarter of total fees. The study attributes this to fee models that fail to account for variable proving costs tied to zero-knowledge virtual machines (zkVMs) arXiv^[3].

To mitigate these risks, the authors advocate for “multidimensional” fee mechanisms that separately price computation, data posting, and proving. Aligning fees with actual resource use, they argue, would enhance resistance to spam and improve cost predictability for users. Tools like dynamic adjustment, partial batching, and cost disclosure are proposed as solutions. While some rollup teams are experimenting with adaptive fee curves and real-time modeling, the study notes that no standardized framework exists yet. The findings coincide with Ethereum’s push for rollup-centric scaling, where zero-knowledge proofs and zkVMs are central to throughput goals. However, without addressing fee variability, models risk breaking under stress, leading to inconsistent user experiences and degraded service Decrypt.co^[1].

For developers and investors, the study underscores the importance of scrutinizing fee calculations beyond headline metrics like throughput or nominal low fees. “Incentives are security,” the authors argue, urging the Ethereum community to integrate transaction pricing into consensus design. As rollups scale, the interplay between fee structures, user behavior, and network security will become increasingly critical. The study’s release follows broader debates on Ethereum’s economic model, including criticism of proposals to impose fees on Layer-2s, which some argue would undermine scalability and ecosystem growth CryptoPotato^[2].

The research highlights a pivotal challenge for Ethereum’s future: balancing accessibility with security. While Layer-2s have reduced costs and increased capacity, their fee models must evolve to prevent exploitation. As the ecosystem matures, multidimensional pricing could become a standard, ensuring that rollups remain both efficient and resilient against adversarial behavior. For now, the call to action is clear—addressing misaligned incentives in fee design is not just a technical fix but a foundational step toward sustainable blockchain scaling Decrypt.co^[1].