Crypto Law Seminar #5 (28 June 2024): “Regulatory Implications of MEV Mitigations”

Notes from the Crypto Law Seminar #5 on the paper “Regulatory Implications of MEV Mitigations” (Yan Ji and James Grimmelmann).

  1. There was significant debate among the seminar participants about whether and to what extent MEV activities fall within the scope of market abuse as defined under Title VI of the Markets in Crypto-Assets Regulation (MiCA).
    • A key point of discussion was distinguishing between “good” and “bad” forms of MEV:
      • Good MEV was described as MEV that promotes healthy market functioning and efficiency. Examples included MEV that enables arbitrage between decentralized exchanges (leading to price convergence), MEV that enables liquidations to happen quickly to protect lenders, and MEV that generally reduces gas costs. The argument is that without good MEV, DeFi markets would be less efficient.
      • Bad MEV was described as MEV that is extractive and harms other market participants. Clear examples of bad MEV that could constitute market manipulation include:
        • Sandwich attacks, where a MEV bot or searcher places orders before and after a victim’s trade to extract value. There was debate on whether sandwiching with public mempool info is inherently manipulative.
        • Time-bandit attacks that involve a malicious MEV bot reorging blocks for their own gain. This harms the integrity of the blockchain.
        • Usage of inside information obtained from “dark pools” or private mempools to trade in an advantaged way. This raises issues of fairness.
    • However, participants noted the difficulty with clearly defining and identifying all forms of bad MEV that cross the line into market abuse. Regulators may be tempted to take an overly broad view that most MEV is manipulative.
    • There was a general view that more clarity and nuance is needed from regulators on precisely what specific MEV activities would be considered market abuse, rather than a blanket categorization. Relevant factors could include the intent behind the MEV, the information advantages employed, and the effects on market efficiency and integrity.
    • Some participants expressed concern that an overly expansive approach to regulating MEV as market abuse under MiCA Article 92 could harm DeFi - it could lead to the “death of the public mempool” if professional market makers feel they cannot engage in any MEV at all without risking liability. An overly restrictive approach could push towards centralization.

Motivation and Background

  • The paper was motivated by a desire to understand the legal implications of MEV (Miner Extractable Value) exploitation.
  • From a computer science perspective, MEV extraction is often seen as unethical because it harms ordinary users.
  • The authors wanted to explore what the law says about MEV and build upon previous legal analysis, particularly in light of rapidly evolving MEV practices and landscapes.
  • The paper examines the intersection of technical details and legal treatment of software systems, particularly in blockchain and DeFi contexts.

Proposed Alternative Standard for Liability

  • The authors propose an alternative standard for evaluating liability in MEV cases.
  • This standard focuses on whether an actor profits from routing a user’s transaction contrary to the user’s intentions or using information not designated for such purposes.
  • The approach is based on ideas from trade secret law and implied confidentiality.
  • It considers the user’s intent as legally relevant, especially in cases where the party receiving the transaction wouldn’t have received it without some expectation of how they’ll deal with the information.

Reasonable User Expectations in DeFi

  • The seminar discussed the challenge of determining what constitutes “reasonable” user expectations in the DeFi space.

  • Considerations include whether standards should differ for retail versus institutional participants. However, the difficulty of identifying which kind of market participant one is transacting with, and of measuring or observing these expectations was highlighted.

  • Methods discussed included user surveys and observing on-chain behavior.

    • The seminar included a significant discussion on the methodological approaches to understanding norms accepted among technology users, particularly in the context of blockchain and DeFi. This is a crucial question for both researchers and policymakers trying to determine what constitutes “reasonable” expectations or behavior in these rapidly evolving technological ecosystems. The discussion centered around two main approaches: surveys and observing behavior.

    • Survey Approach:

      • The seminar participants discussed the possibility of using surveys to gauge user expectations and norms. This method involves directly asking users about their beliefs, expectations, and understanding of various practices within the blockchain and DeFi space. However, several challenges were noted:
        • 1. Representativeness: Ensuring a truly representative sample of users across different levels of expertise and engagement with the technology is difficult.
        • 2. Complexity of questions: Given the technical nature of many blockchain and DeFi concepts, formulating survey questions that are both accurate and understandable to a wide range of users is challenging.
        • 3. Rapidly changing landscape: The fast-paced evolution of the technology means that survey results might quickly become outdated.
        • 4. Potential disconnect between stated beliefs and actual behavior: Users might express certain expectations or ethical standards in a survey that don’t align with their actual behavior when using these technologies.

    • Observing Behavior Approach:

      • The alternative method discussed was observing user behavior directly on the blockchain. This approach has several advantages:
        • 1. Objectivity: It provides hard data on how users actually interact with blockchain systems and DeFi protocols.
        • 2. Scale: It potentially allows for analysis of a much larger set of users and transactions.
        • 3. Real-time insights: Behavioral observation can provide up-to-date information on evolving norms and practices.
      • However, this approach also has limitations:
        • 1. Interpretation challenges: Observed behavior doesn’t always clearly indicate user intent or expectations.
        • 2. Privacy concerns: Extensive behavioral analysis might raise ethical questions about user privacy, even on public blockchains.
        • 3. Technical complexity: Accurately interpreting on-chain behavior requires sophisticated analysis tools and deep understanding of the protocols involved.
    • The seminar participants seemed to lean towards favoring the behavioral observation approach, suggesting that it might be more likely to be used by courts when determining what constitutes normal or expected behavior in the blockchain space. However, they also acknowledged that a combination of both methods might provide the most comprehensive understanding.

    • It was noted that there is currently a lack of robust research in this area. One participant mentioned knowing of only one paper from ETH Zurich that attempted to document user expectations in DeFi, but even this was described as potentially anecdotal rather than a comprehensive study.

    • The discussion highlighted the need for more research in this area, emphasizing the importance of developing methodologies that can accurately capture and interpret the rapidly evolving norms and expectations in the blockchain and DeFi space. This understanding is crucial not only for academic purposes but also for informing policy decisions and legal interpretations in this emerging field.

  • As technical mitigations become commonplace, they may influence legal interpretations of reasonable expectations.

    • The seminar participants discussed how the use of robots.txt files became a norm for controlling web crawlers' access to websites. Initially, when web crawling became prevalent, there wasn’t a standardized way for website owners to indicate which parts of their sites should or shouldn’t be crawled. The robots.txt file emerged as a solution, and it quickly became an accepted standard in the web community.
    • The example illustrates how, in the early days of the web, if a bot crawled a website and the site didn’t have a robots.txt file, it might have been considered that the site owner gave an implied license for the bot to view and index the content. However, once robots.txt became widely adopted and understood, the legal interpretation likely shifted. If a website had a robots.txt file in place specifying certain restrictions, and a bot ignored these instructions, it would be more likely to be seen as a violation of the site owner’s wishes and potentially subject to legal consequences.

Recent Legal Developments

  • The discussion touched on the recent indictment of two brothers for an attack on MEV boost.
  • The use of wire fraud statutes in this case, rather than securities laws, was noted as an interesting choice.
  • This case was seen as potentially implicitly approving of sandwiching or at least seeing it as normal, as the indictment didn’t suggest anything wrong with sandwiching.

Emerging Technologies and Their Impact

  • The seminar explored the potential impact of new technologies like Trusted Execution Environments (TEEs).
  • Protocols allowing users to express more detailed intents for their transactions were discussed.
  • These developments may influence future legal analyses of MEV and related practices - for instance, the fact that a vast majority of Ethereum users opt for MEV mitigating technologies like MEV-Boost may inform a legally relevant norm that certain MEV strategies (e.g., sandwich attacks) are unethical.
  • The discussion touched on projects like MEV-share and SUAVE, which aim to give users more control over their transaction information and intents.

Applying Traditional Legal Concepts to Blockchain

  • The conversation addressed the challenges of applying traditional legal concepts like property rights and contracts to blockchain transactions.
  • The need for careful disaggregation of the technical steps involved in decentralized exchanges was emphasized.
  • The discussion explored how to classify different parts of a blockchain transaction in legal terms.

Broader Implications for Blockchain Technology

  • The seminar touched on the potential impact of legal frameworks on censorship resistance in blockchain systems.
  • The legal status of validators and block producers was discussed, particularly in light of proposals like Ethereum’s inclusion lists.
  • The concept of “legally credible neutrality” for validators and block producers was introduced.

Education and Understanding

  • The importance of educating both lawyers and non-lawyers in the blockchain space about legal concepts was emphasized.
  • The need to help people think about law in a different way, moving beyond simple rules to understanding complex systems and concepts, was discussed.
  • The challenge of balancing regulation with the inherent properties of blockchain systems, such as public transparency, was highlighted.

Ongoing Challenges

  • The seminar emphasized the ongoing challenge of developing legal principles that can adapt to rapidly changing underlying technologies in the blockchain and DeFi space.
  • The difficulty of finding appropriate analogies between traditional finance and DeFi was noted.
  • The need for continued research and discussion in this area was emphasized.