N Y U EW ORK NIVERSITY S L CHOOL OF AW LAW AND ECONOMICS RESEARCH PAPER SERIES WORKING PAPER NO. 24-13 The Illusion of Democracy—Why Voting in Decentralized Autonomous Organizations Is Doomed to Fail Xuan Liu June 2024 Electronic copy available at: https://ssrn.com/abstract=4441178 The Illusion of Democracy— Why Voting in Decentralized Autonomous Organizations Is Doomed to Fail Xuan Liu1 Abstract This article presents the first in-depth analysis of the reality of voting within Decentralized Autonomous Organizations (DAOs) by examining over 4,963 voting events through a combination of quantitative and qualitative research methods.2 DAOs have risen to prominence as an innovation in business organization models, with advocates highlighting their potential to offer democratic, non-hierarchical governance, lower agency costs, and evade regulatory oversight. However, this study reveals that DAOs are inherently susceptible to low voter participation, contradicting their decentralized ethos by potentially centralizing decision-making power and casting doubt on their democratic promise. These findings carry significant implications for the classification of DAO tokens as securities, hinging on their genuinely decentralized decision-making attributes. This article suggests policy interventions to stimulate voter engagement and foster genuine decentralization. Additionally, it introduces a novel regulatory framework—the “probationary business classification”— for DAO classification, considering the dynamic nature of DAO governance. This approach seeks to reconcile the decentralized ambitions of DAOs with the realities of their operational structures, potentially reshaping the regulatory landscape for DAO tokens. 1 Wagner Fellow of Law and Business, New York University. 2 A ‘voting event’ refers to a decision-making process regarding a particular proposal within a DAO. The number of participants in each event varies, with voter turnout ranging from a handful to several thousand. This study’s primary data comprises these voting records, which contain critical details pertinent to the voting outcomes. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table of Contents I. Introduction ...................................................................................................................................... 3 II. Overview of DAOs and their Voting Mechanisms........................................................................... 6 A. From Bitcoin to Ethereum: Transitioning from a Decentralized Payment System to an Open- Source Smart Contracts Application Platform .......................................................................................... 6 B. The Ideological Appeal of DAOs ..................................................................................................... 8 1. Ideological Motivations and Special Interests Attracting Investors to DAOs ............................. 9 2. DAOs as a Protective Layer for Blockchain Entrepreneurs ....................................................... 11 C. The Evolution and Investment Landscape of DAOs ...................................................................... 13 D. DAO Token Economics and Market Dynamics ............................................................................. 15 III. Hypothesis of Low Voter Participation and the Underlying Theoretical Basis ............................. 17 IV. Data Construction and DAO Governance Fundamentals: A Comprehensive Overview ............... 20 A. The Dataset Construction ............................................................................................................... 20 B. Treasury Size, Token Holder Count, and Industry Distribution ..................................................... 21 C. An Overview of DAO Voting Rules .............................................................................................. 23 D. Voting Proposal Topics and Support Received .............................................................................. 26 V. Empirical Findings: Validating Hypotheses on DAO Voting Behavior ........................................ 31 A. Overall Low Voter Turnout Rates .................................................................................................. 32 B. Concentration of Voting Power among Large Token Holders ....................................................... 33 C. Wide Disparity in Token Holder Interest Among Different Proposals .......................................... 35 D. Large Token Holders Dominate DAO Voting Activities through Active Participation ................ 38 E. A Deeper Look into Uniswap’s Voting Data: Low Voter Participation and Misaligned Interests 41 VI. Moving Forward: Three Ways for Improving the DAO Voter Participation ................................. 44 A. Creating Fluid Majorities: The Role of Quadratic Voting in Mobilizing the Latent Majority and Enhancing DAO Governance .................................................................................................................. 44 B. Enhancing DAO Governance through Committee-Based Decision Making and Special Incentives for Proactive Participants ........................................................................................................................ 49 C. Standardized Information Disclosure Requirements and Guidance for Enhancing Voter Participation and DAO Governance ........................................................................................................ 51 VII. Moving Forward: One Novel Regulatory Approach for DAOs Classification .............................. 54 VIII. Conclusion ...................................................................................................................................... 56 Appendix: .................................................................................................................................................... 58 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 I. Introduction Humans distinguish themselves from other species by their ability to construct and disseminate narratives about previously non-existent concepts. These narratives enable individuals to form collective beliefs, which in turn can garner followers and instigate changes in behavior across various domains, be it religion, money, or institutions and political structures.3 The Decentralized Autonomous Organization (DAO) exemplifies such a narrative, serving as a unique type of business organization that challenges traditional organizational form. This paper examines the extent to which DAOs fulfill their narrative of decentralized governance and identifies the implications of their operational realities. To decentralize or not to decentralize: this question holds significant importance. Blockchain technology has not only revolutionized data ownership but also transformed voting, enabling individuals to initiate and participate in votes anytime and anywhere, rendering physical attendance at annual shareholder meetings obsolete. However, the alleged democratic nature of DAOs has been a subject of controversy. Skepticism surrounds the authenticity of its decentralized decision-making process, as concerns mount over the potential for influential token holders—like early blockchain entrepreneurs, tech giants, and wealthy venture capitalists—to wield significant power in the DAO voting process. This calls into question the system’s fairness and leads to a lingering inquiry: is the DAO genuinely decentralized, or is it merely a facade for a new form of plutocracy? Lawmakers today face the challenge of effectively regulating decentralized organizations. The Securities and Exchange Commission (SEC) has stated that in most cases, it will consider DAO tokens as securities (specifically, as investment contracts), which necessitates that DAOs comply with federal securities laws.4 This compliance will involve registering tokens and adhering to ongoing public reporting requirements. Proponents of DAOs argue that they are fundamentally distinct from traditional centralized corporations, as their decision-making processes are entirely 3 Yuval Noah Harari, Sapiens: A Brief History of Humankind 93-122 (HarperCollins 2014). 4 Securities and Exchange Commission, Report of Investigation Pursuant to Section 21(a) of the Securities Exchange Act of 1934: The DAO, Release No. 81207 (July 25, 2017), https://www.sec.gov/litigation/investreport/34-81207.pdf (last accessed January 30, 2024) EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 decentralized. As such, these proponents believe that DAOs warrant unique and more favorable legal treatment. The SEC has also acknowledged that if a DAO demonstrates genuinely decentralized decision-making, its tokens may not be classified as securities.5 Hasty regulatory efforts without a comprehensive understanding of DAOs’ context may lead to unintended consequences.6 Consequently, it is imperative to examine the decentralized nature of DAOs and gain a comprehensive understanding of their operations before implementing any regulatory measures. Furthermore, if DAOs do not exhibit distinctive features compared to traditional corporations and partnerships, their existence would be rendered unnecessary. DAO proponents argue that these organizations facilitate egalitarian and widespread participation in decision-making. Unlike traditional corporations, where shareholder voting is restricted to certain decisions like electing directors or approving mergers, every DAO token holder can initiate proposals and vote on them, with smart contracts executing the decisions once they achieve majority support. It is this key difference that sets DAOs apart from conventional corporate structures and justifies their existence.7 Nonetheless, drawing on Olson’s (1965) theory of collective action and the Chicago School of Law and Economics (Easterbrook and Fischel, 1991), the likelihood of achieving such broad-based involvement in a large, voluntary organization with numerous members seems improbable. The reality, rather than the ability, that everyone does vote and actively participates 5 William Hinman, Dir., Div. of Corp. Fin., Sec. & Exch. Comm’n, Remarks at the Yahoo Finance All Markets Summit: Crypto: Digital Asset Transactions: When Howey Met Gary (June 14, 2018), https://www.sec.gov/news/speech/speech-hinman-061418 (last accessed January 30, 2024) 6 If regulations are overly lenient, it could lead to tax revenue losses and provide a pathway for evading securities regulations. An increasing number of businesses may adopt the DAO model to circumvent legal oversight. On the other hand, if regulations are excessively strict, they may stifle innovation, inadvertently suppressing the development of new technologies and limiting the growth of the blockchain and DAO ecosystem. Additionally, such regulations could drive innovation to other jurisdictions, resulting in a loss of talent and economic growth in the US. 7 Mancur Olson illustrated that large voluntary associations prosper because they can perform functions – such as addressing demands, advancing interests, or fulfilling needs – more effectively than other organizational forms. These demands or interests serve as incentives for the establishment and sustainment of voluntary associations. If traditional corporate structures can meet these demands and interests, the need for alternative forms of organizations may be called into question if they cannot achieve purposes beyond economic benefits, such as democracy. See David Lowery, Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups, in The Oxford Handbook of Classics in Public Policy and Administration 205 (Steven J. Balla et al. eds., 2016) EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 in decision-making serves as a core incentive, if not the sole incentive, for adopting the DAO structure. This empirical study lends support to the hypothesis that DAO governance, in its current state, is predominantly centralized and merely presents an illusion of democracy. The dataset was compiled from all voting records obtained from Snapshot and Tally.8 The analysis focuses on voting behavior and its correlation with various variables such as proposal topic, voter participation rate, total votes for and against the proposal, the number and identity of key voters, among others. The findings unveil a consistent pattern of low voter participation rates and decision-making dominance by large token holders. This article is structured as follows: Section I introduces the topic. Section II provides an overview of DAOs, their voting rules, and the rationale behind the choice of DAOs by investors and entrepreneurs. Section III explores the theoretical framework and examines the hypothesis of low voter participation in DAO voting. Section IV outlines the dataset and empirical methodologies employed in this study. Section V presents the empirical findings and offers a comprehensive analysis of the results. Section VI and VII provide policy recommendations. Section VIII concludes the article. Lastly, the Appendix contains supplementary information on the dataset and research methodology utilized in this study. 8 Both Tally and Snapshot serve as platforms for voting and recording voting results for DAOs. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 II. Overview of DAOs and their Voting Mechanisms DAOs are self-governing entities that operate on blockchain networks, utilizing smart contracts to automate their various processes.9 These smart contracts execute predefined rules and conditions, streamlining operations and transactions without depending on centralized authorities or individual third-party intermediaries. Leveraging blockchain technology, DAO members can participate in decision-making and consensus-building processes, utilizing tokens as a form of ownership analogous to shares in a corporation.10 Navigating the world of crypto could be daunting for beginners, as terminology such as Crypto, Bitcoin, Blockchain, Smart Contract, Protocol, and DAO is scattered throughout various contexts, yet rarely do explanations provide a coherent storyline to connect their relationships and logical progression. This section endeavors to provide a brief introduction to the technological underpinnings and historical context of DAOs, with the goal to enhance readers’ understanding of the decision-making processes and dynamics within this domain. A. From Bitcoin to Ethereum: Transitioning from a Decentralized Payment System to an Open-Source Smart Contracts Application Platform At the heart of the cryptocurrency industry is the principle of decentralization, which aspires to eliminate reliance on traditional trusted third-party intermediaries, such as banks and governments.11 Blockchain technology, initially developed in 2009 as a peer-to-peer electronic cash protocol, empowers individuals with the capacity to create, transfer, and store money without depending on centralized entities. This process entails generating currency without a central 9 Blockchain technology provides a secure and distributed method for storing and transferring data using cryptographic techniques and a network of computers. This system can generate permanent and transparent records for various activities, ensuring data integrity and security. 10 Daniel Drescher, Blockchain Basics: A Non-Technical Introduction in 25 Steps (Springer International Publishing 2018). 11 The inception of blockchain technology and Bitcoin was rooted in early developers’ distrust towards banks, stemming from historical events such as the 2008 financial crisis and the hyperinflation crisis (with 500 percent annual inflation rates) that Chile experienced in the 1970s. This context partly explains why some early crypto pioneers originated from South America. These individuals harbored skepticism towards banks controlling the money supply and distrusted the centralized control exerted by banks and other financial institutions. Blockchain technology presents a decentralized, transparent, and accessible alternative, addressing many concerns associated with traditional banking systems. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 authority, transferring funds without banks or financial institutions, and securely storing wealth transparently without intermediaries.12 Although the initial focus was centered on displacing banks and financial institutions, the blockchain landscape experienced a transformative shift in 2015. In 2015, Vitalik Buterin unveiled the Ethereum network, a distinct blockchain-based decentralized platform with open resources. Unlike Bitcoin, which was originally designed as a way to bypass traditional financial institutions and the centralized control they exert over monetary transactions, the Ethereum blockchain was created to enable developers to build various use cases of decentralized applications (dApps) that could operate autonomously and free from third-party intervention. Ethereum’s most notable feature is its support for smart contracts—self-executing agreements in which the terms between buyers and sellers are coded directly into the contract itself.13 Ethereum’s open and adaptable features facilitate the development of dApps capable of automating complex contracts and transactions. By supporting smart contracts and dApps, Ethereum enables a wide range of potential use cases beyond mere cryptocurrency transactions, such as decentralized finance (DeFi), non-fungible tokens (NFTs), gaming, identity management, and other applications. Ethereum also has its own native currency called Ether (ETH), which is used to pay for gas fees and power smart contracts.14 12 In 2008, an individual or group using the pseudonym “Satoshi Nakamoto” published a paper titled “Bitcoin: A Peer-to-Peer Electronic Cash System.” The first version of the Bitcoin software was released in January 2009, allowing users to create and manage their own transactions. The blockchain technology underpins the Bitcoin system and is maintained by a network of computers called nodes that validate and record transactions. Every participant in the blockchain network possesses a copy of the complete ledger, which records every transaction on the network. When a new transaction is initiated, it is broadcast to all participants who then authenticate and add it to the ledger. The ledger is a permanent, tamper-proof record of all transactions and does not require a centralized intermediary to manage or control the data. This makes the blockchain highly secure and resistant to tampering or hacking attempts. 13 Cryptocurrency enthusiasts often use the analogy of a vending machine to describe smart contracts. When someone wants a bottle of soda, they insert money into the vending machine, which automatically dispenses the drink based on a pre-programmed algorithm. The algorithm calculates the remaining balance according to the deposited amount, eliminating the need for a salesclerk. This process removes human interaction and related costs, as well as any positive or negative interpersonal interactions, streamlining the transaction. 14 A gas fee is the amount of Ether (ETH) required for a user to execute a transaction on the Ethereum blockchain network. These fees compensate Ethereum miners for verifying transactions and securing the network. Gas prices are expressed in gwei, a unit of ETH, with each gwei equal to 0.000000001 ETH (10-9 ETH). Gas fees may vary depending on transaction complexity and network demand. As of March 2023, the average gas fee ranges from 25 to 75 gwei, equivalent to $0.05 to $0.14 per transaction. For more information, refer to Ethereum.org, “Gas: Understanding Ethereum Fees” at https://ethereum.org/gl/developers/docs/gas/ (last accessed March 27, 2023). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 At its core, Ethereum provides the infrastructure and serves as the platform for smart contracts, with DAOs representing a specific application of these smart contracts. DAOs function as decentralized, autonomous business organizations, leveraging a bundle of smart contracts. Consequently, most DAOs initially launched on, and many still operate within, the Ethereum platform.15 B. The Ideological Appeal of DAOs Aside from the intricate terminology and technological advancements, it is crucial to recognize that DAOs fundamentally serve as a form of business organization. Therefore, they can be ultimately distilled down to the relationships between individuals and the associations among them.16 Within this organizational framework, individuals acquire voting rights and enjoys corresponding shares of profits or bears losses based on their token holdings. Both the underlying business owner and investors, motivated by ideologies and specific interests, prefer DAO as their organizational form and choose to invest in such entities. While the technological features of blockchain—such as decentralized data distribution and storage across a network of diverse nodes—are often highlighted, the more critical aspect lies in its ideological underpinnings. Technological advancements allow investors to cast vote and access results seamlessly, fostering a participatory governance model where stakeholders exercise decision-making power. This ideology resonates with the culture and values advocated by the crypto industry, making the decentralized governance model a natural fit. Furthermore, the innovative business applications developed on the blockchain via smart contracts are entirely new, necessitating fundraising efforts and the attraction of investors for growth. In early-stage ventures, promoting a distinct and appealing ideology can serve as a strong selling point. 15 Other platforms, such as EOS, NEO, and Cardano, also support smart contracts and DAOs and offer unique features like low latency, multiple programming language support, and a focus on scalability and sustainability. 16 A DAO fundamentally embodies a consortium of individuals and their interrelations, with each investor holding voting rights within the action-driven collective. DAO members share a mutual interest in the underlying business, and the organization’s foundation is built on the assumption that token holders, acting as self-interested entities, will contribute to the group’s objectives. This is because the group’s success is reflected in its tokens’ value, thereby affecting their investments. Nonetheless, as Mancur Olson’s research indicates, collective action dilemmas are inevitable in such groups. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 1. Ideological Motivations and Special Interests Attracting Investors to DAOs Investors are drawn to DAOs by three main ideologies: the promotion of democracy and community involvement, a belief in technological optimism or techno-utopianism, and a combination of entrepreneurship with speculation. The ideologies advocating for democracy and participation foster a sense of belonging and the perception that one’s voice holds value within the organization. The emergence of blockchain and Bitcoin concepts can be partly attributed to the 2008 economic crisis, during which many developed deep mistrusts towards government fiscal policies and the bailouts of large-scale financial institution, questioning their commitment to safeguarding the assets of the average person. Furthermore, memories of hyperinflation episodes in South America lingered.17 As a result, technologically literate individuals sought to create a decentralized, networked structure to replace centralized forms. They also harbored a distrust towards corporate managers, viewing them as mere components of a hierarchical system. Instead, they placed their faith in individual decision- making, believing that each person would act rationally and ultimately contribute to a rational organization. Thus, DAOs are seen not just as entities managing relationships between individuals and the organization but as embodiments of democratic principles themselves. Participants perceive DAOs as not only a means to achieve democracy but as an intrinsic part of democratic objectives themselves. Consequently, attaining a high voter participation rate is a fundamental goal for DAOs to ensure their democratic principles are upheld. While hierarchical structures are not intrinsically negative, they have been subject to criticism in the wake of corporate scandals like Enron.18 Some have also criticized the layers of management as inefficient and redundant.19 Proponents of DAO have called for its adoption as a replacement 17 Argentina underwent hyperinflation in 1989 and 1990, with annual rates exceeding 3,000% and 2,000%, respectively. Similarly, Brazil endured high inflation and hyperinflation from the mid-1980s to the mid-1990s, peaking at over 2,000% in 1993. These episodes help shed light on why many South Americans have become active participants in the crypto industry. 18 Many has seen the hierarchical structure as a means for upper management to prioritize their own interests over those of minority shareholders, resulting in decisions that may not serve the best interests of the company or its stakeholders. 19 Bethany McLean & Peter Elkind, The Smartest Guys in the Room: The Amazing Rise and Scandalous Fall of Enron 169-97 (2013). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 for the traditional corporate structure, highlighting the removal of hierarchical structures and the implementation of a flattened model as key benefits and appeals.20 However, the fundamental distinction between corporations and DAOs extends beyond these surface differences. Corporations are characterized by the separation of investor and management roles, while DAOs merge these roles, creating a unified entity where decision-making and investment responsibilities are integrated.21 Technical optimism, or techno-utopianism, embodies the belief that technological progress will yield more benefits than drawbacks for humanity, ultimately improving societal conditions. This mindset, characterized by the “underestimation and neglect of uncertainty” in favor of a “widely shared speculative promise,” positions technology as a driving force for advancement, innovation, and solutions to the world’s problems.22 This mindset is highly prevalent in the crypto industry. This optimism is evident in many YouTube interviews and podcasts about the crypto sector, where a recurring theme is the conviction that the crypto community is distinctively transformative, thanks to blockchain and related technologies. Old rules are being shattered, and new ones established, leading to the belief that many conventional assumptions and challenges no longer apply in the crypto sphere. Therefore, even though individuals are aware that various conflicts of interest, gaming, and competition can arise in human interaction and collaboration, they are overly optimistic in believing that technology can either eliminate these issues entirely or at least substantially mitigate their effects and implications. Entrepreneurship and speculation in the crypto industry represent an opportunistic approach in which individuals seek to capitalize on emerging opportunities and potential gains through their 20 Steinberg, R. (2022). Enhancing Traditional Corporate Governance with DAOs. Forbes. https://www.forbes.com/sites/forbesfinancecouncil/2022/03/31/enhancing-traditional-corporate-governance- with-daos/ (last accessed December 26, 2023) 21 Easterbrook, F. H., & Fischel, D. R. (1983). Voting in Corporate Law. The Journal of Law and Economics, 26(2), 395-427. Available at https://chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=2358&context=journal_articles (last accessed January 30, 2024) 22 Hochschild, J., Crabill, A., & Sen, M. (2012). Technology optimism or pessimism: How trust in science shapes policy attitudes toward genomic science. Issues in Governance Studies, 21, 1-19. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 involvement in DAOs. Under this ideology, the significance of the crypto industry and DAOs is multifaceted, offering participants not only the chance to earn economic returns but also to establish their identity within the industry. For those lacking coding skills, engaging in investment and management activities serves as a crucial entry point. The anticipation of the crypto industry’s immense potential for future growth motivates individuals to engage in DAOs’ contributions and management as early as possible. Establishing a track record in the crypto industry sooner rather than later can lead to greater opportunities in the future. And the past few years have witnessed a significant increase in the value of Bitcoin, Ether, and DAO tokens, which has attracted a substantial number of speculators. Many of these individuals have not engaged in careful deliberation but have instead spread their capital across the crypto industry. As a result, DAO tokens have emerged as an essential investment avenue for these participants. 2. DAOs as a Protective Layer for Blockchain Entrepreneurs Regulatory risk presents a major hurdle for entrepreneurs in the crypto industry, with the choice of business structure—partnership or corporation—entailing significant legal and tax implications. Securities regulations, in particular, pose an even greater threat. Following Bitcoin’s remarkable value surge in 2013, which thrust the crypto industry into the spotlight, the Securities and Exchange Commission (SEC) has undertaken 127 enforcement actions targeting a range of issues within the sector from 2013 to 2022.23 Crypto entrepreneurs often grapple with the uncertainty of whether their tokens will be classified as securities and struggle to navigate the evolving legislative landscape. DAOs provide early-stage entrepreneurs with a protective layer through a novel organizational structure, underpinned by technological barriers and distinct ideologies. Crypto tokens that facilitate economic alignment can be deemed securities under the Howey Test.24 Managing a distribution of securities to a large community of users can be challenging and expensive for a 23 Connor Sephton, Clear Rules for Crypto Industry Already Exist, SEC’s Gary Gensler Claims, Yahoo Finance (Mar. 30, 2023), https://finance.yahoo.com/news/clear-rules-crypto-industry-already-092656864.html. 24 SEC v. W.J. Howey Co., 328 U.S. 293 (1946). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788

involvement in DAOs. Under this ideology, the significance of the crypto industry and DAOs is multifaceted, offering participants not only the chance to earn economic returns but also to establish their identity within the industry. For those lacking coding skills, engaging in investment and management activities serves as a crucial entry point. The anticipation of the crypto industry’s immense potential for future growth motivates individuals to engage in DAOs’ contributions and management as early as possible. Establishing a track record in the crypto industry sooner rather than later can lead to greater opportunities in the future. And the past few years have witnessed a significant increase in the value of Bitcoin, Ether, and DAO tokens, which has attracted a substantial number of speculators. Many of these individuals have not engaged in careful deliberation but have instead spread their capital across the crypto industry. As a result, DAO tokens have emerged as an essential investment avenue for these participants. 2. DAOs as a Protective Layer for Blockchain Entrepreneurs Regulatory risk presents a major hurdle for entrepreneurs in the crypto industry, with the choice of business structure—partnership or corporation—entailing significant legal and tax implications. Securities regulations, in particular, pose an even greater threat. Following Bitcoin’s remarkable value surge in 2013, which thrust the crypto industry into the spotlight, the Securities and Exchange Commission (SEC) has undertaken 127 enforcement actions targeting a range of issues within the sector from 2013 to 2022.23 Crypto entrepreneurs often grapple with the uncertainty of whether their tokens will be classified as securities and struggle to navigate the evolving legislative landscape. DAOs provide early-stage entrepreneurs with a protective layer through a novel organizational structure, underpinned by technological barriers and distinct ideologies. Crypto tokens that facilitate economic alignment can be deemed securities under the Howey Test.24 Managing a distribution of securities to a large community of users can be challenging and expensive for a 23 Connor Sephton, Clear Rules for Crypto Industry Already Exist, SEC’s Gary Gensler Claims, Yahoo Finance (Mar. 30, 2023), https://finance.yahoo.com/news/clear-rules-crypto-industry-already-092656864.html. 24 SEC v. W.J. Howey Co., 328 U.S. 293 (1946). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 startup to handle—even industry giants like Airbnb and Uber struggle to achieve it.25 However, insights from recent SEC commentary and enforcement actions suggests that genuine decentralization may enable a startup’s token to transform from a security to a non-security if operations can be sufficiently decentralize, eliminating information asymmetry or dependency on the founding team for value creation.26 Compound, one of the largest protocols on the blockchain, had its general counsel explicitly state that adopting the DAO structure was motivated by the fact that decision-making authority was entirely in the hands of dispersed, non-aligned token holders. This arrangement could prevent their tokens from being regarded as investment contracts. Similarly, the prominent crypto venture capital firm a16z emphasized that achieving decentralized community control is essential for regulatory compliance.27 Furthermore, a16z recommends entrepreneurs in its portfolio adopt a strategy known as “progressive decentralization.” This approach entails that, following the creation of a product and the formulation of basic governance guidelines, the founding team should swiftly transition control to the community, ultimately attaining a sufficient level of decentralization.28 To avoid tokens from being classified as investment contracts and thus securities, early on-chain entrepreneurs created and adopted DAOs as their preferred organizational form. As the number of blockchain businesses grew, subsequent entrepreneurs gravitated towards this established route. Given the explosive growth in the diversity and quantity of on-blockchain businesses in recent years, DAOs have inevitably emerged as a focal point of interest in the business world. 25 Ali Yahya & Jesse Walden, Progressive Decentralization: Crypto Product Management, a16z Crypto, https://a16zcrypto.com/content/article/progressive-decentralization-crypto-product-management/ (last accessed January 30, 2024). 26 One key test for whether an investment might be deemed a security is whether investors are relying on the efforts of others with the expectation of profits. Post-network launch, provided that the network is sufficiently decentralized, the nature of the token can change from a security to a non-security, owing to the fact that the holder of the token is no longer relying on the efforts of others. William Hinman, Dir., Div. of Corp. Fin., Sec. & Exch. Comm’n, Remarks at the Yahoo Finance All Markets Summit: Crypto: Digital Asset Transactions: When Howey Met Gary (June 14, 2018), https://www.sec.gov/news/speech/speech-hinman-061418 (last accessed January 30, 2024) 27 Katherine Wu, Mutability: SEC & Recent Cases, a16z Crypto, https://a16zcrypto.com/content/article/mutability-sec-recent-cases/ (last accessed January 30, 2024). 28 Yahya & Walden, Progressive Decentralization. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 C. The Evolution and Investment Landscape of DAOs The concept of DAO first emerged in 2016 when a group of developers launched TheDAO, a decentralized venture capital fund allowing members to vote on proposals for fund allocation and investment in various projects.29 Since then, the proliferation of smart contract applications, coupled with enthusiasm from investors and prominent venture capital funds, has led to significant growth in the number of DAOs.30 As a result, DAOs have become a widely adopted business structure within the blockchain ecosystem. As of March 23, 2023, there were 11,123 DAOs, with a combined treasury of $13 billion.31 Among these, 156 boasted treasuries exceeding $1 million, 79 had over $10 million, and 21 had amassed over $100 million.32 DAOs encompass a broad range of sectors and activities, with a significant portion operating within Decentralized Finance (DeFi) sector. This is consistent with the fact that DeFi are among the most numerous and prevalent business projects on the entire blockchain. DeFi DAOs provide a wide range of services from lending and borrowing platforms to decentralized exchanges and insurance. This growth of DeFi DAOs stems from the blockchain’s original focus on digital currency, payment, and value storage purposes. As the blockchain industry evolves, it has led to the development of various financial tools and instruments. Furthermore, DeFi DAOs exemplify 29 The first DAO in history, known as TheDAO, was launched in 2016 with the purpose of coordinating investment in and governance of Ethereum projects. TheDAO successfully raised over $150 million worth of Ether, Ethereum’s native currency from more than 11,000 investors through a token sale, making it one of the largest crowdfunding campaigns in history. However, it soon encountered a significant setback when an attacker hacked and drained one-third of its funds. After the hack, TheDAO was effectively defunct. See CryptoSlate, Decentralized Autonomous Organizations (DAOs): An Overview, (last visited Feb. 5, 2023), https://cryptoslate.com/guides/what-are-daos/ and David Siegel, Understanding the DAO Attack, CoinDesk (last visited March 22, 2023) https://www.coindesk.com/learn/understanding-the-dao-attack/ 30 Notable figures and entities involved in early DAOs included billionaire investor Mark Cuban, who has invested in OlympusDAO and NounsDAO and considers DAO as “the ultimate combination of capitalism and progressivism.” Ethereum co-founder Vitalik Buterin, who advocates for DAOs as a means to coordinate collective action and governance on the blockchain. venture capital firm a16z, which has invested in and actively participated in various DAO projects; and Polychain Capital, a crypto-focused venture capital fund that has invested in multiple DAOs and DAO platforms, such as Compound, MakerDAO, MolochDAO, Aragon, and Colony. 31 Among them, Uniswap has a treasury of $2.5 billion, and its treasury once reached an all-time high of $11.5 billion in September 2021; BitDAO has a treasury of $2.5 billion, which is now approaching its treasury’s peak; Gnosis has a treasury of $1 billion, with its historical highest treasury also occurring in September 2021 at $3.1 billion. DeepDao. 2 Feb. 2023 deepdao.io/organizations. 32 However, it is also worth noting that 10,711 DAOs had no traceable treasury or voting activity so far. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 the dynamic nature of financial markets, fueling “creative destruction,” a process that continually challenges and supersedes old ideas and organizations with more innovative ones.33 These DAOs are celebrated for their robustness and adaptability, empowering a new generation of on-chain “bankers” to bypass government regulations and provide cross-border financial services. DAOs focused on developing blockchain infrastructure comprise the second largest category.34 This predominance is attributed to the fact that blockchain is still being in the early stages of development and the primary operational activities are centered around improving the technical infrastructure. Gnosis Safe and Rainbow are two examples that provide specialized blockchain infrastructure services to users.35 NFT DAOs hold the third position, owing to the recent surge in NFT transactions in the past two years.36 The significant surge in NFT prices and trading volume has drawn attention to NFT DAOs and the DAO structure has also provided a favorable environment for collective creators, enabling them to participate in the creation and management of NFTs with greater transparency and decentralized control.37 Investing in DAOs provides investors with the opportunity to engage in innovative and decentralized ventures, potentially earning a return on their investment while actively contributing to the organization’s direction and governance through DAO tokens. These tokens allow investors to share in the profits or losses of the DAO’s underlying projects, proportional to their investment. 33 Rajan, R. G., & Zingales, L. (2004). Saving capitalism from the capitalists: Unleashing the power of financial markets to create wealth and spread opportunity. Princeton University Press, p. 3. 34 An example of such a DAO is Compound, which is an algorithmic, autonomous interest rate protocol designed for developers, with the aim of unlocking a plethora of open financial applications. 35 Gnosis Safe facilitates the creation and management of multisignature wallets on the Ethereum network, where transactions require multiple signatures to increase security and accountability. Rainbow, on the other hand, offers a colorful and user-friendly wallet interface for Ethereum and non-fungible tokens (NFTs), allowing users to explore, collect, and trade NFTs as well as access various decentralized finance (DeFi) protocols and platforms. 36 NFT stands for Non-Fungible Token. It is a unique digital asset stored on blockchain that represents ownership or proof of authenticity of a piece of content, such as art, music, videos, or even tweets. The trading of NFTs has experienced explosive growth, with the market hit $17.6 billion in 2021, a staggering increase of 21,000% from 2020, as reported by Nonfungible.com. Even with months of declining sales and falling prices, the overall NFT sales volume in 2022 nearly matched the 2021 peak, underscoring the continued interest in NFTs as a valuable asset class. 37 Some examples of NFT DAOs include Flamingo DAO, which holds over 9,000 NFTs, including rare ones such as Cryptopunks and Bored Apes, as well as PleaserDAO, which focuses on acquiring and showcasing high- profile NFTs and supporting NFT artists through grants and collaborations. Another example is PartyDAO, which allows individuals to pool funds and bid on NFTs collectively. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Moreover, DAOs offer a diversified investment portfolios that includes everything from gaming, NFTs, and DeFi to social initiatives, allowing investors to choose based on their preferences and risk tolerance. Furthermore, investing in DAOs provides investors exposure to disruptive technologies and business models that are often inaccessible through traditional investment channels, while also fostering a sense of community and belonging to a new and revolutionary movement in the blockchain ecosystem. D. DAO Token Economics and Market Dynamics DAO tokens represent ownership or membership in a DAO and can serve various functions and rights, depending on the DAO’s design and purpose. These functions may include voting, governance, profit sharing, access to private forums or information, or utility. In most cases, DAO tokens serve as a combination of these functions, acting as both investment certificates and voting credentials. DAOs determine their total token quantity based on various factors, including their purpose, vision, governance model, funding needs, and chosen token economics. A universal formula for ascertaining an ideal token amount for a DAO does not exist, as different DAOs possess unique objectives and strategies. Some DAOs elect to adopt a fixed token quantity, wherein a predetermined number of tokens are generated and distributed at the organization’s inception.38 In contrast, other DAOs opt for an uncapped or unlimited token supply, allowing for the perpetual creation of new tokens.39 However, the majority of DAOs prefer a flexible token supply, facilitating the minting or burning of tokens in accordance with market demand, supply dynamics, and the organization’s overall performance. Minting refers to the process of generating new tokens and augmenting the existing token supply, while burning entails the elimination of extant tokens 38 Compound has a fixed supply of 10 million governance tokens. Approximately 4.2 million will be distributed to users over four years, 2.4 million to Compound Labs’ shareholders (the company behind Compound’s software development), and 2.2 million to the founders (subject to a four-year vesting schedule). Moreover, 775,000 tokens are allocated for governance incentives, and 332,000 for future team member incentives. See “Compound Governance,” Medium, accessed at https://medium.com/compound-finance/compound- governance-5531f524cf68 (last accessed January 30, 2024). 39 Uniswap has no cap on its UNI tokens; initially, 1 billion UNI were minted, with allocations over four years as follows: 60% to community members, 21% to team members and future employees (with a four-year vesting), and 18% to investors (also with a four-year vesting). A perpetual inflation rate of 4% per year commences after the four-year period. See “Uniswap Governance,” Uniswap Blog, accessed at https://uniswap.org/blog/uni/(last accessed January 30, 2024). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 and the subsequent reduction of the token supply. Both minting and burning mechanisms are executed through smart contracts. DAO tokens can be offered in both primary and secondary markets, depending on the stage of the DAO and token availability. In primary token markets, common methods for introducing tokens include Initial Coin Offerings (ICOs), resembling Initial Public Offerings (IPOs), where DAOs sell tokens to the public before launching their platforms, raising capital for development and marketing.40 Another method for introducing tokens to the market is through an Airdrop, where a DAO distributes its tokens for free to a selected group of recipients, usually based on some criteria or eligibility.41 This can help reward early adopters and contributors but may dilute the value of the tokens and create tax implications for the recipients. Additionally, DAOs often sell tokens to the public after launching its platform or service. This can also help raise funds for further development and expansion, as well as attract new investors. In secondary token markets, trading DAO tokens on decentralized exchanges (DEXs) and centralized exchanges (CEXs) provides investors with liquidity and facilitates easy token 40 For example, Uniswap conducted its ICO on September 1, 2020, but instead of selling tokens directly to the public at a fixed price, it distributed them for free to past users to reward their participation and loyalty. Each user who interacted with Uniswap before received 400 UNI tokens, valued at approximately $1,200 at the time of distribution (now around $2,400). In total, 150 million UNI tokens (15% of the total supply) were distributed, with the remaining tokens allocated to the founding team, investors, advisors, and future community treasury. Conversely, BitDAO's ICO on August 17, 2021, was a traditional Dutch auction, with token prices starting high and decreasing until all tokens were sold. The auction lasted 24 hours, attracted over 10,000 participants, and raised over $230 million from investors such as Peter Thiel, Founders Fund, Pantera Capital, and Dragonfly Capital. Each token was priced at $0.10, and 2.3 billion tokens (23% of the total supply) were sold during the ICO. The remaining tokens were allocated to the founding team, advisors, partners, and future community treasury as well. See Uniswap Protocol, Home, https://uniswap.org/ (accessed March 27, 2023); and Olga Kharif, “BitDAO Raises $230M for Decentralized Crypto Investment Fund,” CoinDesk, August 17, 2021, https://www.coindesk.com/business/2021/08/17/bitdao-raises-230m-for- decentralized-crypto-investment-fund/ (last accessed January 30, 2024). 41 Some examples of DAO token airdrops include Collab.Land, a tool that helps DAOs manage token-based access to their chats and content, which airdropped its governance token to more than 2 million eligible users that use the tool. Arbitrum, a scaling solution for Ethereum that reduces fees and transaction time, airdropped its token to users who have interacted with its network and transitioned to a DAO model. Among these, 1.1% of the total supply of Arbitrum tokens was airdropped to DAOs that operate in its ecosystem, such as Uniswap, SushiSwap, Compound, and others. See “Arbitrum Shows Just How Messy and Tricky Crypto Airdrops Can Be,” CoinDesk, https://www.coindesk.com/tech/2023/03/29/arbitrum-shows-just-how-messy-and-tricky- crypto-airdrops-can-be/ (last accessed December 27, 2023). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 transactions.42 Uniswap is the most widely utilized decentralized exchange for purchasing and trading DAO tokens, while PancakeSwap, another popular decentralized exchange, operates on the Binance Smart Chain. As of March 24, 2023, the 24-hour trading volumes of DAO tokens on Uniswap were notably high: $106,555,843 for Uniswap (17,865,026 tokens), $127,438,249 for LidoDAO (59,747,887 tokens), $3,794,828,823 for Arbitrum (2,659,335,666 tokens), and $108,701,058 for ApeCoin (26,656,949 tokens).43 The substantial trading volumes underscore the significant market liquidity of DAO tokens and indicate the growing interest and confidence in these decentralized organizations. III. Hypothesis of Low Voter Participation and the Underlying Theoretical Basis DAOs represent a compelling fusion of technological advancements and ideological aspirations. They are founded on voluntary relationships between individuals and groups within business organizations, but this organizational structure is not entirely novel in the realms of law and economics. Contrary to the expectations of DAO advocates, token holders may not actively engage in voting simply because they possess the right to do so.44 Theories by Mancur Olson on collective action and those by Easterbrook and Fischel on rational decision-making suggest that rational decision-making may deter individuals from contributing to voluntary business organizations. Large organizations, in particular, are prone to the free-rider problem, wherein individuals refrain from contributing their fair share of effort or resources towards shared goals.45 As a result, DAOs 42 Decentralized exchanges (DEXs) and centralized exchanges (CEXs) differ in the degree of control and custody they have over users’ funds and data. Centralized exchanges, like Coinbase, Binance, and Kraken, act as intermediaries between buyers and sellers, requiring users to deposit funds into the exchange’s wallets and entrusting the platform with trade execution, asset security, and personal information protection. In contrast, decentralized exchanges, such as Uniswap, Curve, and SushiSwap, allow users to trade crypto assets directly with each other without third-party intermediation, enabling users to maintain full control and custody over their funds and data. See TradeSanta. “What is a Decentralized Exchange.” https://tradesanta.com/blog/what- is-a-decentralized-exchange. (Last accessed December 29, 2023) 43 See CoinMarketCap, DAOs https://coinmarketcap.com/view/dao/(last accessed December 24, 2023), 44 The intuitive belief and idealized notion held by DAO advocates is that groups will invariably act in their self-interest, based on the rational and self-interested behavior of their constituents—a concept rooted in Adam Smith's theories concerning the benefits of individual self-interest for the broader community. However, the reality of this situation is far more nuanced than this simplified perspective suggests. 45 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups 5-52 (Harvard University Press 1971). Olson’s argument centers around the nature of collective goods, which cannot be exclusively consumed by one individual once provided by any group member. Consequently, individual members only reap a fraction of the benefit from any expenditure they incur in pursuit of the collective good. In any group or organization, regardless of its size, the overarching objective is to attain benefits that EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 are predisposed to low voter participation, especially among smaller stakeholders. When only a limited number of token holders participate in voting, it becomes inevitable that DAOs will fail to materialize as the decentralized decision-making organizations their advocates envision or promote. As a result, the centralized decision-making aspect of DAOs remains indistinguishable from that of traditional corporations, eliminating the need and justification for considering them as distinct legal entities. The hypothesis of low voter participation in DAOs is grounded in Mancur Olson’s collective action theory. This theory posits that when an organization grows so large that an individual’s contribution or decision-making influence has a negligible effect on others, rational individuals will opt not to participate. This results in the classic free-rider problem.46 Applying this theory to DAOs suggests that voter participation rates are expected to be quite low, particularly among small token holders. Consequently, decision-making power within DAOs tends to be concentrated in the among a select few large token holders, who are typically the founders, early investors, and major venture capitalists.47 Low voter participation can also result from individuals’ rational choice not to contribute due to the division of labor and expertise. Easterbrook and Fischel contended that shareholders’ non- participation in corporate management is rational.48 This viewpoint emphasizes the efficiency gained by assigning specialized individuals to specific tasks. They referred to this phenomenon as “shareholders’ rational ignorance,” which is grounded in a shareholder’s cost-benefit analysis.49 According to their theory, the primary economic virtue of the public corporation does not lie in its intrinsically serve the interests of all members involved. While all members share a common interest in securing these collective goods, they may not necessarily hold a mutual interest in bearing the associated costs required to provide such collective goods. Rather, each individual would often prefer that others bear the full extent of these costs yet continue to enjoy the benefits regardless of their personal contributions. Accordingly, the larger the group, the more it tends to fall short of providing an optimal level of a collective good. 46 Id. 47 Note that the theory of collective action holds true even when there is unanimous agreement in a group about the common good and the methods of achieving it. 48 Frank H. Easterbrook & Daniel R. Fischel, The Economic Structure of Corporate Law 55-78 (Harvard Univ. Press 1991). 49 DAO advocates argue that the non-hierarchical structure eliminates agency costs, but they overlook the fact that the costs associated with every token holder participating in corporate governance may deter token holders from investing their time and effort. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 capacity to pool vast amounts of capital.50 Rather, it is the hierarchical decision-making structure that makes it well-suited for running large business enterprises when shareholders lack expertise or the desire to acquire it.51 Consequently, shareholder voting should serve as a last resort for accountability, rather than the primary mode of decision-making.52 The principle of separating ownership and control in organizations predates many corporate governance features and thus merits recognition for its enduring relevance. 53 By adhering to the rational ignorance theory, most token holders will have little interest in acting as managers without proper compensation, and consequently, DAOs will experience low voter participation.54 The third reason underpinning the low voter participation hypothesis relates to the alignment of individual token holders with the management team’s ideas and practices, as well as their choice to exit the organization by selling their tokens when they disagree. Upon acquiring DAO tokens, individuals essentially endorse the existing management principles and operational methods of the business organization.55 As a result, they may be more inclined to preserve the status quo instead of initiating significant changes. Furthermore, as blockchain technology and businesses built upon it are still in their early stages with limited competition, major shifts or adjustments in operational 50 The development of modern corporate structures can be traced back to the late 16th century when European monarchs created chartered companies to pursue their dreams of imperial expansion.50 This structure allowed multiple investors to pool their resources and share ownership. Through the 18th and 19th centuries, this structure evolved to accommodate more sophisticated operations brought on by the industrial revolution and to address the need for professional management. Brian R. Cheffins, The History of Corporate Governance, in The Oxford Handbook of Corporate Law and Governance 46 (Jeffrey N. Gordon & Wolf-Georg Ringe eds., 1st ed. 2013). 51 Easterbrook & Fischel, supra note 42, at 56. See also L. Meade, Director Primacy and Shareholder Disempowerment, 1 J. Corp. L. & Ethics 117 (2022). 52 Id. Easterbrook and Fischel further emphasized that the individual apathy towards voting is a rational choice made by weighing the costs and benefits. The high costs associated with obtaining, processing, and evaluating information demand a substantial investment of time and resources that most individuals are unwilling or unable to make. As a result, individuals opt to abstain from actively evaluating specific proposals due to the high cost and futility of their efforts. 53 Walter Werner, Corporation Law in Search of Its Future, 81 COLUM. L. REV. 1611, 1637 (1981). 54 Easterbrook & Fischel, supra note 42, at 46-49. See also Kenneth J. Arrow, The Limits of Organization, 74 Cal. L. Rev. 1363 (1974). 55 In particular, many DAOs adopt a progressive decentralization model, where the founding team first establishes the DAO and then gradually distributes decision-making power through the use of tokens. Before transferring decision-making authority to the general public, these founding teams have already established basic governance rules and relatively successful products. Individual token holders join based on these reasons and are less willing to make changes. Moreover, the original founding teams generally hold a large number of tokens, so these small individual token holders are also willing to follow their decisions. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788

strategies may not be urgently needed. Therefore, token holders may favor passively retaining tokens as long-term investments and overlooking decisions that do not substantially affect the organization’s core operations. Moreover, should small token holders find themselves at odds with the decisions made by large token holders, they can opt to exit the DAO by “voting with their feet” through selling their tokens, rather than casting dissenting votes. Regardless of the individual or combined influence of these three theories, DAOs are likely to exhibit low voter participation. A considerable proportion of token holders will abstain from engaging in the decision-making process, rendering them incapable of expressing their viewpoints through voting. If empirical evidence corroborates this hypothesis of low voter participation, it would suggest that achieving a genuinely decentralized decision-making process is an elusive goal. As a result, DAOs should not be granted special recognition under corporate law and securities regulations. IV. Data Construction and DAO Governance Fundamentals: A Comprehensive Overview This empirical study was conducted by gathering and organizing all DAO voting activities.56 The dataset comprises both on-chain and off-chain voting data, gathered from Snapshot and Tally platforms, which are responsible for executing and recording these voting activities.57 To fill in gaps related to voting information and uncover patterns within the dataset, an in-depth manual examination of influential DAOs’ governance forums was conducted over an extended period. The observed patterns and insights will be discussed in the following section of this article. A. The Dataset Construction To ensure the reliability and comparability of the data analysis, 50 DAOs with 4,983 voting activities were selected for in-depth examination based on a comprehensive evaluation of three 56 It should be noted that since most DAOs require 2 to 3 rounds of voting for each individual voting proposal, some proposals may not progress to subsequent rounds due to an insufficient number of votes received. As a result, the number of voting activities should not be equated with voting proposals. There were 10,493 voting activities collected. 57 Snapshot and Tally are both software that offer voting services for DAOs and keep track of virtually all on- chain and off-chain voting activities. Specifically, Snapshot is used to conduct and record off-chain voting activities, while Tally is employed to conduct and record on-chain voting activities. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 critical factors: treasury size, number of token holders, and number of voting proposals. This approach enabled a comprehensive and representative evaluation of the current state of DAO governance. The data was analyzed using a set of variables and was cleaned through both manual and automated processes, such as the elimination of proposals that were simply betting on market trends and lacked implementation significance. For each voting activity, the following data was collected: (1) Proposal subject; (2) Total number of votes cast for the proposal; (3) Number and rate of voter participation; (4) Percentage of total votes in support of the proposal ; (5) Percentage of total votes against the proposal; (6) Final voting results; (7) The number of first, second, third, and fourth largest votes for or against a specific proposal; Subsequently, a manual evaluation was conducted to determine: (8) The influence of key voters on the voting result; (9) The consistency of key voters’ voting behavior in other voting activities; and (10) The identification of key proponents, influential voters, and their token ownership within various DAOs. B. Treasury Size, Token Holder Count, and Industry Distribution Out of the 50 DAOs selected for further investigation in this empirical study, 29 had a treasury ranging from $25 million to $100 million, 17 had a treasury from $100 million to $300 million, and 4 DAOs had a treasury size exceeding $300 million. Concurrently, the average number of token holders for DAOs with a treasury size between $25 and $100 million was 38,000, for those with a treasury size between $100 million and $300 million, the average number of token holders was 70,000, and for the 4 DAOs with a treasury size above $300 million, the average number of token holders was 116,000. A positive relationship between treasury size and token holder count was observed, suggesting that larger treasuries may attract more interest, investment, or participation. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Figure 1: DAOs categorized by treasury size and average token holder count The study of 50 DAOs shows that the DeFi sector is the most popular among DAOs, with 31 operating in this area. This is consistent with the broader trend in the crypto industry, indicating that DeFi is currently the most sought-after application of blockchain technology. The second most popular category of DAOs is blockchain infrastructure-focused, with 10 operating in this sector. This shows the increasing importance of building robust and scalable blockchain networks to support the growth of the entire ecosystem. NFT-related DAOs come in third with 5, demonstrating the increasing interest in digital art and collectibles. Lastly, gaming and social DAOs each account for two DAOs, indicating potential growth in these sectors. The findings highlight the versatility of DAOs, and the strong focus on DeFi suggests that decentralized finance is likely to continue as a driving force for the development of blockchain technology. Figure 2: DAOs categorized by industry EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 C. An Overview of DAO Voting Rules DAOs employ a variety of voting rules, including token-based quorum voting, majority voting, holographic consensus voting, conviction voting, multisig voting, delegation voting, and various combinations of these methods.58 The most widely used voting method is quorum majority voting, which mandates a minimum level of participation from DAO token holders to approve a proposal. Upon reaching the participation threshold, the proposal with the highest number of votes is deemed the winner. Conversely, if the threshold is not met, the proposal is considered to have failed. DAOs voting can also be categorized into two distinct types: off-chain voting and on-chain voting. Off-chain voting, also referred to as soft voting, is the process of casting votes outside of the blockchain network. This form of voting often takes place through Snapshot, a user-friendly voting interface that enables individuals to express their opinions and gauge sentiment towards specific proposals. On the other hand, on-chain voting, also known as hard voting, refers to the act of voting that is executed directly on the blockchain. Many DAOs utilize Tally, an on-chain voting tool, to facilitate on-chain voting. On-chain votes are recorded on the blockchain as transactions that are processed by “miners.”59 These “miners” are responsible for running, storing, and adding transactions to the blockchain, using their computational power to process votes and append them to the blockchain.60 Therefore, on-chain voting is achieved through the use of smart contract code, allowing for changes or transactions to be executed automatically without human intervention. As a result, on-chain voting can be costly, as it requires transaction fees, or “gas fees,” to compensate the miners for the energy and effort required to process transactions.61 58 Each DAO employs a unique voting mechanism, typically outlined in the governance section of its website. The selection of these voting rules is influenced by factors such as the DAO’s size, its operational and managerial philosophy, and the industry in which it operates. For an in-depth overview of specific voting mechanisms, please refer to Appendix 3. 59 Miners are participants in a blockchain network who use their computational power to validate transactions and create new blocks of data. 60 Ian Vanagas, Governance: On Chain or Off Chain? Upstream Blog (August 30, 2020), https://blog.upstreamapp.com/on-chain-and-off-chain-governance/ (last accessed January 30, 2024) 61 Gas fee for a voting activity may vary depending on the complexity of the voting contract and the data involved. A standard transaction on Ethereum has a gas limit of 21,000 units. The gas price for a voting is determined by the network congestion and the user’s preference of confirmation speed. For example, if one wants to vote on a proposal on a DAO on Ethereum, she gets the gas limit to 50,000 units and the gas price to 100 gwei (gwei is a subunit of ETH). The gas fee for her voting transaction would be Gas fee=Gas limit*Gas EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table 1: Comparison of Off-Chain and On-Chain Voting Voting Method Off-Chain Voting (Soft Voting) On-Chain Voting (Hard Voting) Definition A voting process that casting votes A voting process that executing outside of the blockchain network. voting directly on the blockchain Often through a third-party platform. using smart contracts. Cost Free, but may involve fees charged May incur gas fees for executing by the third-party platform for using transactions on the blockchain. their services. Platforms Snapshot Tally Example 1: Uniswap’s Voting Rules Uniswap, an Ethereum-based decentralized exchange platform for cryptocurrencies and DAO tokens, serves as a prominent example of voting rules employed by DAOs. Its voting process typically involves multiple phases, beginning with one or two rounds of off-chain voting to gauge sentiment regarding proposed changes. Proposals that pass the off-chain phase proceed to on-chain voting, where the proposal initiator must write code for a smart contract that can be written onto the blockchain. If the proposal secures enough votes during this phase, the smart contract is executed, and the change becomes irreversible on the blockchain. Uniswap, this Ethereum-based decentralized exchange platform, functioning as a cryptocurrency and DAO tokens exchange, operates under the DAO structure for governance and incorporates its own governance token, UNI. With the largest treasury size and most active token holders among DAOs, Uniswap’s voting rules are also highly representative of the DAO ecosystem. The voting process in Uniswap consists of three phases: (1) Proposal: UNI holders must post their proposals on the governance forum for a minimum of 7 days to allow the community to review, comment, and ask questions. (2) Temperature Check (soft voting): A 3-day snapshot vote requires 10 million UNI to move to the next phase. (3) On-chain Voting (hard voting): The final 7-day voting period on Tally requires a quorum of 40 million UNI to pass. price, which is 50,000*100 gwei fee = 0.005 ETH. At the current market price of ETS ($3,500), the gas fee for a vote would be equivalent to $17.5. see more at Gas and fees | ethereum.org EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 This updated voting rule system replaced the previous four-phase system, where a Consensus Check (soft voting) phase was eliminated, and the quorum for the Temperature Check phase was increased from 50,000 UNI to 10 million UNI.62 This change aims to simplify and streamline the governance system, as the previous process was considered too complex, resulting in low participation and engagement from the community. By merging the temperature check and consensus check into a single phase, the new process is expected to boost community engagement and expedite decision-making.63 Figure 3: Voting Rules of Uniswap Example 2: MakerDAO’s voting rules MakerDAO, one of the well-established and largest DAOs in the crypto space, governs the Maker Protocol, a decentralized platform that enables users to create and manage a stablecoin called 62 Uniswap Community Governance, Community Governance Process Update, Jan 2023, https://gov.uniswap.org/t/community-governance-process-update-jan-2023/19976 (last visited Feb 9, 2023). 63 Id. The specific voting rules for each DAO are delineated in their respective governance forums. When changes to voting rules are necessary, stages of voting must be conducted to approve them. These alterations are subsequently reflected through both on-chain and off-chain mechanisms. Off-chain rules resemble traditional voting rules within corporate structures, relying on human execution and supervision. Conversely, on-chain rules, such as hard voting quorums, are documented on the blockchain using smart contracts. Upon reaching both the quorum and majority vote requirements, a proposal is automatically enacted by the smart contract, eliminating the possibility of human interference. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 DAI.64 Unlike Uniswap, MakerDAO’s voting system employs a continuous approval voting mechanism, allowing voters to select any number of options without ranking them.65 The option with the most approval at any given time becomes the winner. The governance of MakerDAO is entrusted to its token holders, who vote on various aspects such as interest rate adjustments, collateral type additions, and governance system enhancements. The MakerDAO governance token is MKR. Off-chain voting occurs in the Maker Governance Forum, where community members can express their opinions on a range of subjects. MakerDAO offers two forms of on-chain voting: Polling and Executive.66 Polling is designed to gauge the sentiment of MKR holders, with votes weighted according to the amount of MKR each voter stakes. Polling votes serve as a guide for Executive voting. Proposals can be submitted either as signal requests or Maker Improvement Proposals (MIPs). For a Polling to pass, it must receive more yes votes than no votes, while an MIP poll requires at least 10,000 MKR for approval. Executive votes can be cast at any time, though they are typically posted on Fridays.67 These votes include technical changes supported by signal requests and governance polling. By adopting a continuous voting approach, a new executive becomes the official state of the protocol once it amasses sufficient votes. D. Voting Proposal Topics and Support Received DAOs across various industries typically initiate and vote on proposals that emphasize unique aspects of their protocols. These proposal topics not only embody the underlying business associated with a particular DAO, but also represent the management issues that token holders can engage in and are interested in addressing. By analyzing the subjects of voting proposals and the support they receive, one can determine which matters fall within the purview of token holders for 64A stablecoin is a type of cryptocurrency that tries to maintain a stable value by pegging its price to some external reference, such as a fiat currency (like the U.S. dollar), a commodity (like gold), or another cryptocurrency. See “What is a stablecoin?” on Coinbase. 65 MakerDAO Governance. MakerDAO. Available at: https://makerdao.com/en/governance/ (last accessed January 30, 2024). 66 MakerDAO Governance Portal, https://vote.makerdao.com/ (last accessed January 30, 2024) 67 Id. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 governance beyond existing DAO smart contracts. Furthermore, this analysis helps identify the particular issues that token holders care about and are willing to participate in for governance. While the overarching themes of voting proposals exhibit similarities across industries, the quantity and distribution of proposals within distinct categories may diverge. To elucidate industry-specific priorities, an examination of proposal topics from representative DAOs is presented as the follows. Topic of Voting Proposals in Different Sectors In the DeFi sector, major DAOs, including Uniswap and PancakeSwap as examples, show a variety of proposals that token holders participated in. These proposals can be categorized into several groups. The first category involves deploying protocols on multiple chains and broadening their presence on other platforms.68 Uniswap, for instance, has proposed launching on alternative chains like Arbitrum and Polygon, while PancakeSwap suggests adding support for new tokens and expanding to additional chains such as Binance Smart Chain. The second-largest group of proposals pertains to enhancing governance processes, including upgrading the governance contract to Governor Bravo, increasing proposal submission thresholds, and extending voting delays.69 Incentivizing liquidity and encouraging user participation within their ecosystems are also common themes in DeFi DAO proposals. Uniswap features proposals related to incentivizing liquidity across different chains and bolstering governance participation, while PancakeSwap puts forth proposals to incentivize liquidity in specific pools and establish new Syrup Pools for user engagement.70 Additionally, DeFi DAOs present proposals related to funding other organizations and initiatives within their communities. For example, Uniswap has proposals to create a Uniswap 68 Deploying protocols on multiple chains means that the underlying business can run on different blockchain networks, such as Ethereum, Binance Smart Chain, Polygon, etc. This allows DAOs to leverage the strengths and features of different blockchains. See https://ethereum.org/bridges (last accessed January 30, 2024) 69 Governor Bravo is a governance module used by some DAOs for more flexibility and functionality in governance, such as adding abstention as a voting option, allowing freeform comments with votes, and enabling governance parameters to be adjusted after deployment. See Understanding Governor Bravo. A review of key changes. https://www.tally.finance/post/understanding-governor-bravo 70 Uniswap employs an automated market maker (AMM) mechanism to facilitate trading without intermediaries, using liquidity pools locked in smart contracts containing two tokens, such as ETH and DAI. The price of each token is determined by the ratio of tokens in the pool. To expand its business, Uniswap attracts users to contribute tokens to swap pools by offering incentives such as UNI tokens, flexible fee tiers, and price ranges for different pool types. See https://uniswap.org/ (last accessed January 30, 2024) EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Foundation and fund organizations like the Nomic Foundation and Izumi Finance, while PancakeSwap proposes creating new Syrup Pools to support different tokens and fund charity initiatives. Finally, these DeFi DAOs also have miscellaneous proposals that fall outside of the aforementioned categories. These proposals can vary widely in their scope and impact on the platform, including retroactive airdrops or the removal of established funds or delegates. GnosisDAO serves as an exemplary case in the blockchain infrastructure category, offering a range of products such as Gnosis Safe, Cow Protocol, Conditional Tokens, Gnosis Auction, and Zodiac that provide essential support for blockchain activities. A thorough examination of GnosisDAO’s voting proposals reveals several categories, each of which pertains to distinct aspects of the DAO’s operations and development. The most prevalent category is funding and grants for projects and initiatives, with 13 proposals, followed by governance and treasury management, with 9 proposals. Ecosystem growth and partnerships and miscellaneous proposals, both have 7 proposals. Technical improvements and upgrades feature 5 proposals, while incentivizing participation and liquidity has 4 proposals. In the NFT sector, Rarible Protocol serves as an example of the NFT marketplace. The category with the most proposals in Rarible Protocol is user experience and platform features, with 16 proposals. These proposals aim to enhance user experience and introduce new features on the Rarible platform, such as creating white-label stores and improving methods to discover NFTs. The second-largest category encompasses miscellaneous proposals, with 15 proposals. The third category is dedicated to technical improvements and upgrades, featuring 13 proposals. Funding proposals, such as those designating budgets, and partnership proposals with other entities, each consist of 12 proposals. Representative DAO Proposal Topics Distribution and Voting Outcomes The findings presented in Table 2–6 illustrate the distribution of proposals across five selected DAOs, chosen based on substantial token holder participation, well-established governance forums, and a high number of submitted proposals. The selected DAOs encompass a diverse range of both DeFi and blockchain infrastructure, operating at varying scales. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 As can be observed from the tables, the pass rates for different DAOs are significantly distinct. The failure of DAO proposals is a complex issue that encompasses both power dynamics and participation rates. Minority proposals often encounter vetoes by a dominant token holder wielding a majority vote, while insufficient participation by token holders can result in failure to attain the quorum. For most DAOs, a quorum of 4% of the total tokens in circulation is mandated for proposals to pass, posing challenges for proposals without support from large token holders to succeed. Conversely, in some cases, a supermajority vote of 66.7% is necessitated for proposal approval, rendering proposals more difficult to pass.71 Table 2: Uniswap’s Proposals at Temperature Check Phase72 71 For example, BancorDAO employs a supermajority for its voting proposals to pass. 72 Core business decisions encompass decisions related to deployments on various blockchain networks or platforms, as well as transaction fee-related proposals. As of Nov. 3, 2023, there are 45 proposals in total in Uniswap’s Snapshot voting platform. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table 3: GnosisDAO’s Proposals73 Table 4: QiDAO’s Proposals74 73 GnosisDAO Builds Decentralized Infrastructure for the Ethereum Ecosystem. As of Nov. 3, 2023, there are 80 proposals in total in its Snapshot voting platform. 74 QiDao is an Overcollateralized Stablecoin Protocol that Allows Users to Mint Stablecoins (MAI) Against the Value of Their Decentralized Token Collaterals. As of Nov.3, 2023, there are 100 proposals in total in its Snapshot voting platform. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table 5: LidoDAO’s Proposals75 Table 6: BancorDAO’s Proposals76 V. Empirical Findings: Validating Hypotheses on DAO Voting Behavior This empirical study results align with the established theories of collective action in large businesses or voluntary organizations and the rational behavior of rational individuals. The empirical evidence confirms that the overall willingness of token holders to participate in voting is relatively low, especially for small token holders who cannot exercise decisive influence on the outcomes. The voting power within DAOs is firmly held by big token holders, with the vast 75 Lido is a DAO that manages the liquid staking protocols by deciding on key parameters, such as setting fees, assigning node operators and oracles, etc. As of Nov.3, 2023, there are 100 proposals in total in its Snapshot voting platform. 76 Bancor is a blockchain protocol that allows users to convert different virtual currency tokens directly. As of Nov.3, 2023, there are 423 proposals in total in its Snapshot voting platform. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 majority of voting activities being determined by a maximum of four large token holders. These large token holders tend to cast consistent votes and have a higher propensity to form coalitions within individual DAOs. Furthermore, given the nature of DAOs as a novel form of organization, a higher number of stockholder proposals are approved in DAO voting. This phenomenon can be partly attributed to the fact that DAOs are ideologically driven, as well as to the close connections that large token holders share with the broader crypto industry. Consequently, they are more inclined to approve proposals that contribute to the overall growth and development of the industry. A. Overall Low Voter Turnout Rates The voter turnout rates across the 50 DAOs and 4,936 voting activities substantiate the low level of voter participation hypothesis, as evidenced by the calculated summary statistics. The mean voter turnout rate is 1.77%, signifying that a small percentage of eligible voters participated in the decision-making process. The median turnout rate is even lower, standing at 0.10%, which indicates that over half of the proposals experienced an even lower proportion of voting engagement. The substantial standard deviation of 5.46% points to a significant variation in voter turnout rates among different voting proposals.77 The Pearson correlation coefficient (r) between voter turnout rate and the number of total token holders is -0.2217004, indicating a weak negative linear relationship between the two variables. This suggests that as the number of total token holders in a DAO increases the voter participation rate tends to slightly decrease. However, the relationship is weak, implying that other factors may also contribute to variations in voter participation rates across different DAOs. 77 The number of token holders is calculated based on the addresses holding DAO tokens as of November 3, 2023. As the number of addresses holding DAO tokens continually change, this data should be considered for reference purposes only. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table 7: Statistics of Voter Turnout Figure 4: Correlation between the Number of Token Holders and Voter Turnout Rate B. Concentration of Voting Power among Large Token Holders The empirical findings suggest that the governance structure within a DAO is predominantly concentrated among a select few token holders, primarily due to the low participation rates of smaller token holders. This observation further undermines the assertions made by DAO proponents that it functions as a decentralized decision-making entity. Empirical results demonstrate that, on average, the voter with the highest token holdings among participants could sway the outcome of a vote in 1,737 instances, accounting for 35% of total voting activities. Moreover, the collaboration of the top two token-holding voters among participants had the capacity to influence the outcome in 2,726 cases, adding 989 cases, which equates to 55% of all EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 activities. A trio of voters holding the highest, second-highest, and third-highest number of tokens were able to impact the results in 3,107 cases, adding 381 cases or 63% of all voting activities. Meanwhile, a quartet of voters (holding the highest, second-highest, third-highest, and fourth- highest number of tokens) possessed the power to determine the outcome in 3,266 cases, adding 159 cases. Collectively, 66% of voting outcomes could be dictated by a group of four or fewer voters. A further observation was noted, as the leading four voters exhibited a remarkable level of agreement in their voting patterns, with only slight variations in their views. In particular, the top four voters persistently voted in a unified manner for every proposal. In some DAOs, these same four voters consistently ranked among the top voters across all proposals, indicating that a small contingent of three to four individuals held substantial influence over the majority of decision outcomes within the DAO.78 Table 8: Distribution of Voting Power among Token Holders79 78 See the following analysis on token holders’ interest alignment. 79 In order to determine the number of cases where a small group of voters could determine the outcome of voting activities, a specific calculation method was employed. The method involved several steps, with the first being to divide the number of votes cast by the first voter in terms of vote volume by the total number of votes cast. If the result is greater than 0.5, which assumes that all proposals are subject to simple majority voting, then the count is 1. If the result of the first step is less than 0.5, the number of votes cast by the first and second voters are summed and divided by the total number of votes. If the result is greater than 0.5, then the count is also 1. This process is repeated for up to four voters, yielding the results of 1,373 for the first step, 989 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788

Table 7: Statistics of Voter Turnout Figure 4: Correlation between the Number of Token Holders and Voter Turnout Rate B. Concentration of Voting Power among Large Token Holders The empirical findings suggest that the governance structure within a DAO is predominantly concentrated among a select few token holders, primarily due to the low participation rates of smaller token holders. This observation further undermines the assertions made by DAO proponents that it functions as a decentralized decision-making entity. Empirical results demonstrate that, on average, the voter with the highest token holdings among participants could sway the outcome of a vote in 1,737 instances, accounting for 35% of total voting activities. Moreover, the collaboration of the top two token-holding voters among participants had the capacity to influence the outcome in 2,726 cases, adding 989 cases, which equates to 55% of all EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 activities. A trio of voters holding the highest, second-highest, and third-highest number of tokens were able to impact the results in 3,107 cases, adding 381 cases or 63% of all voting activities. Meanwhile, a quartet of voters (holding the highest, second-highest, third-highest, and fourth- highest number of tokens) possessed the power to determine the outcome in 3,266 cases, adding 159 cases. Collectively, 66% of voting outcomes could be dictated by a group of four or fewer voters. A further observation was noted, as the leading four voters exhibited a remarkable level of agreement in their voting patterns, with only slight variations in their views. In particular, the top four voters persistently voted in a unified manner for every proposal. In some DAOs, these same four voters consistently ranked among the top voters across all proposals, indicating that a small contingent of three to four individuals held substantial influence over the majority of decision outcomes within the DAO.78 Table 8: Distribution of Voting Power among Token Holders79 78 See the following analysis on token holders’ interest alignment. 79 In order to determine the number of cases where a small group of voters could determine the outcome of voting activities, a specific calculation method was employed. The method involved several steps, with the first being to divide the number of votes cast by the first voter in terms of vote volume by the total number of votes cast. If the result is greater than 0.5, which assumes that all proposals are subject to simple majority voting, then the count is 1. If the result of the first step is less than 0.5, the number of votes cast by the first and second voters are summed and divided by the total number of votes. If the result is greater than 0.5, then the count is also 1. This process is repeated for up to four voters, yielding the results of 1,373 for the first step, 989 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 C. Wide Disparity in Token Holder Interest Among Different Proposals The level of interest that token holders have towards various proposals is highly disparate. Within a single DAO, the number of voters can range from a minimum of 76 to over 20,000, indicating that not all token holders exhibit an equal degree of interest in every decision within a DAO, regardless of its size or impact. 80 Taking the following five DAOs as example—Uniswap, QiDAO, GnosisDAO, LidoDAO, and BancorDAO—we observed distinct similarities and differences in the types of proposals that garnered both the most and least voter participation. Uniswap mainly concentrated on deploying to various networks, fostering innovation, and adjusting fee structures. QiDAO’s voters demonstrated a primary interest in collaborations, incentives and rewards, treasury management, and the establishment of lending markets. GnosisDAO emphasized strategic initiatives, funding external projects, GNO tokenomics, and community programs. LidoDAO concentrated on external partnerships, funding allocation, governance and management proposals, and technical upgrades. BancorDAO voters gave priority to liquidity mining rewards, token whitelisting and co-investing. The sector in which each DAO operates influences the proposal types that attract voters. For example, Uniswap, a decentralized exchange, focuses on fee structures and network deployment, while LidoDAO, a staking solution provider, is more interested in technical upgrades.81 Despite these differences, collaboration and partnership themes are prevalent across all analyzed DAOs, highlighting their relevance to the broader DAO ecosystem. High voter participation in collaboration and partnership proposals can be particularly reflective of token holder preferences. First, these proposals generally have a greater impact on a platform’s for the second step, 381 for the third step, and 159 for the fourth step. For a more in-depth understanding of the calculation procedure, see appendix 1. 80 The WonderlandDAO saw a voter turnout of 76 for the proposal to remove the multisig feature and replace it with on-chain governance, while 22,509 voters participated in the vote to wind down the Wonderland DAO. 81 Lido is a liquid staking solution that allows users to stake their Ethereum tokens and receive stETH tokens in return, which represent the value of their staked ETH plus daily rewards. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 growth, reach, and functionality, making them attractive to voters focused on long-term success. Second, these proposals tend to be less technical and more strategic, which makes them more approachable and easier to comprehend for a diverse range of voters. This ease of understanding promotes participation, as voters may feel more confident voting on proposals that concern the platform’s wider direction as opposed to those dealing with specific, technical governance issues. While in examining the least voted proposals across the five DAOs, several common themes are identified that did not elicit high voter participation. These themes include technical and specialized proposals, governance-related adjustments, specific tasks and initiatives, treasury management and risk mitigation, and incentive programs and partnerships. The lower participation rates in these proposals may be attributed to factors such as complexity, limited impact, and lower stake. Voters tend to be more engaged in broader strategic decisions and proposals with more significant implications for the platform. In contrast, they may be less interested in voting on technical or specialized matters, which could be due to a lack of understanding or perceived relevance. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Figure 5: Voter numbers per proposal among different DAOs82 82 The X-axis represents the progression of time, with voting order indicated from left to right. The Y-axis represents the number of voters participating in voting, determined by the unique blockchain wallet addresses. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 D. Large Token Holders Dominate DAO Voting Activities through Active Participation It has been observed that large token holders wield a disproportionate amount of power in the decision-making process of DAOs, owing to their higher level of participation in voting activities compared to smaller token holders. This pattern parallels the situation in traditional corporate settings, where large shareholders similarly exhibit greater engagement in voting activities.83 The impact of their larger token holdings is compounded by their greater frequency of participation in voting processes, giving these large token holders significant influence over decisions within the DAO. This creates a power dynamic that disproportionately favors large token holders, raising questions about the fairness and democratic nature of the current voting mechanisms. Figure 6 illustrates the distribution of token holders involved in voting and their respective token holdings across 4,963 DAO voting events. The observed distribution is the skewness value of - 0.2832 indicates a significant left skew. This data suggests that a larger proportion of participating token holders are situated on the right side of the distribution, where the mode is greater than both the median and the mean. This observation leads to the conclusion that large token holders exhibit a higher level of participation in voting activities and wield more significant influence in the decision-making process within DAOs. Conversely, small token holders hold a relatively limited influence. This pattern is consistent with the prior findings regarding the concentration of power in the hands of a few token holders, suggesting that the governance structure of DAOs is not as decentralized as initially intended. 83 The same situation also occurs in the corporate context. According to studies, large shareholders tend to participate more than small shareholders in corporate voting for several reasons. First, large shareholders have more voting power and influence over the outcome of the vote, and therefore more incentives to vote. Second, large shareholders may have more information and resources to analyze the proposals and make informed decisions. Third, large shareholders may face fewer coordination problems or free-riding issues. See John Coates, What Determines Participation in Corporate Voting? Harv. L. Sch. F. on Corp. Governance (June 16, 2017), https://corpgov.law.harvard.edu/2017/06/16/what-determines-participation-in-corporate-voting/. Shareholder participation in corporate voting varies depending on the size and type of shareholders. Some studies find that large shareholders have higher turnout rates (88% versus 76%) and vote differently than small shareholders, especially on management proposals (17% versus 9%) and shareholder proposals (24% versus 15%). See Cvijanovic et al., 2012; Brav et al., 2022; Fisch and Schwartz, 2023. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Figure 6: Distribution of Token Holders’ Participation Rate in Relation to Votes Cast per Voting84 Figure 6 depicts the overall relationship between the number of tokens held by token holders and their participation rate in voting activities, utilizing the entire dataset of 50 DAOs. To reinforce the conclusions drawn, Figures 7, 8, and 9 individually analyze three specific DAOs, offering a more in-depth exploration of the connection between token holdings and voting participation within each DAO. In all three instances, the figures display a left-skewed pattern, indicating that in these particular DAOs, large token holders exhibit a greater degree of participation in voting activities and assume a more prominent role in the decision-making process. Figure 7: Uniswap’s distribution of token holders’ participation rate in relation to votes cast per voting 84 Figure 6 presents the data after applying a natural logarithmic transformation to the token holdings of voters. The transformed data reveals a minimum and maximum average number of votes cast by voters of 0 and 23.8, respectively, accompanied by a standard deviation of 3.528. To streamline analysis, the token holdings were divided into 12 equal bins, each representing a range of 0 to 24. The observed distribution of voters within each bin is illustrated in the figure, providing a detailed depiction of the distribution for the logarithmically transformed data. For detailed data and calculation procedures, please refer to the Appendix. Figures 7, 8, and 9 underwent similar processes. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Figure 8: LidoDAO’s distribution of token holders’ participation rate in relation to votes cast per voting EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Figure 9: BancorDAO’s distribution of token holders’ participation rate in relation to votes cast per voting E. A Deeper Look into Uniswap’s Voting Data: Low Voter Participation and Misaligned Interests This section provides an in-depth analysis of empirical findings derived from Uniswap’s voting data. It underscores the issue of low voter participation and occasional misalignment between the interests of large and small token holders. In certain instances, large token holders have single- handedly overturned the supermajority’s stance, thereby contradicting the consensus of the wider community. Uniswap’s Snapshot voting platform comprises a total of 79 proposals. To assess the voting intentions of individual participants, it is crucial to consider the number of voters without taking into account the number of tokens each voter casts. Votes that concur with the ultimate winning outcome are designated as “winning votes.” The winning percentage is calculated by dividing the number of winning votes by the total number of voters. A higher winning percentage signifies a stronger alignment of interests among the voters. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 The dataset encompasses 79 data points, ranging from a minimum value of 0.32% to a maximum value of 100%.85 When categorized into four bins, 11 data points reside within the low alignment range (0-25%), signifying limited consensus among voters. Conversely, only 3 data points are found in the moderate-low alignment range (25-50%) and another 3 data points in the moderate- high alignment range (50-75%), both indicating varying levels of concurrence among participants. A significant majority of proposals (62 out of 79) are situated in the high alignment range (75- 100%), showing that voters' interests are often closely aligned. This analysis indicates that, in the majority of cases, a strong consensus exists among voters, with just a few proposals displaying low or moderate alignment. With an average of 85.93% and a median of 98.11%, the data exhibits a positive skew, as numerous proposals demonstrate high alignment. The standard deviation of 24.28% suggests that the dispersion of data points is relatively large. A smaller proportion of proposals exhibit low or moderate alignment, emphasizing the infrequency of such instances. Figure 10: Interest Alignment Percentage of Uniswap's Voting Proposals Additionally, a total of 202,246 voting records were documented.86 By aggregating the voting records based on voting addresses, it was determined that 25,842 unique addresses (representing 85 See Appendix 3 for more detailed information. 86 See Appendix 4 for more detailed information. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 distinct voters) took part in the voting process.87 It is important to note that Uniswap has 359,300 token holders; thus, only about 7.92% of token holders ever engaged in the platform’s voting. Among the 25,842 engaged token holders, 29 voted more than 50 times, 4,039 voted between 10 and 50 times, and the remaining 21,774 voted fewer than 10 times. Out of the 79 voting events, the four largest voters reached unanimous agreement in the majority of the votes, occurring 55 times. This means that most of the time the large token holders’ interests are well-aligned. There were only 24 cases showed disagreement among the four largest voters. Within these cases, 12 instances involved just one token holder dissenting from the other three. Furthermore, there were 10 instances in which one of the largest voters went against the opinions of nearly all other voters, resulting in a winning percentage of single digits.88 This examination of Uniswap’s voting data reveals that although most proposals exhibit high alignment, the overall voter participation rate remains low. Instances where large token holders single-handedly influence the outcome, contradicting the wider community’s consensus, also raise concerns. 87A blockchain address is a unique identifier composed of a long string of alphanumeric characters. This address represents a voter’s connected blockchain wallet and is used to track their participation in the voting process. Each blockchain address typically corresponds to a unique token holder, enabling a clear representation of voting engagement among participants. 88 The proposals that resulted in single digit winning percentages were related to incentivizing measures and liquidity. These proposals include the Optimism-Uniswap LM Program: Update to Incentivized Pools, Managing Systemic Risk in Uniswap's Community Treasury using KPI Options, and Merge the Uniswap Grant Program into Uniswap DAO Governance. See Appendix 4 for more detailed information. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 VI. Moving Forward: Three Ways for Improving the DAO Voter Participation Based on the analysis above, DAOs have failed to achieve the democracy and efficiency for which they were advocated. To address this issue, three methods for improvement can be proposed: (1) Quadratic voting, (2) Small group decision-making mechanisms, and (3) Standardized information disclosure requirements and guidance. A. Creating Fluid Majorities: The Role of Quadratic Voting in Mobilizing the Latent Majority and Enhancing DAO Governance A man who tried to hold back a flood with a pail would probably be considered more of a crank than a saint, even by those he was trying to help.89 Such is the sentiment of a small token holder in a vast DAO, feeling that their individual vote is but a drop in the ocean. In situations with numerous participants, the typical participant recognizes that their individual efforts may not significantly impact the overall outcome. The first approach seeks to address the concerns of small token holders who may feel that their limited influence renders their time and effort incapable of effecting substantial change. To address this issue, the weight of voting power is adjusted by quadratic voting mechanism to mobilize the latent organization, thereby enabling any individual to potentially become a major voter on a single issue and influence the voting direction. This approach acknowledges the necessity for certain individuals within a large organization to possess substantial weight and significant decision-making authority. 90 However, unlike static hierarchies, the identities of these influential individuals are fluid, reflecting the “big voter” concept’s dynamic nature. Quadratic voting inherently introduces a system of variable voting strengths within the organization. 89 Thomas S. Kuhn, The Structure of Scientific Revolutions 2nd ed. (Univ. of Chicago Press 1970) 90 A group characterized by members with highly disparate degrees of interest in a collective good, and one that values a collective good (at some level of provision) far more than its cost, is more inclined to provide the good than other groups of equal size. The capacity of a group to act in its collective interest hinges on whether the individual actions of one or more members are perceptible to other group members. See Olson, collective goods. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Quadratic voting is a unique voting mechanism that seeks to provide proportional representation of individual preferences and give a voice to minority opinions. By granting token holders a specific number of voting credits that are proportional to their token holdings, which they can distribute among various proposals, and once cast, these voting credits cannot be recovered. This method also ensures that the cost of adding tokens to an option increases quadratically, thereby encouraging voters to critically evaluate the options and make informed decisions.91 In the context of DAO governance, one token could purchase a single vote credit for a specific issue, while 4 token could buy two vote credits, and 400 tokens would procure only 20 vote credits; and so on.92 This voting method ensures that minority opinions are given equal consideration and representation. In a conventional majority voting system, the majority often dominates and suppresses the voices and opinions of the minority. In contrast, quadratic voting gives minority opinions added significance, reducing the risk of them being overshadowed. Consequently, quadratic voting emerges as a more democratic and equitable approach to DAO governance, ensuring that all token holders, regardless of their stake, have an equal voice in decision-making and that minority interests remain safeguarded. Figure 11: Tokens and Their Vote Credits under Quadratic Voting 91 Casella, A, Macé, A. 2020. "Does Vote Trading Improve Welfare?" Annu. Rev. Econ. 13: Submitted. https://doi.org/10.1146/annureveconomics-081720-114422. 92 Posner & Weyl. supra note 41, at 107-110. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Table 9: Votes and Their Cost under Quadratic Voting93 Quadratic voting has gained attention from academics and policymakers alike and has been discussed in several publications. Eric Posner and Glen Weyl, in their book “Radical Markets: Uprooting Capitalism and Democracy for a Just Society,” provide a compelling argument for the rationality and superiority of quadratic voting.94 Quadratic voting has also gained a lot of attention and support from industry practitioners in recent years. One notable advocate is Vitalik Buterin, the founder of Ethereum, who has long been a vocal proponent of quadratic voting.95 In addition to the benefits of minority protection, quadratic voting can address other issues that plague traditional voting methods. One of the issues is vote-buying, which can happen when individuals or groups with greater resources buy up large numbers of voters to swing an election in their favor. Another issue that quadratic voting can address is the unequal distribution of power. The present study furnishes empirical evidence that an issue of unequal distribution of power is present in the current governance of DAOs. The purpose of this article is not to expound on the theoretical foundations of quadratic voting but to contribute to the ongoing discourse by presenting empirical evidence to substantiate the need 93 Id. 94 Eric A. Posner & Glen Weyl, Radical Markets: Uprooting Capitalism and Democracy for a Just Society 107- 125 (Princeton Univ. Press 2018). 95 Vitalik Buterin, "Voting 3.0: A Third Generation Voting System for DAOs," Vitalik Buterin's Blog, August 16, 2021, https://vitalik.ca/General/2021/08/16/voting3.html. For other scholars’ articles, see: Juan Cárdenas, César Camilo, Román Mantilla, David Zárate Purchasing Votes without Cash: Implementing Quadratic Voting Outside the Lab Posted: 2014. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 for its adoption in DAO governance. The recommendation for the use of quadratic voting is based on two crucial factors. A primary rationale for recommending quadratic voting in DAO governance lies in its parallel to a market for proposals. In market transactions, individuals convey the intensity of their preferences for goods and services by the amounts they’re willing to spend. Quadratic voting echoes this by assigning a value to each voter’s preference, enabling them to denote the relative importance of their choice. It’s this ability to capture preference intensity that many economists believe makes the price system inherently more efficient than a simple majority vote. 96 Considering the governance forum’s role as a market for proposals—where token holders “purchase” proposals that resonate with their managerial vision—quadratic voting stands out as the intuitive choice for DAO governance. Beyond merely indicating a preference for or against a proposal, it allows token holders to articulate the intensity of their support or opposition. Blockchain technology could streamline the integration of quadratic voting into DAO governance, facilitating the practical execution of backing or contesting proposals via voting credits. Consequently, quadratic voting has the potential to become a cornerstone of DAO governance, paving the way for enhanced collective decision-making. The utilization of quadratic voting is further justified by it is ability to incentives to proposal proponents to present higher quality proposals while also compensating them for the labor they exert. Behavioral psychology suggests that purely altruistic behavior is not sustainable in a marketplace, and a combination of good citizenship and incentives provide a more effective means of motivation.97 By allocating a proportion of voting credits to reward implementable proposals that receive a high level of support, quadratic voting achieves this objective. This approach avoids 96 “Witness, for example, the frequent citation of this theorem in a recent poll of economists asking whether an ideal voting system exists. IGM Forum, Primary Voting (March 7, 2016), http://www.igmchicago.org/surveys/primary-voting.” 97 Jean Tirole, Economics for the Common Good 166-183 (Princeton University Press 2nd ed. 2017). See also, Samuel Bowles, The Moral Economy: Why Good Incentives are No Substitute for Good Citizens, Yale University Press (2015). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 relying solely on economic incentives since short-term evaluations of proposal quality can be challenging, and it can be difficult to accurately estimate the corresponding economic return. Therefore, quadratic voting provides an effective incentive mechanism for token holders to devote time and effort towards creating high-quality proposals that benefit the DAO ecosystem. Secondly, the adoption of quadratic voting is more equitable as it provides better protection for minority interests by taking into account both the quantity and strength of their opinions. In doing so, it creates a more democratic and just system of decision-making. The findings of this study suggest that, on average, a small number of token holders (less than 4) could make major decisions for a given DAO. This calls into question the fairness and democratic nature of the current voting mechanism, which can be considered a thinly veiled form of elitism. The founders, Key Opinion Leaders, investment institutions, and early investors hold a disproportionate amount of control over decision-making processes due to their possession of a relatively larger number of tokens compared to the total number of tokens in circulation. This gives them a substantial numerical advantage when they choose to vote, leaving smaller token holders with the perception that their opinions and contributions are not valued. This, in turn, results in their reluctance to participate in the voting process, resulting in the formation of a silent majority. Quadratic voting holds the potential to address this issue by providing a means to reduce the disproportionate voting power of large token holders and enhance the representation of small token holders in the voting process. This is, as demonstrated by the empirical findings of this article, because small token holders are not always interested, understanding or have strong preferences for all proposals within the DAO. Quadratic voting can break the shackles of one person one vote, enabling small token holders to exert greater influence on issues they care about while sacrificing some influence on issues they care less about.98 Therefore, at least on some issues, small token holders have the power to compete with and have decision-making power over large token holders. The capacity to save up voting credits enables small token holders to concentrate their voting resources on the proposals of great significance to them, providing a way for their voices to be heard and their opinions to be taken into account. While large token holders may still hold 98 Eric A. Posner & Glen Weyl, supra note 46. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 significant power and influence over the decision-making process, the ability for small token holders to exert their influence on proposals that align with their personal ideology or interests serves to enhance democratic principles and foster increased voter participation. The implementation of quadratic voting can also help to address the issue of voter apathy by providing token holders with a greater sense of agency and control over the decision-making process. By allowing for more proportional representation of individual preferences and empowering minority opinions, quadratic voting can lead to a more equitable and just DAO governance system. B. Enhancing DAO Governance through Committee-Based Decision Making and Special Incentives for Proactive Participants Transforming a large decision-making group into smaller committees and subcommittees is an effective means of enhancing group decision-making efficiency. This approach can curtail the phenomenon of free riding, as theorized by Mancur Olson. In practice, many DAOs have adopted this approach as a direction for governance transformation by establishing decision-making committees and mandating that token holders delegate their voting rights to committee members. Alternatively, token holders may be restricted to voting solely on proposals presented by the committees. Such measures aim to enhance the effectiveness of DAO governance through targeted decision-making and greater participation by all stakeholders. Apart from committee-based decision-making, separate and selective incentives are also essential factors that can promote member participation and enhance DAO governance. By incorporating committee membership as part of separate and selective incentives, DAOs can foster greater participation and incentivize members to actively engage in the decision-making process. These incentives can play a significant role in promoting informed decision-making and ensuring that the organization benefits from the collective wisdom of all stakeholders. As previously discussed, rational individuals in large groups are unlikely to make any sacrifices to achieve shared objectives. Thus, there is accordingly no presumption that large groups will organize to act in their common interest. It is only in small groups or in cases where there is an independent source of selective incentives that groups are more likely to organize and act to EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788

significant power and influence over the decision-making process, the ability for small token holders to exert their influence on proposals that align with their personal ideology or interests serves to enhance democratic principles and foster increased voter participation. The implementation of quadratic voting can also help to address the issue of voter apathy by providing token holders with a greater sense of agency and control over the decision-making process. By allowing for more proportional representation of individual preferences and empowering minority opinions, quadratic voting can lead to a more equitable and just DAO governance system. B. Enhancing DAO Governance through Committee-Based Decision Making and Special Incentives for Proactive Participants Transforming a large decision-making group into smaller committees and subcommittees is an effective means of enhancing group decision-making efficiency. This approach can curtail the phenomenon of free riding, as theorized by Mancur Olson. In practice, many DAOs have adopted this approach as a direction for governance transformation by establishing decision-making committees and mandating that token holders delegate their voting rights to committee members. Alternatively, token holders may be restricted to voting solely on proposals presented by the committees. Such measures aim to enhance the effectiveness of DAO governance through targeted decision-making and greater participation by all stakeholders. Apart from committee-based decision-making, separate and selective incentives are also essential factors that can promote member participation and enhance DAO governance. By incorporating committee membership as part of separate and selective incentives, DAOs can foster greater participation and incentivize members to actively engage in the decision-making process. These incentives can play a significant role in promoting informed decision-making and ensuring that the organization benefits from the collective wisdom of all stakeholders. As previously discussed, rational individuals in large groups are unlikely to make any sacrifices to achieve shared objectives. Thus, there is accordingly no presumption that large groups will organize to act in their common interest. It is only in small groups or in cases where there is an independent source of selective incentives that groups are more likely to organize and act to EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 achieve their objectives.99 In line with this, John James has eloquently suggested that “committees should be small when you expect action and relatively large when you are looking for points of view.”100 When committees are kept small, they tend to be more agile and action-oriented, whereas larger committees tend to offer a broader range of perspectives and ideas. By leveraging these insights, DAOs can identify the most effective committee size for a given situation and tailor their decision-making mechanisms accordingly. Given the challenges associated with large-group decision-making, organizations often turn to small groups such as committees, sub-committees, and leadership groups to address these obstacles.101 These groups have demonstrated a crucial role in promoting effective decision- making and achieving desired outcomes. This tendency is reflected in several DAOs, where committee-based decision-making and targeted incentivization strategies have been adopted to enhance participation and promote collective action. An emerging governance strategy for DAOs is the transition to a small group model, characterized by the formation of committees with distinct levels of responsibility. Some DAOs are already embracing this approach, tapping into the inherent advantages of smaller groups, including heightened engagement and domain-specific expertise. However, to reconcile this committee- centric governance with the democratic and non-hierarchical ideals cherished by DAO enthusiasts, it’s imperative to foster a fluid committee system. This can be achieved by granting the broader community of token holders the periodic authority to elect committee representatives and by implementing robust oversight and member removal protocols. By integrating these measures, the small group governance model can resonate seamlessly with the foundational principles of DAOs. This need becomes even more evident when we consider the inherent challenges large organizations face in sustaining collective action. The very fact that a goal or purpose is common to a group means that no one is the group is excluded from the benefit or satisfaction brought about 99 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Harvard University Press, 1971), 166-167. 100 John James, “A Preliminary Study of the Size Determinant in Small Group Interaction,” American Sociological Review, XVI (August 1951), 474-477. 101 Id. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 by its achievement. And just as a state cannot support itself by voluntary contributions, or by selling its basic services on the market, neither can other large organizations support themselves without providing some sanction, or some attraction distinct from the public good itself, that will lead individuals to help bear the burdens of maintaining the organization. An individual’s own efforts will not have a noticeable effect on the situation of his organization, and he can enjoy any improvements brought about by others whether or not he has worked in support of his organization. Large organizations that are not able to make membership compulsory must also provide some noncollective goods in order to give potential members an inventive to join. Given the challenges associated with collective action in large organizations, incentivizing individual DAO members becomes imperative. Offering tangible rewards, such as compensation for gas fees or entrusting enhanced managerial duties, can mitigate the looming free-rider issue and spur members to immerse themselves in the decision-making process. This approach aligns personal ambitions with collective goals, potentially paving the way for more proficient governance and decision-making in DAOs. C. Standardized Information Disclosure Requirements and Guidance for Enhancing Voter Participation and DAO Governance Upon examining numerous DAO governance forums, it becomes apparent that proposals using sophisticated language and persuasive techniques, even when lacking detail, can successfully attract support. Furthermore, having one or two allies post concise yet enthusiastic comments emphasizing the proposal’s advantages can effectively sway the conversation in its favor. These methods allow for quickly steering the direction of the discourse. In case of opposition, it is beneficial to avoid immediate responses and engage in dialogues that support the proposal’s stance, which helps suppress dissenting voices and maintain focus on the positive aspects. If multiple parties contest the proposal, crafting extensive comments that downplay opposing views can be a valuable strategy. Regrettably, many individuals tend to equate the length of a response with the strength of the argument, as they lack the patience to thoroughly read and process lengthy text. While these tactics might be viewed as elementary public relations EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 strategies, they continue to be effective in DAO governance forums and can significantly influence and secure votes. Consequently, both those who disclose information and the commenters can potentially exert significant influence, manipulating the narrative within the DAO governance forum. Given this, the emphasis on information disclosure should not solely rest on deciding what to disclose but rather on devising strategies to effectively sift through information surges and pinpoint key details. In this context, clear guidance on information disclosure becomes paramount. To further this aim, establishing standardized requirements and guidance for information disclosure is a crucial approach to enhancing voter participation and improving DAO governance. By standardizing the disclosure process, individuals can more easily access and understand the information presented to them, enabling them to make informed decisions. Drawing inspiration from established regulations, the adoption of proxy rules akin to Section 14 of the Securities Exchange Act of 1934 could redress the challenges inherent in DAO governance forum proposals. Such rules are designed to bolster transparency and accountability by ensuring token holders are equipped with precise and pertinent information. This includes: (1) General disclosure provisions to provide token holders with the necessary and accurate information for informed decision-making; (2) A mandate for proponents to disclose any potential conflicts of interest; (3) An anti-fraud provision prohibiting the use of false or misleading statements; and (4) A fixed voting period, such as a designated day each month or quarter, to prevent token holders from forfeiting their right to vote on important decisions due to missed voting opportunities.102 Implementing proposal standards can significantly improve the quality of proposals and discussions within DAO governance forums. By setting minimum requirements, such as a 5,000- word length or mandating answers to a set of predetermined questions, low-quality proposals can be filtered out, and unproductive discussions and misinformation can be reduced. 102 The principal components of proxy rules were pointed out by Frank Easterbrook and Daniel Fischel. See Frank H. Easterbrook & Daniel R. Fischel, The Economic Structure of Corporate Law 82-83 (Harvard Univ. Press 1991). EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Moreover, the influence of large token holders and Key Opinion Leaders should be taken into account during this process, as they often hold considerable sway over the community with their opinions on proposals. To ensure that they act in the best interest of the DAO community and foster fair and transparent decision-making, it might be necessary to impose a fiduciary duty on these individuals. Such obligations could entail disclosing any conflicts of interest and requiring that recommendations and voting decisions be based on reliable and accurate information. To optimize the decision-making process within DAOs, it is crucial to ensure that the disclosure rules and systems in place exhibit appropriate biases, maintain a well-defined focus, provide sufficient but not extraneous data, and incorporate internal plurality.103 103 Downs, Anthony. An Economic Theory of Democracy. New York: Harper & Row, 1957, pp. 207-219. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 VII. Moving Forward: One Novel Regulatory Approach for DAOs Classification One of the most urgent legal quandaries currently surrounding DAOs is their classification. Should they be categorized as corporations, partnerships, or recognized as an entirely new form of business organization? Tied closely to this classification debate is the question of whether DAO tokens ought to be designated as securities. The diverse governance models adopted by the plethora of existing DAOs mean that the level of centralization can differ markedly from one DAO to another. Furthermore, even within a singular DAO, the depth of management and decision-making can evolve based on its progression. As such, it becomes both intricate and potentially unsuitable to bestow a static classification upon all DAOs in the context of business entities. Instead, these classifications should ideally be dynamic, adjusting based on the real-world governance of each DAO, with a particular focus on the degree of centralization in decision-making. As emphasized throughout this article, blockchain technology offers unparalleled advantages in terms of transparency, openness, and meticulous on-chain records. All data related to corporate governance, especially voting data, can be accessed and logged in real-time without incurring additional costs. This immediacy paves the way for a fluid entity classification approach.104 Legislators and regulators need not exhaustively analyze the characteristics of DAOs, especially when the concept of DAOs itself is rapidly evolving, to assign a fixed business classification. Instead, they can leverage the inherent transparency of blockchain data to provide newly established DAOs with a probationary period. For instance, if a DAO achieves a voting participation rate of 30% within its first six months of establishment, it could qualify as a special business entity. This would allow it to avoid the pitfalls of double taxation and the more stringent regulatory requirements found in securities regulation. Conversely, a DAO that fails to sufficiently democratize its decision-making processes during its initial phase essentially mirrors traditional corporations and should not benefit from any special classification considerations. 104 Drawing parallels from corporate law, the proposed fluid classification for DAOs bears semblance to the “hybrid entity” or “check-the-box” regulations. In such frameworks, entities, given specific conditions, possess the discretion to determine their tax classification — be it as a corporation, partnership, or a disregarded entity. This analogous regulatory malleability, which adjusts in accordance with the distinct nature and attributes of the entity, might offer insightful precedents for the classification methodology being advanced for DAOs herein. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 In subsequent governance, DAOs could annually report their voting participation rates and other pertinent voting data to determine if decision-making power is highly concentrated among a few individuals. DAOs exhibiting such concentrated power would forfeit the privileges associated with the special business classification. On the other hand, DAOs initially classified as corporations can showcase their progression towards decentralization through their annual voting participation rates. If they meet specific criteria, they could transition to being recognized as special business entities, thereby sidestepping issues like double taxation and additional regulatory constraints. Another crucial aspect to consider is why a fluid classification for DAOs is preferable to a straightforward categorization as either a corporation or a partnership. The answer lies in the intrinsic characteristics of blockchain businesses. Given that the vast majority of DAO management, communication, and execution occur on the blockchain, DAOs are not strictly bound or influenced by geographical constraints. This flexibility allows DAOs to opt for registration in jurisdictions like the United States, Switzerland, or even a Caribbean Island. With this mobility comes the potential for tax collection leakage as DAOs can strategically shift their registration to benefit from tax havens. Recognizing DAOs as novel avenues for economic growth, and considering their inherent detachment from strict national affiliations, there's a compelling case for granting them additional favorable legislative considerations. However, this preferential treatment must be balanced. Highly centralized DAOs might merely be traditional corporations seeking to operate under the guise of a DAO. It's imperative to ensure that the DAO structure doesn't become a conduit for corporations to sidestep tax obligations and securities regulations. In this light, the proposed probationary business classification, grounded in voting data, emerges as a fitting regulatory response to the unique challenges posed by DAOs. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 VIII. Conclusion This empirical study confirms the hypothesis that DAOs cannot rely solely on a substantial number of voters to voluntarily contribute through voting. Merely offering token holders the opportunity and conditions to vote does not ensure their actual participation in the voting process. As a result, decentralized decision-making may be an unattainable concept. The voting results from existing DAOs lend support to this hypothesis. Upon analyzing 50 DAOs, a discernible pattern of low token holder engagement in voting emerged. For instance, a mere 7.92% of Uniswap token holders have ever participated in voting. DAO decision-making power is concentrated among a few large token holders, with one large voter capable of swaying 35% of all voting outcomes. In total, 66% of voting outcomes could be influenced by four or fewer voters. The concentration of power is intensified by large token holders exhibiting greater enthusiasm compared to small token holders, resulting in higher levels of participation in the voting process and, subsequently, increased decision-making power. In the majority of cases, these large token holders maintained consistent positions, causing some instances where voting results did not align with the preferences of the small token holders, who constituted the supermajority in terms of voter count, but not in token quantity. This convergence of factors further diminished the influence of small token holders’ decision-making within DAOs, leading to a decline in their motivation to participate. In conclusion, decentralized decision-making, characterized by high voter participation, serves as both the primary purpose of a DAO’s existence and its most critical distinguishing feature from traditional corporations. If token holders of DAOs do not participate in the voting process—either because their input carries little weight or they choose to side with management—genuine decentralized decision-making remains elusive. Recognizing this challenge, this paper has proposed three potential solutions: the implementation of quadratic voting to empower minority voices, the restructuring of DAOs into smaller decision-making committees for increased efficiency, and the standardization of information disclosure to enhance transparency and trustworthiness in the decision-making process. Each of these methods offers a pathway to bolster EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 voter participation, thereby reinforcing the decentralized essence of DAOs. The proposed “probationary business classification,” grounded in voting data, offers a potential regulatory solution to address the classification challenges posed by varying levels of decentralization in DAOs. However, until voter participation consistently meets the stipulated criteria, DAOs should neither be granted special considerations as innovative entities nor receive preferential treatment from a legal standpoint. EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 Appendix: 1. Detailed Calculation for Table 7 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 2. Detailed Calculation for Figure 5,6,7 and 8 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 3. Uniswap’s Voting Data for Token Holders’ Interest Alignment Analysis Proposal Total Votes Winning Votes Winning Vote Percentage Largest 4 Voters Position Optimism-Uniswap LM Program: Update to Incentivized Pools 5563.00 18.00 0.32%Minority(1) Temperature Check - Should Uniswap incentivize Liquidity on Optimism and Arbitrum? 232.00 3.00 1.29%Minority(1) Temperature Check: Create Accoutability Committee 8397.00 186.00 2.22%Minority(1) Managing Systemic Risk in Uniswap's Community Treasury using KPI Options 201.00 5.00 2.49%Minority(1) Temperature Check - [Merge the Uniswap Grant Program into Uniswap DAO Governance] 136.00 6.00 4.41%Minority(1) Should the DEF move its assets to a custodial wallet viewable on chain? 252.00 33.00 13.10%Minority(4 ) Consensus Check - ABOLISH Delegates and change the UNI Governance Voting System 471.00 69.00 14.65%Minority(1) Funding the Fee Switch Activation 335.00 56.00 16.72%Minority(2) Temperature Check—Support izumi Finance as Uniswap V3 official partner of Liquidity Mining Platform 253.00 46.00 18.18%Minority(1) Optimism-Uniswap Protocol LM Program: Pool selection 3026.00 551.00 18.21%Minority(1) Optimism-Uniswap Protocol LM Program: Structure 3071.00 630.00 20.51%Minority(1) Optimism-Uniswap LM Program: Update to LM Managers 6095.00 2353.00 38.61%Minority(1) Retroactive Airdrop For Proxy Contract Users (e.g. 1Inch, Dharma, Matcha, etc.) 40.00 18.00 45.00%Minority [Temperature Check] Which bridge should Uniswap v3 use for cross-chain governance messaging between8 E4t3h8e.r0e0um and BNB Ch4a7in5?2.00 56.32%Majority (3) Minority(1) Donate $50M worth of UNI to Chris Blec to leave DeFi for good 516.00 291.00 56.40%Majority Funding a Political Defense of DeFi 462.00 301.00 65.15%Majority (3) Minority(1) [Proposal] Removal of established cranial follicles fund. 523.00 403.00 77.06%Majority (3) Minority(1) Temperature Check - Deploy Uniswap V3 on Harmony through an Additional Use Grant to Hermes DeFi 275.00 230.00 83.64%Majority (2) Minority(2) Begin Uniswap Liquidity Program v0.1 276.00 235.00 85.14%Majority Consensus Check — Funding a Political Defense of DeFi 413.00 354.00 85.71%Majority (3) Minority(1) Temp Check: Larger Grant Construct // CEA + No Negative Net UNI 464.00 406.00 87.50%Majority (3) Minority(1) Temperature Check: Raise the proposal quorum threshold 369.00 327.00 88.62%Majority Create a CAPITAL PROTECTION Tool for v3!! 350.00 325.00 92.86%Majority (3) Minority(1) Consensus Check - Should Uniswap Provide Voltz with v3 Additional Use Grant 362.00 337.00 93.09%Majority Should Uniswap lower the proposal submission threshold? 360.00 337.00 93.61%Majority Temperature Check - Allocate 1.5M UNI to Go-Ethereum long-term Talent-Acquisition & Retention Grant (G51E0T.H00-TARG) 484.00 94.90%Majority (3) Minority(1) [Temperature Check] Post-BSL cross-chain deployment process and creation of new uniswap.eth subdomai7n936.00 7623.00 96.06%Majority [Temperature Check] Should Uniswap v3 be deployed to BNB Chain? 6495.00 6265.00 96.46%Majority (3) Minority(1) [Temperature Check] Deploy Uniswap V3 to Boba Network 3754.00 3630.00 96.70%Majority Is uniswap supported to accelerate layer2? 467.00 453.00 97.00%Majority (3) Minority(1) Consensus check - Should Uniswap incentivize liquidity on Optimism and Arbitrum? 275.00 267.00 97.09%Majority Uniswap DAO can help 80K UNI to Turkey to relief after the big earthquake disaster 6800.00 6610.00 97.21%Majority Temperature Check - Managing Systemic Risk in Uniswap’s Community Treasury using KPI Options 408.00 397.00 97.30%Majority Temperature Check - Should Uniswap incentivize liquidity on Arbitrum and Optimism? 300.00 292.00 97.33%Majority [Consensus Check] Deploy Uniswap V3 to Boba Network 3514.00 3425.00 97.47%Majority (3) Minority(1) Can UGP increase their Q2 budget to $1.5M? 570.00 557.00 97.72%Majority [Temperature Check] Deploy Uniswap v3 on Avalanche 6413.00 6270.00 97.77%Majority Consensus Check - Should Uniswap v3 be deployed to Polygon? 379.00 371.00 97.89%Majority Consensus Check - Add 1 Basis Point Fee Tier 389.00 381.00 97.94%Majority [Temperature Check] Enable 1bp Fee Tier for UniswapV3 on Arbitrum 9707.00 9508.00 97.95%Majority [Consensus Check] Deploy Uniswap V3 to zkSync 3807.00 3735.00 98.11%Majority Temperature Check: Upgrade Governance Contract to Governor Bravo 296.00 291.00 98.31%Majority [Temperature Check] Deploy Uniswap V3 on Gnosis Chain 6702.00 6594.00 98.39%Majority [Consensus Check] Fix the Cross Chain Messaging Bridge on Arbitrum 6743.00 6636.00 98.41%Majority Consensus Check: Community-Enabled Analytics via Yield-Generating Investment Strategies 330.00 325.00 98.48%Majority [Temperature Check] Deploy Uniswap v3 onto Kava EVM Network 4934.00 4863.00 98.56%Majority Uniswap V3 Launch on zkEVM 9445.00 9318.00 98.66%Majority Consensus check- Uniswap DAO can help 80K UNI to Turkey to relief after the big earthquake disaster 7259.00 7164.00 98.69%Majority [Temperature Check] Deploy Uniswap V3 on Scroll testnet 15219.00 15022.00 98.71%Majority (3) Minority(1) [Temperature Check] Fix the Cross Chain Messaging Bridge on Arbitrum 5238.00 5171.00 98.72%Majority Community Governance Process Changes 8603.00 8498.00 98.78%Majority Should we deploy Uniswap v3 on Aurora? 3045.00 3010.00 98.85%Majority Turning on the fee switch for V2 354.00 350.00 98.87%Majority Consensus Check - Create Uniswap Volume KPI Options to help divest Uniswap’s Community Treasury into 1S8ta1b.0le0coins 179.00 98.90%Majority Should Uniswap governance contribute funding to the Nomic Foundation? 851.00 844.00 99.18%Majority [Consensus Check] - Enable 1bp Fee Tier on UniswapV3 Optimism(L2) 1315.00 1305.00 99.24%Majority Should UGP be reinstated with existing funds? 533.00 529.00 99.25%Majority Temp Check: "Fee switch" activation for ETH/USDC & USDC/USDT 1744.00 1731.00 99.25%Majority [Consensus Check] Create the Uniswap Foundation 2199.00 2184.00 99.32%Majority [Temperature Check] Deploy Uniswap V3 to zkSync 2457.00 2441.00 99.35%Majority Consensus Check - Should Uniswap governance contribute funding to the Nomic Foundation? 1414.00 1405.00 99.36%Majority [Temperature Check] Create the Uniswap Foundation 1860.00 1849.00 99.41%Majority Temperature Check - Should Uniswap Provide Voltz with v3 Additional Use Grant? 354.00 352.00 99.44%Majority Should we deploy Uniswap V3 on Celo? 914.00 909.00 99.45%Majority [Consensus Check] "Fee switch" Pilot for Three Pairs 2373.00 2361.00 99.49%Majority Should we deploy Uniswap V3 on AVAX? 228.00 227.00 99.56%Majority Temp Check - Should Uniswap add the 1bp fee tier to Polygon? 1031.00 1027.00 99.61%Majority [Temperature Check] Deploy Uniswap V3 on Moonbeam 1301.00 1296.00 99.62%Majority [Consensus Check] Should we deploy Uniswap V3 on Celo? 1391.00 1386.00 99.64%Majority (3) Minority(1) Should Uniswap v3 be deployed to Polygon? 300.00 299.00 99.67%Majority Consensus Check: Upgrade Governance Contract to Governor Bravo 605.00 603.00 99.67%Majority [Temperature Check] Should the Uniswap community participate in the Protocol Guild Pilot? 1139.00 1136.00 99.74%Majority Consensus Check - Should Uniswap add the 1bp fee tier to Polygon? 1338.00 1335.00 99.78%Majority Temperature Check - Should Uniswap v3 be deployed to Gnosis Chain? 909.00 907.00 99.78%Majority Should Uniswap provide Rage Trade with an additional use grant? 3144.00 3140.00 99.87%Majority [Consensus Check] Should Uniswap v3 be deployed to Gnosis Chain? 1697.00 1695.00 99.88%Majority [Consensus Check] Deploy Uniswap V3 on Moonbeam 1477.00 1476.00 99.93%Majority Uniswap Temperature check - [Should we increase the votingDelay?] 6.00 6.00 100.00%Majority Deploy Uniswap v3 on Arbitrum 871.00 871.00 100.00%Majority EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788 4. Uniswap’s Voting Data for Token Holders’ Interest Alignment Analysis EElleeccttrroonniicc ccooppyy aavvaaiillaabbllee aatt:: hhttttppss::////ssssrrnn..ccoomm//aabbssttrraacctt==44444411117788