Abstract: systematically document governance centralization in Decentralized Autonomous Organizations (DAOs) and examine its drivers and economic implications. Using multiple data sources and granular onchain transaction data, we find that DAOs –including most Decentralized Finance (DeFi) projects – exhibit low participation rates and highly concentrated voting power, with the top decile of voters controlling 76.2% of total votes. Further analysis reveals significant abnormal trading prior to proposal creation: blockvoters accumulate tokens to influence outcomes, while proposal managers engage in insider trading, earning average market-adjusted returns of 9.5%. These conflicts of interest can be value-destructive, particularly during crises when effective governance is critical to organizational survival. Yet, recent innovations such as quadratic voting and delegation mechanisms show promise in advancing the goals of decentralization by empowering minority token holders and mitigating agency conflicts. Overall, our study highlights both the persistence of agency problems in DAOs and the potential for well-designed governance mechanisms to improve outcomes in these emerging digital organizations.
Presentation: AFA 2026 (scheduled); HARC 2026 (scheduled)*; 2025 Global AI Finance Research Conference (scheduled); 6th Annual RCF-ECGI Conference (scheduled); FMA 2025 (scheduled); ABR-Fudan Joint Conference; 2025 CEBRA Annual Meeting*; 2025 SAIF Annual Research Conference*; 5th Annual CBER Conference; 18th International Behavioural Finance Conference; 4th Hong Kong Conference for Fintech, AI, and Big Data in Business; 5th Machine Lawyering Conference; 2025 Dishui Lake International Conference in Finance; ABFER 2025; CFMR 2025*; 2025 Forensic Finance Conference*; Digital Economy and Financial Technology (DEFT) Lab Meeting; Harvard Business School Seminar*; 2024 Tokenomics Conference, Waseda Workshop in Decentralized Finance; 2024 Singapore Scholars Symposium; NUS Accounting Brownbag
Revise & Resubmit at Research Policy
Abstract: This study examines how team reputation affects both the funding success and the subsequent innovation progress in early-stage crowdfunded projects, with a comparative focus on blockchain-based and conventional crowdfunding platforms. Using development progress data for blockchain projects on GitHub, we find that while teams with stronger reputations raise more funds, their subsequent innovation productivity, measured by development activity, tends to lag. Further analysis reveals an association between higher team reputation and increased turnover of tokens post-fundraising, suggesting team members may strategically exercise exit options soon after securing financing. In contrast, analysis of Kickstarter campaigns reveals a more consistent alignment between team reputation, funding success, and project delivery. We attribute this contrast to differences in platform governance; Kickstarter's structured oversight appears to mitigate early team disengagement and enhance project delivery compared to the blockchain context. These findings underscore the importance of effective governance mechanisms in crowdfunding platforms to ensure that team reputation signals translate into sustained innovation outcomes.
Abstract: This paper examines how open source innovation influences entrepreneurial activity. Using textual analysis and machine learning techniques, I classify open source software (OSS) released between 2008 and 2020 on GitHub into 226 software industries defined by Crunchbase. The analysis shows that an increase in OSS releases within a market leads to more startup entries and greater investment in the following year. The effects are particularly pronounced for non-forked, highly-starred, organization-created, and permissively licensed repositories, as well as for ventures whose founding team members are active on GitHub. To establish causality, I exploit inconsistent access to GitHub in mainland China and find that startup formation and financing decline relative to other countries when access is restricted. These findings suggest that open source innovation reduces entry costs by expanding access to technological resources, generating positive externalities and stimulating entrepreneurship.