DAOs can reshape society by bringing democratic control to important economic and social activities previously governed through more hierarchical means. But as many people have pointed out, DAOs have not yet achieved their goal of true decentralization.
Participation in DAO governance is too low, and tokenholders often lack the expertise and time to make informed decisions. Many DAO decisions are taken by a small number of actors, and we lack formal mechanisms to ensure these decisions are aligned with the interests of the tokenholders at large.
As it turns out, these challenges are identical to the governance challenges faced by democratic societies in the physical world. Lessons we have learned from two millennia studying and designing democratic governments provide useful insights into improving DAO governance.
As a professor of political science and political economy with a particular interest in DAOs, I have some ideas for how they can improve their governance structures by looking to antecedents in the physical world.
Make Voting Simpler
While there is a lot of variation in how different DAOs handle proposals and votes, they often ask tokenholders to vote on many specific decisions the DAO makes.
But referendum-based voting systems like this rarely succeed, because we cannot expect voters to devote many hours each week to studying issues and casting votes.
Asking voters to vote on a small set of the most important decisions can increase participation. While this may seem less democratic, it focuses voters’ attention and therefore increases their influence over the most consequential decisions the DAO makes.
Build More Robust Systems of Delegation
Faced with the infeasibility of government by referendum, democracies have evolved representative systems of governance. More and more DAOs are now building nascent systems of delegation, too.
To help voters find and retain good delegates and fire bad ones, democracies have developed a number of tools: news media ecosystems to monitor and report on politics; party-based campaigns to explain issues to voters; and transparency and disclosure requirements to mitigate conflicts of interest and foster accountability.
For DAOs, a great first step is employing tools that track what delegates do . Yetinattentive voters need analysis, not just information. Coverage, analysis, and endorsements of delegates by trusted figures — akin to news coverage and interest groups in the physical world — will help DAO voters more than raw information can.
Make Delegates Stand for Reelection at Defined Intervals
Today, DAOs that use delegation typically allow voters to pledge their tokens to delegates continuously, reallocating them whenever they want.
But democracies most often rely on periodic elections at fixed or otherwise predictable intervals. Holding elections at special moments coordinates voters’ attention, rather than expecting them to continuously stay up to date on how their delegates are behaving. And it gives representatives predictable milestones to shape their behavior.
DAOs should experiment with periodic elections and special “Election Days” that serve as focal points for voters to discuss, debate, and vote together. They should also experiment with different term lengths, compensation structures, and term limits for delegates, as in the physical world.
To align incentives, DAOs often try to subject the recommendations of their workforce experts to direct votes by tokenholders. But tokenholders are unlikely to be able to pay sufficient attention or participate enough to make informed decisions about the abstruse technical decisions that face DAOs on a daily basis.
Instead, acting as a council, delegates could appoint managers, like ministers in parliamentary systems, who could take certain operational decisions more expertly than tokenholder voting could. Though unelected, these managers would still be accountable to tokenholders because they can be fired by the elected delegates at any time. This would bear similarities to corporate governance, as well.
A “constitution” could help to spell out what managers can do unilaterally (e.g., hiring and firing), what must require approval by the delegates (e.g., passing a budget), and when and how the delegates can remove them.
Future Research on DAO Governance
DAOs offer a unique opportunity for us to democratize the way we work and the way we organize our activities and our resources. To succeed in this lofty endeavor, we must continue to innovate and experiment with different ways to organize the collective DAO decision-making process.
Democratic governments have been working on this same challenge for more than a thousand years, and while they have fallen short in some ways, they offer invaluable lessons that can help us to build a brighter future.
Andrew Hall is a Professor of Political Science at Stanford University and a Professor of Political Economy (by courtesy) at the Stanford Graduate School of Business. He serves as an advisor on democratic governance issues at Meta, Inc. For advice and support with this research, Hall thanks the DAO Research Collective.