Issue #70: Governance Experiments

Despite growing acknowledgement that simple token-based governance models have (arguably) critical flaws, the amount of experimentation in the governance realm is low. I have certain assumptions about why that I’ll get into, but the key observation is that more experimentation is needed if decentralized governance is going to be sustainable.

Early experiments at improving efficiency involved vote/decision delegation, and those have largely been successful. Nevertheless, most projects still blindly cling to a straightforward token model.

A second wave of experimentation seems to be emerging around NFTs. I think this is the right direction, and opens up a universe of new models and incentives that I hope will inspire the next generation of projects.

Let’s zoom in…

Delegation

The first wave of experimentation in decentralized governance was delegation.

Delegation is the assignment of authority to another person or group to carry out certain activities. It is often more efficient for constituents to assign their voting power to another person better suited to perform the activities. Most constituents don’t want to be, nor are they in the best position to, make frequent and important decisions. They don’t have the time, and they aren’t well informed.

The act of assignment can be inherent to a governance system, or at the discretion of the individual voter. We see both forms of delegation in Web3:

Inherent. This form has roots in traditional corporate structure, where authority over daily decisions is delegated to management, which are then further delegated to distinct operational units. In 2021, YFI tokenholders approved an operational structure that resembled traditional corporate governance, where authority over core operations were delegated to small, specialized operational units (yTeams).

Discretionary. This form has roots in representative democracies. Ethereum Name Service and Gitcoin tokenholders have the option of delegating their voting power (1 token = 1 vote) to a delegate they believe to be informed and will vote in the best interest of the project.

By and large delegation has been a successful at improving efficiency, and I’m hopeful we’re on on the cusp of a second, more radical phase of experimentation involving NFTs.

Bicameral Governance

When Optimism announced the OP token in late April, they also unveiled plans for a novel bicameral governance system. The Optimism “community” will be represented by two distinct bodies:

• Token House - comprised of OP token holders. Token holders will be able to vote on the distribution of project incentives as a part of a Governance Fund, protocol upgrades and more, akin to other governance tokens.

• Citizens’ House - comprised of holders of non-transferable or “soul-bound” NFTs. The process for distributing “citizenship” will be determined by the Optimism Foundation with input from the Token House. Citizens will facilitate and govern a process to distribute retroactive public goods funding.

Let me just first say…it’s about damn time!

We’ve been experimenting with decentralized governance for 7-ish years and this is the first example of a governance framework I’m aware of incorporating someone’s reputation (i.e. non-transferrable credentials).

Two reactions to the Optimism plan:

• Citizens should have more responsibility. I know this is an experiment, but I would like to see citizens assume a larger role in governance. Why are we limiting what I assume will be the most thoughtful and knowledgeable voices in the community to retroactive public goods funding? I’ll get into this later when I talk about governance debt, but my hunch is Optimism needs to (a) justify the existence of the OP token and is therefore (b) unwilling to remove the bulk of governance decisions from tokenholders’ plates.

• Citizenship requirements. I’m particularly interested to see the requirements for citizenship. There’s a lot of design space here. If I was in charge I would start conservatively and restrict citizenship to the most active/value-added community members. You can always loosen requirements in the future, but going the other direction is problematic and risks too much overlap with the Token House, which could prove fatal.

Another reason an NFT-based governance model is exciting is because there are obvious shortcomings to a straightforward token-based model.

Shortcomings of Tokenholder Voting

This critique isn’t new. If you want a longer read, Vitalik wrote a great post last year articulating the shortcomings of token-based governance models. Here’s a quick summary:

• Inefficient. Absent some form of delegation, token-based models are inefficient. Decisions are slow, and most tokenholders don’t care.

• Conflicts of interest. Token-based governance prioritizes the interests/incentives of tokenholders over other members of the community. This often leads to over-valuing the goal of “token go up”.

• Voting Buying / Attacks. Token-based models can be manipulated by simply buying a significant amount of the token. Small cap projects are the most at-risk, but this presents a problem for larger projects too.

Another shortcoming I haven’t seen articulated at all is the concept of “governance debt”.

“Governance Debt”

Projects with a token-based governance model are less likely to experiment with new models because of what I call “governance debt”. This isn’t so much a shortcoming as it is a result of implementing something that you know needs to be redesigned later.

Just like technical debt in software development, governance debt represents the implied cost of choosing a simplistic or “quick and dirty” governance model instead of a more thoughtful, robust model. The larger the debt, the more painful it is to implement changes because you become paralyzed under the weight of the current system.

My hunch is a growing number of founders and projects acknowledge (mostly privately) that simple token-based governance models have critical flaws. However, they can’t or won’t consider alternatives because their projects are too far down the governance token path.

In many cases, governance participation is the only utility for a token. Removing that right renders the token completely worthless. Token-based governance also allows for insiders (founders and early investors) to maintain control while claiming decentralization, something that has both advantages and disadvantages.

Parting Thoughts

Because of governance debt, the majority of governance experiment/innovation will come from new projects like Optimism that are implementing decentralized governance for the first time and conscious of not repeating the mistakes of the previous generation of projects.

They don’t have switching costs. And as long as you set expectations with your community, there will be some tolerance for trial and error.

No pun intended, I’m optimistic about NFT-based governance structures and I hope it inspires other projects thinking about governance design to try something different too.

If you’re interested in another argument in favor of NFT-based governance, I highly recommend this post from @anonalyx - NFTs > ERCs.

Thanks for reading,

Andy

—

Not a subscriber? Sign up below to receive a new issue of 30,000 Feet every Sunday.