Web3 treasuries and polycentric governance
Blockchain communities need to embrace more tools for polycentric governance
Governing the many billions of dollars of digital assets in web3 treasuries is hard.
How big should your treasury be? Should you diversify? Who makes decisions? Who executes decisions? What do good decisions look like?
My aim in this piece is to argue that treasuries should embrace polycentric governance to solve these problems. And we need better DAO tools to facilitate that polycentric innovation.
Why a treasury
Treasuries are pools of collectively owned digital assets.
The obvious role of a treasury is to fund things: public goods, or local public goods, or club goods, or ecosystem goods. Economic definitions can get in the way here. It doesn’t really matter — the point is that web3 platforms don’t have governments to tax-and-spend on the things communities might need, so those things are funded through community treasuries.
Treasuries tend to (or aim to) fund research, development, marketing, token distribution and ecosystem incentives. Each of these have elements of positive spillovers to the broader ecosystem (or web3 generally). Those spillovers might be, for instance, network effects through incentivising sides of the market to join.
Another (less-obvious) role that treasuries play is as a costly signal. Unwholly large treasuries demonstrate to investors, the community, and the market long-term intentions. They evoke reputational benefits for new protocols and teams (including anons). Treasuries are both credible commitments and moats against attack from would-be forkers.
But unless you simply want an enormous treasury of your native token then you’re going to need to make some spending (and diversification) decisions.
Spending your treasury
Let’s start with some basics about the problem web3 communities face in spending (or not spending) their treasury.
You don’t know what you want to buy now, let alone in the future (a knowledge problem)
Bad people want your bags and you don’t know who’s good and who’s bad (an incentive problem)
There’s two ways we can fix this.
The first way feels scientific and rational. Here we look for a single answer. If we can design the one correct set of rules (e.g. token voting aggregation formulas) then those rules will create just the right amount of knowledge coordination and protection against attack. Perfect.
This monocentric approach feels clear and familiar. If only we can set the right uniform parameters the answers to our treasury problems will fall out the other end.
To be sure, thinking about new voting models is important.
But those innovations must sit within a broader conception of polycentricity.
In theory, polycentricity means many centres of decision making with partial and overlapping autonomy within a set of overarching rules.
In practice, polycentricity means SubDAOs, working groups, delegation, committees, pods, advisory committees, councils. Hierarchy.
Many DAOs are already transitioning to more polycentric governance. They’re doing so for operational reasons: some decisions need to be made fast and in secret, outside of the gaze of the Discord.
But polycentricity should be embraced for deeper reasons. It’s important because of what polycentricity tends us towards: solutions to our knowledge problem and our incentive problem.
Polycentricity makes sense when we admit that (1) we don’t know what we’re doing; and (2) there are bad people and machines. Governance must acknowledge these problems.
There is uncertainty at multiple levels in these new communities. Obviously there’s day-to-day uncertainty about what to do (e.g. when should we turn down those ecosystem incentives?). But at a higher level there’s uncertainty about the rules by which we should make those decisions.
A system that is polycentric brings innovation and institutional diversity. There’s more room for experimentation (we’ve written about this elsewhere in terms of grant programs).
Smaller units make decisions using their local knowledge. Some (maybe lots) will make poor decisions. That’s fine. It’s a discovery process (but we must maintain feedback loops). We don’t just get information about the bad decisions they made, but about the rules by which they made those decisions.
Indeed, greater polycentricity means many blockchain ecosystems will look more like competitive laboratory federalism.
More generally, because polycentricity can also help mitigate the threats of attack (smaller areas of overlapping responsibility might be more difficult to attack than one central process) these systems create a more robust political economy.
The need for (more) DAO tools
Polycentricity should emerge through a discovery process. This need makes the rapid development of DAO tools even more important. Thankfully we’re seeing DAO tooling that helps:
ensure security and voting (e.g. Gnosis Safe, reality.eth)
reveal, measure and compensate contributions (e.g. Source Cred, Coordinape)
The risk today is that those tools are deployed in the same combinations across the suite of existing (and future) DAOs. We should be asking ourselves how these tools can be leveraged to facilitate more polycentric and diverse web3 networks.