In our last piece, we discussed the difference between managerial capitalism and entrepreneurial capitalism—that in the latter the bigger equity holders in the company also run the company, and in the former the interests are dispersed more widely across more shareholders. In entrepreneurial capitalism, incentives are often better aligned since the people-making decisions are also motivated primarily by equity.
Perhaps I was too dismissive of managerialism though. Managerialism’s drawback is that no one is significantly incentivized to serve the company with its long-term interests in mind because equity is too dispersed. However, the charitable view is that you can now align the incentives of not just your employees and investors, but also your customers too. And perhaps this financial incentive is the thing that can finally disrupt the network effects of Web 2. After all, why else would people shift away from Facebook, AirBnB, or Uber if not for money?
Indeed, this is the promise of DAOs and web3 more broadly.
***
As you’re probably already aware of, there has been an explosion of interest in DAOs in the past few years. From decentralized crypto exchanges (Uniswap) to crypto lending (Compound) to a failed attempt to buy an original copy of the constitution (ConstitutionDAO). To briefly summarize for the uninitiated, DAOs are decentralized institutions — AKA decentralized versions of traditional organizations. DAO stands for:
Decentralized: Workforce, governance, and infrastructure.
Workers: The workforce in a DAO can be located anywhere.
Infrastructure: DAOs are typically accessed via the web browser, but their backend is typically the Ethereum network.
Governance: DAOs tend to be direct democracies, where any member can propose an idea and everyone can vote on every decision.
Autonomous: Rules are automatically enforced. DAOs can run autonomously, without the blessing of a government, or other higher body.
Organization: Coordinating around shared objectives. A group of people united by a common interest or purpose.
Per Packy McCormick:
“DAOs are a new way to finance projects, govern communities, and share value” — native to the internet”
(Credit, Patrick Rivera)
One of the key selling points of DAOs is their governance structure. Governance in DAOs is generally a bottoms-up, radically democratic community governed by a predetermined set of smart contract rules. Rules are set by community members and these rules are public and auditable. This allows everyone to see the rules and understand how decisions are made.
DAOs aim to create all the benefits of democracy — both on the ownership and the governance level. In politics we call this direct democracy. You don't have representatives, Congress, the Senate, or any of our traditional governance structures. Instead you have the crowd voting on each and every issue.
While this is a noble mission, when you look back you see that this structure has struggled historically. California, of course, runs largely in this way — which is part of why the state’s local government is currently a mess in so many ways. Similarly, the corporate version of this idea is holacracy, which also has a poor track record.
Direct democracies are impractical: if everyone has to argue about every single issue, it simply takes too long to get anything done. As anyone who’s worked in groups knows, nothing gets done if everyone has an equal say.
Direct democracies also suffer from information problems. Voters have little incentive to educate themselves on the issues. It’s actually irrational for them to spend much time investigating any given democratic choice, because the odds that a single vote matters (breaks a tie) are very low.
To quote Bryan Caplan:
“Voters’ lack of decisiveness changes everything. Voting is not a slight variation on shopping. Shoppers have incentives to be rational. Voters do not. The naive view of democracy, which paints it as a public forum for solving social problems, ignores more than a few frictions. It overlooks the big story inches beneath the surface. When voters talk about solving social problems, their primary aim is to boost their self-worth by casting off the workaday shackles of objectivity.
Marc Andreessen on a podcast once recommended an essay written by a prominent feminist political activist named Jo Freeman called The Tyranny of Structurelessness. Jo was a feminist living in communes in in the 1970s and the goal of those communes were to create a complete democracy. Jo recounted that when you have a large group of people trying to coordinate on something, you don't get pure democracy. What you get instead is cliques — groups formed in an effort to exercise influence over the whole. Those groups begin to take over and rule on behalf of the group, but they rule in an undefined and informal way which makes them unaccountable since their power isn’t formalized.
This evolution from democracy to oligarchy is known as the Iron Law of Oligarchy.
The Iron Law of Oligarchy
Coined by Robert Michels in his book Political Parties and summarized well in The Populist Delusion, the Iron Law of Oligarchy states that all organizations, including those committed to democratic ideals and practices, will inevitably succumb to rule by an elite few (an oligarchy).
The theory states that true democracy can't happen mechanically, because a large number of people cannot coordinate or organize. Only small groups can organize. So whenever you have “democracy” or a large number of people who are supposed to coordinate, they simply can’t. So instead a small number of people who can coordinate seize all the power.
For example, Michels noted that the socialist parties of Europe in the early 20th century, despite having a strong democratic ideology and mass participation, always ended up being dominated by a small number of elites.
To be sure, the founding founders saw this coming. America was famously established as a republic, or a representative democracy, not a true direct democracy. They designed Congress and the presidency to be offsets to what would otherwise become a mob. In companies, we run an even more hierarchical model. Generally, companies are run similarly to a combination of monarchy and oligarchy. There’s a CEO, a management team, and a small group of people who define the rules for a larger number of people.
Before we break down the Iron Law of Oligarchy, we need to first describe the three broad governance structures:
Monarchy (rule by one): Monarchies are efficient because one person can make all of the decisions quickly. However, if the monarch is bad or ineffective, this can lead to poor outcomes.
Oligarchy (rule by a few): Oligarchy is rule by the elite few. Because no one person can make decisions, it is difficult for oligarchies to make radical changes. While oligarchy is the most stable form of government, oligarchies tend to serve themselves and not the entire population. This is closest to representative democracy we have today.
Democracy (rule by many): Direct democracy is the least stable of the governance forms. Information costs are high (does your average person know much about intricate monetary policy?), and decision-making is cumbersome.
It is important to understand that there is an evolution between monarchy, oligarchy and democracy. The Greeks called this cyclical theory of governance anacyclosis. Monarchy is the most efficient form of government (top-down control is how successful organizations are run). However, monarchies have high variance because sometimes you get a bad or weak monarch. When this happens, power gets diffused, and the governance structure turns into an oligarchy. Oligarchies are not very efficient and tend to be bad at getting things done (imagine rule by committee). As the cycle progresses, eventually enough people get fed up with the oligarchy and democracy sweeps in to elect a new monarch.
The cycle from democracy to oligarchy often looks like this:
First, a large democratic group is formed. Everyone tries to have a say. This creates conflict, and all discussions go on for a long time as they try to resolve the conflict. Most people leave because the decision-making process is taking too long.
Eventually, fewer people remain and some of them present their plans. The crowd picks the plans that make the most sense and the planners get designated as leadership.
The planners slowly turn into the elite and start to reinforce their power. These elites also begin to control access to information and become progressively less and less accountable to the rank and file members.
Over time, the oligarchy develops values that are at odds with its regular members. Elites become higher status, and feel that they must have some way of differentiating themselves from the democratic masses. They slowly begin to push for things that are good for elites but not necessarily good for the rest of the group.
Once it is in place, democratic members lack the means to displace the oligarchy. It would require coordination that democracies simply don’t have. Additionally, the oligarchy won out as the most effective structure due to its strategic capabilities. For this reason, in any formally democratic organization, power will always slowly disperse away from members and into the hands of an elite.
We’ve established that all democratic organizations eventually become oligarchies. Why does this matter for DAO governance? Well, any democratic governance structure you put together for your DAO will lend itself to becoming an oligarchy over time. Is it possible to slow down this process through financial incentives and pre-programmed rules within the DAO? Maybe! But it’s worth knowing that the default within most democratic organizations is to trend towards oligarchy.
The pure democracies of web3 are going to have the same problems that democracies of every other period for the last 500 years have also had: that while they might start as democracies, they won’t stay that way. Instead, they're going to end up as some combination of oligarchy and monarchy — at least at a governance level. Anecdotally, I’ve seen builders in web3 start to decouple equity and governance. If builders can prove that you can get the benefits of entrepreneurial capitalism (sufficient authority and upside for owners) and the benefits of managerial capitalism (huge long tail of aligned shareholders who add value to the network) while avoiding the pitfalls of both, then DAOs will be truly game-changing organizations.
Thanks to Molly Mielke
"First, a large democratic group is formed. Everyone tries to have a say. This creates conflict, and all discussions go on for a long time as they try to resolve the conflict. Most people leave because the decision-making process is taking too long."
I think this is the part of the political process that social media changes. Social media dramatically lowers the cost of trying to have a say; you no longer have to spend all afternoon at the town hall. Instead, the masses remain (virtually) present in a way that prevents planners from organizing and becoming elites. Result: the most notable characteristic of politics in the social media era is that mass uprisings (OWS, BLM, January 6) come and go without creating any proto-institutions that can take effective action.
I think the big picture is accurate, but DAOs do not have to be democracies, just oligarchies with more check & balances better functioning as the tech is enabling:
- easier participation
- more transparency which means more accountability (not saying code is law, but it takes more effort to obscure decisions - or value transfer)
- standards and best practices across difference geographies, cultures, industries