Claim: The current state of on-chain governance is trending more towards oligopolies


We are having a live debate with Meltem Demirors this week about the claim she makes in her recent article that the current state of on-chain governance is trending more towards oligopolies, where each protocol has its own ruling party and its own oligarchs (i.e., exchanges, investors, individuals who may collude to have more influence).

Paul began to outline his argument for why he supports this claim, and also proposes some ways to counter-argue it here.

What do you think? Do you back or challenge this claim? Why?


There’s two camps of people: Those who believe on-chain governance is better, and those who don’t. We can use this conversation to discuss the arguments for both. To do that, it’s important to paint the broader picture around on-chain governance, it’s uses, benefits, etc.

Questions to discuss:

  1. What is “on-chain governance”? How is it different from off-chain governance?

  1. What is the problem that on-chain governance is trying to solve? (i.e. advantages)

  2. What are current approaches for how on-chain governance is handled?

  1. Why do you believe on-chain governance trends towards oligopolies?

  2. Do the same dynamics exist for off-chain governance?

Posting @paulapivat’s analysis here:

The reasons Meltem provides for why on-chain governance trends towards oligopolies (see above) does not seem to be unique to on-chain governance schemes. In off-chain governance schemes (i.e., what some call loosely coupled, hard-fork model), the same challenges remain, namely, the majority of users will be disinterested in many issues and will not want to invest time/energy to vote. Moreover, lobbying and ‘politiking’ are common in off-chain systems as well.

However, the key is that for on-chain governance, unlike off-chain governance or in traditional voting in the real world, vote buying is increasingly easy. Smart contracts, specialty hardware and “dark” DAOs make vote buying easy, particularly for any permissionless system where users can generate their own keys (see Phil Daian’s post here for details). Phil would argue that hard fork-based governance (i.e., off-chain) provides users with the only exit from oligopolies (see here).

For other thoughts on why on-chain governance, more so that other forms of governance, lead to oligopolies, see Vitalik’s posts (here and here).

  1. Is there evidence of cartel-like behavior for protocols that have off-chain governance? If not, what is it about off-chain governance that mitigates cartel-like behavior?

  2. Is there evidence of collusion, cartel-like behavior for NEO, Tezos, Dash, and/or Decred?

  3. What are other disadvantages of on-chain governance?

  • Active participation in governance is time consuming
  • Lower voter turnout (highly dependent on ease of voting, importance of voting issue, risk of voting, personal gain)
  1. Is there such thing as “optimal” governance?

The pursuit of an optimal structure for governance has plagued every ‘tribe (society and organization) since human consciousness; and is still an unproven experiment (outcome TBD) . Look no further than the fragility of our existing ‘democratic structures’ for evidence.
As you’ll likely agree, the optimal form of governance for each of us is that which serves our individual preferences and benefit, regardless of how deeply and ardently we justify and rationalize these arguments
– Meltem

  1. Do all governance schemes inevitably become abused and malignant, devolving into tyranny, oligarchy, and ochlocracy?

  1. Would the separation of powers (trias politica) model be a more appropriate framework than traditional corporate governance or full on-chain governance?

More context here:

  1. Is Identity the solution to solving the oligopoly problem?


A great thread showing an example of low voter participation at Aragon:


Another interesting thread on low voter participation:

One thing that was mentioned was that the likely reason for lack of participation is lack of incentive. The only reason to participate is to “govern”.

I agree with Rocco that if there was a reason for more voters to care, the participation wouldn’t be so abysmal:


I would also like to challenge if on-chain voting is most efficient in generating value for end users of the protocol.
Irrespective of whether if voters can be bribed, are the most suitable voters collecting governance rights? Do we need suitable external market signals (E.g., if voters have graduated from a prestigious university, worked in reputable corporations, developed successful in prior life) to qualify the selection of voters, and can this be enforced on-chain? What process do/can the protocols developers have in place to ensure the quality of decision making is maintained?


Arguments for yes on-chain-governance won’t lead to oligopolies

  • Financial incentives will overpower. @HealingVibes gave an example that hoarding and vote-buying occurs in DPOS systems. For example if Binance owns a lot of a DPOS coins they could proxy votes to themselves or another party they support (which would get a large degree of votes from that relationship)

Arguments for no on-chain governance won’t lead to oligopolies

  • Voting power is more diluted in on-chain governance; off-chain is already an oligopoly. A critic of Bitcoin’s governance said this on Reddit.

BTC, a coin controlled by Blockstream and Bitmain that takes years of debate and social media wars to make small decisions like a slight block size increase, ultimately ending in a resource split. Source

  • Harmful oligopolies won’t happen because the option to fork out the codebase still exists.

  • Is it an oligopoly or a benevolent dictatorship? is an oligopoly bad? A few people having control is not a bad thing for a new product in its early stages. Haseeb Quereshi wrote an excellent blog post on this.