Improving governance participation rate at MakerDAO

Fellow DAOists :slight_smile:

The past few weeks have been very interesting at MakerDAO—we’ve seen a ton of discussions about long-term topics such as Maker’s mission, the RWA strategy, MKR tokenomics and more. I have also read a lot of critique on the state of Maker governance. With this post I’d like to zoom into one particular problem that Maker governance is facing today: Low governance participation rate.

With this post I hope to achieve the following:

  1. Highlight the problem and its potential causes;
  2. List potential solutions on multiple time horizons;
  3. Spark a discussion on how to initiate some of these improvements.

Below are some other critiques on Maker governance that I’ve read over the past few weeks but won’t address in this post. I feel either not mandated or not qualified to address these.

  1. Low delegate retention rate (eg. delegates pursuing other opportunities in the space or laying down their role as delegate);
  2. Discontentment with the MIPs framework (eg. too lengthy, too bureaucratic, greenlight polls not effective);
  3. Insufficient financial alignment between MKR holders and CU contributors;
  4. Legal liability concerns of governance participants.

Problem statement:

“Low governance participation rate in MakerDAO”

Even though Maker is the oldest and biggest DeFi DAO and the quality of the discussions on our Governance Forum is (arguably) high, not a lot of individuals actively participate in governance votes (ie. polls and executive votes). Some data points to support that claim:

  • Of all the MKR tokens in circulation (~901k), only ~172k are currently locked in chief and protecting the protocol from governance attacks. That’s less than 20%. Only ~80k MKR tokens have voted on the latest executive proposal.
  • Three recent governance polls which were arguably important only saw up to 21 unique voters, including delegates—representing no more than 108k worth of MKR voting power (example 1, example 2, example 3).

What’s interesting to note is that the governance participation rate in more recent, smaller DAOs is much higher. Let’s consider BanklessDAO as an example; even though their governance proposals are significantly less complex compared to the proposals we see at MakerDAO, their voter turnout is outright impressive (>700 individual votes for some proposals, without explicit incentives!).

Although there are other protocol DAOs which are dealing with even lower participation rates, I would argue that we should prioritise addressing this challenge in order to further decentralise MakerDAO governance. Whatever long-term direction Maker decides to pursue, it cannot succeed without a healthy and engaged governance ecosystem. We need a diverse set of individuals to consider and research proposals, vote, and actively advocate for their views.

What might be the underlying causes

1. Negative financial incentives:

— High cost of participation due to gas fees. Our voting system relies on Ethereum L1 transactions, which have been increasingly expensive due to high blockspace demand.
— Opportunity cost of locking MKR in Maker governance vs ‘putting it to work’ in DeFi for yield.
— No explicit positive financial incentives to participate in Maker governance (besides defending the protocol from governance attacks).

2. High complexity:

— Proposals are increasingly complex and require specific knowledge and/or due diligence in order to form an opinion on the matter.
— Information relevant to governance proposals is often scattered throughout the Governance Forum, Discord servers, various dashboards, and more.
— Difficulty keeping track of all the proposals and their status (eg. someone engages in discussion on the Forum but does not notice the vote going live).

3. Low perceived impact of individual MKR voting power:

— MKR token voting makes for a plutocracy, where the voting power of some whales and delegates are orders of magnitude bigger compared to most holders. This perceived minimal impact might demotivate active participation.

4. Low composability with the industry:

— Active participation in Maker governance involves moving your MKR tokens around on L1. Although I realise that this mechanism was designed deliberately to prevent governance attacks, it ends up excluding a lot of MKR holders who are not necessarily misaligned with the success of Maker.
— A recent example of this is @Mringz of Index Coop who was eager to get the MKR tokens in the DPI index active in Maker governance—today this accounts for >23M USD worth of MKR! We did not succeed in finding a solution to enable this.
— Although @Growth-Core-Unit and @dux-core-unit have been working hard to help custodians (eg. Anchorage, Coinbase Custody) enabling Maker governance participation for their clients, it involves a lot of custom development and therefore takes a long time.
— Maker governance does not integrate with today’s most popular DAO governance aggregators and tooling such as Snapshot, Tally, Boardroom and Commonwealth.

Challenge statement:

“How might we improve the governance participation rate in MakerDAO?”

Potential solutions related to incentives/costs

Short-term

  • Set up gas cost reimbursement program for recognized delegates. Is @SES-Core-Unit still working on this?
  • Make more use of forum sentiment polls for soft consensus seeking. We might want to consider using less on-chain polls for soft consensus seeking and defer to forum polls instead, since these do not incur gas costs. I personally like @LongForWisdom’s recent thread on gauging sentiment regarding the Clean Money vision.

Medium-term

  • Temporarily use third-party tooling for polling. We might want to consider integrating with Snapshot for gasless voting on some eligible governance proposals, which might incentivise smaller MKR holders to participate.
  • Approach Maker’s voting system as modular and run experiments. Instead of aiming to completely revamp the Maker voting system, we could run isolated scaling experiments against various components in parallel. Examples would be 1) trying to implement a fork of Snapshot into the Governance portal for polling, 2) experimenting with snapshot voting on rollups (eg. StarkWare!), 3) experimenting with moving on-chain governance to L2. Credits to @colby who shared this perspective with me earlier and who’s doing exciting work on this front.
  • Explore incentives for voting. I personally like Snapshot’s built-in POAP support for votes as I’ve experienced this non-financial incentive being sufficiently powerful to activate me. Another idea I read about was exploring unique NFTs for certain voter milestones, similar to the governance forum badges. We might want to run some experiments with this to test effectiveness.

Long-term

  • Completely redesign Maker’s voting system in relation to improved tokenomics. If we end up improving the MKR tokenomics we might want to take the opportunity to redesign the Maker voting system from scratch to address the problems mentioned in this post.
  • Move governance polling to L2. We adopt a L2 scaling solution that does not compromise in the security and trustlessness of L1 Ethereum and fully integrate this solution into the Governance Portal, which would enable near-gasless voting.

Potential solutions related to complexity

Short-term

  • Create tooling for tracking the status of governance proposals. @GovAlpha-Core-Unit did a great job at this through creating the Governance Tracker spreadsheet—I’d love to hear how @Recognized-Delegates have been using this and whether it’s been useful
  • Seeks ways to aggregate relevant information for voters. This is something we’re looking into at @dux-core-unit—we already came up with some relevant feature ideas for the Governance Portal that we’ll prioritise early next year.

Mid-term

Long-term

  • Have designated subject-matter experts in the DAO. We could explore ways to make it easier for voters and delegates to understand which themes are being addressed by a governance proposal, and who are subject-matter experts. This has been partially addressed by @rune’s concept of Decentralized Voter Committees (DVCs). We might even want to consider paid SMEs outside of the DAO.

Closing remarks

I’d like to repeat that the above is not an exhaustive list of problems, causes and potential solutions—it’s merely an attempt at aggregating the comments I have read over the past weeks, complemented with my own views.

My assumption is that the most effective method for increasing governance participation in MakerDAO would be removing friction in the voting process. This can be achieved through 1) lowering costs of participation through reimbursing today’s most important governance participants (ie. our beloved recognized delegates) and seeking to mitigate gas expenses asap—and through 2) enabling effective decision-making through aggregating relevant information related to governance votes. Therefore I would love to collaborate with some of you to set up experiments and initiatives focussed on these two goals.

Please feel free to share thoughts on this post format, any additions to the problems/causes/solutions mentioned above, or any other ideas for improving the governance participation rate. I sincerely hope this post sparks some discussion that eventually leads to initiative.

:sparkles:Thank you for reading!:sparkles:

25 Likes

I will have more comments on this.

I spent a lot of time trying to come up with a rewards model for governance participation that was simple and addressed key concerns. I was waiting for @LongForWisdom to comment to see if it needed adjustment. I posted it to my delegate page mostly because I wanted to solicit private comments before publishing it.

One thing I wanted to highlight was a point you made regarding DAO governance participation in general and that it is low generally.

One point you don’t make that I intend to address with my Maker growth plan is a general issue that ‘governance tokens’ also represent value. So there is this distinct competition between tokens in a place they can govern vs. a place they earn return for the holder. This needs to be changed. (expect more on this when I release my Maker 1T vision)

Since you brought up the issue I propose a potential solution.

This is kind of a first step. Expect more clarity on this within a complete vision post I am trying to finish up. It has been a hellish week, month and year for me that so far is not getting better unfortunately, this is driving my time in a fully interrupted fashion. :frowning: The above is what I came up with after month or two of collecting and analyzing governance data, and then trying to formulate an approach.

3 Likes

Deniz thank you for creating this write-up. Very much needed. From a Delegate point-of-view:

I think you make a very interesting point about reimbursing Delegates for gas fees–but I believe that with the new compensation structure, some Delegates should be able to afford their own gas fees. As you know, there are some VCs who want to delegate and are also providing a gas stipend. This is cool–perhaps what GovAlpha should do is provide a threshold of which delegates should be reimbursed.

As an example, if a Delegate Platform only carries a few hundred, or thousands of MKR – perhaps these Delegates should qualify for a gas reimbursement. But for Delegates that will be compensated well, perhaps not.

Depending on how well Flip Flop Flap Delegate scales, we plan on hiring folks that can provide expertise on a number of subjects. But in the meantime we also would like to communicate more with the Core Units. We believe this will help in getting our heads-around some of these complex subject matters.

What is interesting about a low participation rate is that there’s a lot of people out there, who are looking to work in DeFi/Crypto. I have met a few folks who want to work in this ecosystem and they find the DAO governance structure fascinating. Perhaps @LongForWisdom @prose11 can to set-up a Bounty for Community members to Recruit Delegates? I personally would like to see Women Delegates. Diversity and expansion of Delegate types is super important, IMO. And yes, for sure–its not easy to find qualified people–but we won’t know for sure unless we try.

For Flip Flop Flap Delegate this is very useful. But I can also tell you that some Community folks I have spoken to are still confused as to how things (specifically collateral types) are prioritize. This could be because some folks don’t digest information thoroughly (as you said, there is so much going on), because I believe GovAlpha has done a nice job of communicating such.

Thanks again for putting this post together. I look forward to seeing others chime in.

3 Likes

I will say there is some opaqueness to the collateral onboarding process, but I’d like to think we’ve set it on a good path all things considered. The problem is you have a bunch of reasonably independent groups working together to add new collaterals and they only so much bandwidth. In practice this means the more exciting collaterals (in terms of scaling the DAO) tend to get tackled first because it’s easier for the CUs to know their work won’t be going to waste.

I think there’s a lot of potential with @collateral-core-unit being able to devote much more time to this process. I will be curious to hear if they have any recommendations we can be implementing in the short therm.

4 Likes

The collateral process is a top priority to clarify and iterate on.

I’ll have an update on Friday to talk about the CES progress and initial areas where we can make an impact as our team is coming up to speed. The good news is that we’re almost fully staffed and starting to take over responsibilities in the technical collateral processes.

6 Likes

I would like to experiment with this - I know a lot of small holders do not vote currently because they:

  1. cannot move the needle with their votes
  2. gas is quite expensive
  3. concerned w/ risk touching their MKR

If you eliminate 2 and 3 I think you’ll find a lot more MKR holders will participate in Governance despite not being able to move the needle themselves.

2 Likes

I’m a small MKR holder. I don’t vote due to high gas costs.

4 Likes

It seems like there is a lot of business uncertainty associated with becoming a Recognized Delegate. You have to invest a lot of time and effort to get started and the whales might not even notice you for weeks. Maybe GovAlpha could offer grants to help serious Recognized Delegates get up and running long enough to attract some MKR? Maybe this is a way to address the lack of delegate diversity.

4 Likes

I think this is a good way forward. With increasing complexity it will be impossible for a single person to have in depth knowledge in every subject. In my view the most important part of governance is being able to strike the right balance between opposing needs of experts.

I would say the most pressing expertise is competent legal advice. We have seen some discussions reduced to different interpretations of legal ramifications without a qualified opinion on the matter. Technical expertise can mostly be found in the Core Units, but for the long term it might make sense to have independent technical experts if delegates are meant to be a true check on the Core Units. As Real World Finance grows I would expect the need for valuation and credit experts to emerge.

I envision putting these experts into their own Core Unit and made available to all delegates and other stakeholders. Pooling the resources makes sense to me as the need is shared by everyone involved in decision making. Since the needs are similar it doesn’t make sense to do the same work by multiple entities.

Right now the lack of such a resource is being met by building teams around existing delegates. This makes sense if the delegates are meant to serve several entities, but less so if the goal is to focus on a single one.

3 Likes

My take. I currently delegate, reason being high gas prices. Otherwise i probably wouldn’t delegate for longer periods. I would probably delegate only when i am less engaged.

1 Like

Let me intrude in this interesting thread commenting on the above point with just a consideration taken from:

Warren Buffett’s Berkshire Hathaway corporation has two classes of stocks, Class A voting stock (NYSE: BRK.A) and Class B non-voting stock (NYSE: BRK.B). The Class B stock carries 1/10,000th of the voting rights of the Class A stock, but 1/1,500th of the dividend.

Rationale: it’s OK to have investors not interested in voting.

What we should make sure is that the few that actually take part in voting, well… actually vote well.

I believe the Sagittarium Engine (sketch of) idea of @rune goes into this direction.

4 Likes

So, this is the main reason I have been so in favour of the quadratic funding model that we’re currently trialling, it solves this problem in a better way than having us offer grants. The whole point is that new delegates only need a small amount of MKR in order to start getting non-trivial amounts of DAI (ideally enough to cover gas) quickly enough for it to be worth their time.

If GovAlpha starts offering grants, then it means we’re spending the DAO’s money on our opinion of who would make a good delegate for MakerDAO, rather than allowing MKR Holders to make that choice directly.

The elegance of the current system is that anyone can setup as a delegate, but they need to attract at least some amount of MKR in order to receive compensation, this gives MKR Holders the power to filter the collection of delegates down to those they want, rather than GovAlpha opining on who the DAO should have.

In many cases, delegates also have some of their own MKR which they can delegate themselves, which helps to get them started. This is also a positive filter. We want delegates to have some MKR to their name as that means they are more aligned with the protocol.

If a new delegate cannot either:

  • Attract some MKR quickly enough to cover their expenses; or
  • Acquire some MKR of their own which they can delegate to themselves.

Then they will have to pay out of pocket for gas until such a time as they can attract some delegated MKR.

Note that attracting MKR can begin even before the delegate contract is created and votes start to happen. There is nothing stopping potential delegates from using the forum or chat to lobby for delegated MKR, and to try to get commitments from MKR Holders prior to spinning up the contract.


We offer support to anyone that comes to us and says: ‘I’m thinking about becoming a Recognized Delegate’, including advice, reviewing platforms, discussing any concerns, etc. Ultimately though, it is in MKR Holders hands whether a delegate is successful, or not.

10 Likes

Good to see some engagement here :slight_smile: Some replies below:

Thanks for sharing your thoughts @MakerMan and thanks for sticking around as a delegate, despite the fuzzy situation re: compensation. I think you’re doing fantastic work and I always enjoy hearing your arguments on G&R calls so I’m looking forward to reading your longer vision write-up.

I fully agree that the next implementation of MKR tokenomics should address the ‘functional dilemma’ between voting weight in Maker governance and financial utility in DeFi—right now you have to pick one. With regards to your proposal for delegate compensation—I’ll abstain from the discussion re: the exact implementation since I believe @GovAlpha-Core-Unit covers that base and @LongForWisdom shared an idea below.

Thanks for chiming in @ElProgreso! I believe that delegates/voting committees hiring their own legal advisors is still far away, so until then I think we should try and map out the expertise that we already have available through CU contributors and other members of the DAO.

One thing we could do is leverage the poll categories that @GovAlpha-Core-Unit put together previously (click the poll type filter here) and try to identify 1-3 domain experts for each of these—preferably from within the DAO (eg. CU contributors, active forum members). We can turn this in a reference document that delegates can use to source advice/consultation on various topics. What do you think?

Couldn’t agree more, and this is where I believe we can really score big wrt governance UX. I’m hoping to get something off the ground soon, perhaps in collaboration with @colby and PE!

Interesting take! Thanks for sharing :slight_smile: Let’s wait for @rune’s iteration on the tokenomics proposal and continue the discussion there.

2 Likes
1 Like

I really like the way you are thinking about the problem and the several approaches to dealing with it.

I think the #1 deterrent is gas costs, and so I advocate for a gas rebate program for all voters, not just recognized delegates. Individual investors/voters should not be deprioritized just because the delegate program works well. It’s important we give that class of voter consideration.

Another piece I think will be important is building that database of Subject Matter Experts. Our team, @gov-comms-core-unit, is working on an internal stakeholder database that can include information about an individual’s subject-matter expertise or specialty. Eventually, the internal stakeholder database will be partially ported to a public stakeholder database where anyone in the DAO can essentially find the person with the relevant expertise.

The lowest hanging fruits here are the modular experimentation with off-chain voting and the gas rebates IMO.

Educating voters and keeping them on track with regard to specific issues is another very important field of focus. I wonder how we might be able to systematize and streamline issue-specific educational content and updates to voters who are looking for it. I’ve been envisioning a free subscription-like service that lets you get push notifications around specific things at Maker–whether it be parameter changes themselves, updates about specific issues, publications(Maker Relay, MGR, etc), or even delegate platform updates. The idea is you can custom-configure what you’d like to be aware of. We are working with @Content-Production to flesh out the vision and back-end tech to support it.

Anyway, just some thoughts. This is an important issue and it looks like there are many ways Maker will be improving so I am excited!

6 Likes

I largely agree with this post David and agree gas costs are prohibitive, especially for smaller MKR holders. It certainly isn’t an easy problem to solve.

My main concern with gas rebate for all voters is that it is open to abuse. What’s to stop someone maliciously voting with 0.000000001 MKR multiple times in order to drain DAI from the surplus buffer? Do you have any thoughts on how we might prevent this?

4 Likes

My first instinct is a gas rebate “dust” limit.

ie; Only give rebates to voters of 0.5 MKR or more (or something like that)

or issue the rebate weeks/months after the voting activity, so only issue rebates once per month or per quarter. That way there is huge upfront cost for such an attack and governance has time to identify them and exclude them from the rebate.

2 Likes

This!

I don’t understand why this doesn’t come up more often. Keeping MKR is a financial investment. If you’re a whale it’s worth voting with it because you’re impacting the future of your investment. If you are a smaller holder then you only have influence in principle, in practice you don’t matter. There could be much argument here about the Sagittarius Engine proposed by Rune but the fact is that the proposed cheap or free credit for MKR would solve this problem.

For my part I would take the credit and go play with it, and hand the voting power to a sympathetic delegate who understands in depth what he is voting for. The other option might be to ‘putting it to work’ by another protocol and not vote with it, or sell it (chances are a whale will buy it, it’s plankton to him) and buy something else. It may seem like a simple jerk attitude, but wake up, if you don’t want a few people running the DAO then we should do something like this.

Maybe.

3 Likes

Thanks for chipping in here David :v:

You make a good point, and it would be worthwhile to calculate the potential cost of such a (temporary) program in order to assess whether it’s worth it for the DAO. Also if we end up pursuing either your proposed scheme (or just recognized delegates for now), I think it would be best to start with a trial. I think SES was working on something related, maybe @juan or @wouter can confirm?

Agreed and I can’t wait to explore this further. I’m especially interested in the work between StarkWare and Snapshot—I’m keeping an eye on that project and I’ll try and get something off the ground.

I wasn’t aware that you’re working on this, but I’m happy to read this! Let’s stay in touch—we might be able to build something useful into the Governance Portal (eg. referencing to internal SMEs for each poll category).

2 Likes

I was going to comment on this but @mkrorbkr beat me to it. : )

We should run some basic analysis with some parameters to see how much this would cost the protocol.

@tadeo, maybe DI can take care of this?

2 Likes