I’d like to start a discussion about incentives and rewards for delegates. Being an active delegate requires time and effort. It’s not just about reviewing proposals, asking questions, and voting—it also involves weekly calls, in-person meetings, self-education, staying updated on web3 and ZK technology, and keeping up with ongoing changes by engaging in discussions with builders and others in this space.
At the DC7: ZKsync Delegate Meetup, many active delegates expressed support for an incentive program for ZK Nation. I would like to open a discussion about this.
Active delegates, I’d appreciate your thoughts on the following:
Do you support the idea of a delegate incentive program for ZK Nation?
What types of programs do you prefer? (monthly stipend, retro rewards,…)
Are you or your organization interested in working on a proposal for this?
Here are a few examples of delegate incentive programs at other DAOs:
I support a potential incentive program for delegates and I personally would prefer OP-Style retro rewards, that evolves over time, promoting decentralization.
The reason I became a delegate was to help avoid a centralization of governance within the same big delegates we all know (no hate!) and I believe such a program should try to promote just that; normal users participating.
My view on this has always been relatively simple: delegate work is work. The idea that delegates don’t get paid in an industry as well-funded as ours has always struck me as absurd. We’re not monks!
I would add to the above list Uniswap’s delegate program since it’s well documented and follows the module that answers (2) for me which is monthly. This seems a fairly simple module and mirrors the “real world.”
As they say, you get what you pay for. The reason the above programs - including our direct competitors - have introduced incentivised delegates is because that’s how you ensure quality delegates actually sit down and read your proposals and keeps them attuned. If you want to see a dead governance forum then don’t have one and we’ll slowly see everyone peel off over time.
One small thing we should avoid: rewards based on posting on forums, which encourages spam.
Nice to see this pop-up, I was thinking about this lately as I kept spending a lot more time around ZKsync (learning about the ecosystem, meeting and talking to folks from the ecosystem, meeting and talking to other delegates, learning about TPPs and even opened up two docs for TPP temp checks that I hope to submit next year).
Do you support the idea of a delegate incentive program for ZK Nation?
Yes. I find myself investing a lot of time here regularly. And even though I enjoy it, as a company founder, ens delegate and service provider, and occasional investor, my time is a scarce resource
What types of programs do you prefer? (monthly stipend, retro rewards,…)
Open to discussing this. Type of program, eligibility criteria, compensation structure breakdown. This might be a topic of discussion for the next Delegate call @theshelb.
Arbitrum has a good DIP, as far as I can see. Since most of it is predefined criteria-based, we could possibly automate it through a TPP (I’d love to explore this).
Are you or your organization interested in working on a proposal for this?
No. But I’m happy to contribute to the discussions and make sure this is designed in the best possible way.
Thanks for the post @Tekr0x.eth. Like @polar, i’ve been pushing at “governance work is work” for years and I completely agree with the position that if you want good governance participation you have to pay for it. I also think it is becoming the dominant paradigm that some form of delegate rewards programme is installed in the major DAOs across the space and so it is wise for ZK nation to have a full consideration on this topic.
I will say though, doing this right is exceptionally challenging and could create all sorts of weird perverse incentives if not designed correctly.
Some thoughts to frame the problem:
Incentivising people who turn up and vote does not necessarily incentivise good decision making. It can lead to donkey voting, which is very difficult to detect.
Any kind of system that incentivises engagement on forums, will lead to ‘farming’ of the forum and one of the worst outcomes is that the space that we use for decision making becomes polluted with people chasing token rewards.
Quality deliberation is an intersubjective phenomenon and is getting even harder to detect with LLMs and again, if not thought through correctly, we end up with a forum full of AI slop.
The simple versions of these systems fall back to established esteem games, which could further ossify power around established professional delegates and well known ‘protocol politicians’ and do the opposite of the intention and alienate people from participating.
Retroactive approaches require people to work for free for X period on the hope that they will get paid, other barriers like “6 months of voting history” just reward the OGs on the forum and create undue barriers to entry.
There’s always a line to be drawn somewhere and like in the optimism example above if top 100 delegates get paid, then you get 100 delegates and no more. Realising the reward threshold means there’s no point in participating at all unless you’re in the frame.
Some interesting lines of enquiry:
Sortition is under utilised in DAO governance IMO and could be used in multiple areas. One of which could be this, “you’ve been selected to be an incentivised delegate in this season” etc. Ideally your probability of inclusion would be materially influenced by your participation.
Active recruitment of people to come and contribute to the forum and build a delegate position. There are great minds in the industry out there who don’t participate in spaces like this and it would be great to draw them in. Also intentionally building a collective of minds that have genuine diversity of thought and are rewarded for critique is important.
Vouching / nomination systems building on from the previous point, one of the best decentralised governance mechanisms is vouching IMO and again is under utilised.
Delegate signalling / feedback systems that make the delegate role an actively bi-directional conversation between delgatees and delgators could generate information that could be used to shape incentives. For example, this delegate with X number of delegators has this approval rating, which leads to Y rewards.
Rewarding Re/De-Delegation - recognising not just the work of delegates but the decision making of delegators is under explored. To break the ossification of power around old guard delegates, you could promote more liquid democracy action by having moments where delegators making active decisions about who they delegate to and then reward them for it.
Crowd Appraisal / Peer-Review - the best signal for who is the best delegate and thus who gets rewarded, would be an objective analysis of who has done the best work. Building systems that scale this action is very challenging but is well worth exploring.
Overall
This should be considered, but not necessarily rushed into IMO. It’s very easy to make things worse rather than better with bad incentives and I suspect some of the early moving programmes in other DAOs are going to do just that. Whatever reason people have for being here currently is working to some degree. Votes are hitting quorum easily and there was a lot of very critical and thorough discussion on the Ignite proposal. Less so on the ZIPs, which probably does need more urgent attention.
The end game target has to be a highly engaged DAO that draws on decision making and deliberation from across the wider elastic chain ecosystem and incentives are inevitably going to have to play a part in that. The best DAOs will have the most engagement and this is likely to be a key metric for appraising them (and their L2s) in the future, so this could be a huge value driver in the future, but is also a bit of a minefield.
Thank you all for your feedback. I would like to hear thoughts from some other delegates as well. Maybe @SEEDGov can also share their thoughts here since you have experience running incentive programs?
I’m categorically against any sort of ‘retro delegate rewards’ on a delegate-by-delegate basis. This will just lead to more I-scratch-your-back-you-scratch-mine type dynamics between delegates and whoever decides the rewards. It creates a simp economy and can even distort the way people vote.
Just rewarding being a delegate is also bad–it’s like a participation trophy. Also, it could lead delegates to govern in a way that entrenches their ‘salaries’. Rather, the program should aim to reward being a high-quality delegate, which basically means a track record of ‘voting correctly,’ in some sense.
One possible way to go is a weaker futarchy style…basically reward those who vote for the ‘winning side’. This however doesn’t work unless voting is blind (commit/reveal)–otherwise it will just cause flocking behavior. And we don’t currently have blind voting.
Another idea would be to do retro style rewards, but based on retroactive rating of proposals rather than individual delegates. An existing or new BORG/council could rate each proposal after each governance ‘season’ (with ratings spaced out enough to allow time to show how the proposals played out) and basically just reward each delegate that voted ‘the right way’ in retrospect. So if you voted yes on a proposal that was approved and that proposal still looks good some time later, you get a reward, if you voted no on a proposal that was approved and that proposal later looks like a disaster, you get a reward. And so on. This avoids simping and grifting dynamics because it treats all delegates who voted a particular way on a particular proposal the same as each other.
Another point is that delegates should be completely disqualified from having influence over delegate rewards, beyond establishing the program in itself. i.e. if there is such a BORG/committee as described in #4, no delegates should serve on it.
Later Edit: another thing I’d be interested in, with any delegate reward scheme, is finding a way to disproportionately reward self-delegation. This may sound paradoxical, but I truly believe those with the most skin in the game will make the best decisions.
Here’s a quick recap of opinions on the proposed delegate incentive programme, keeping it simple and straight to the point:
@yonathan supports an incentive programme and prefers OP-style retro rewards to encourage wider participation.
@polar believes delegate work is valuable and deserves compensation, suggesting a monthly stipend similar to Uniswap’s programme.
@cap agrees that delegates invest significant time and are open to discussing programme types and structures, particularly Arbitrum’s DIP.
@drnick cautions that while delegate work is valuable, a poorly designed programme could lead to negative consequences like ‘donkey voting’ and forum spam. They suggest exploring innovative solutions like sortition and vouching systems.
@Gabriel supports the programme and leans towards an OP-style retro reward system, but suggests combining it with a small monthly stipend.
@lex_node is against rewarding delegates individually, proposing instead to reward delegates who vote for successful proposals in retrospect.
I want to ask all other active delegates again to share their opinion in the comments below.
Next week (10th Dec) we can discuss this topic during⚡Standing ZKsync Delegate Call⚡
thanks fro kicking off this discussion which is very important.
Do you support the idea of a delegate incentive program for ZK Nation?
Yes. I think its key to keep good quality in governance.
What types of programs do you prefer? (monthly stipend, retro rewards,…)
I think a blend of channels based on learnings from other ecosystems could be very strong (in the sense of @Gabriel):
A kind of basis stipend, on a monthly basis and maybe for cycles of 6 months, with check points based on voting activity, contributions in the forum, and calls? (although its always hard to measure this and protect against gaming).
A RPGF style reward to make sure that activity is measured afterwards.
A deliberative process with randomly selected members of the community to distribute some part of the rewards (like participatory budgeting a bit or a jury).
Are you or your organization interested in working on a proposal for this?
YES! I am focusing on deliberation and sortition ( kudos @drnick) so I would love to support this.
Just to note I am not necessarily suggesting a monthly stipend. Just that Uniswap does that. But I am a supporter of simpler solutions when it comes to incentives. (I fear only one thing, Goodhart’s Law).
On one hand being a delegate requires significant work and should be compensated as such. On the other hand, wouldn’t compensating delegates bring in less desirable actions such as spamming, lower quality delegates, etc…?
It is also difficult to correctly review such a proposal since we are delegates, and as such could be quite biased. I would be interested in hearing more feedback from “neutral” participants.
In the event a proposal is put together to compensate delegates, we should be extremely careful on compensation criteria since as mentioned before they may incentivize less desirable behaviors. For instance, I would be extremely against:
compensating votes on winning proposals: shouldn’t we compensate votes on losing proposals too? This might create an incentive for people to “yay” most proposals to increase compensation.
forum posts without additional review: this would otherwise create a risk for more forum spamming
other forms of social engagement: delegates are not ambassadors
Maybe an alternative idea would be to have levels or compensation trenches based on various “ranks” among delegates? To that extent, it would be interesting to review what the Polkadot technical fellowship had put in place, I feel it could be adapted to this DAO. Polkadot Technical Fellowship · Polkadot Wiki
This post raises an important question about governance in decentralized ecosystems - I’ve been thinking a lot lately about the role of delegates.
Who does a delegate truly answer to/should be accountable to, mostly from a legal POV?
a. Some delegators have formal agreements with their delegates, often involving compensation. Should these delegates prioritize their delegators’ interests above all else? What happens when these interests conflict with the broader vision outlined in the ZKSync Governance System North Star?
Do delegates have an obligation to serve their delegators’ interests? I would say yes, but up to which point? And let’s also remember delegators can redelegate whenever they desire to.
b. Delegates also have a responsibility to uphold ZKSync’s interests and ZK Nation’s vision. In my perspective, when the ecosystem directly incentivizes delegates, it changes the dynamic between delegators and delegates, I see benefits in this ecosystem-driven incentive structure. When delegates are incentivized by the ecosystem rather than just individual delegators, they might be more likely to make decisions that benefit the wider community of stakeholders, rather than focusing solely on their delegators’ interests.
I would be very interested in understanding different perspectives on these points.
This is really interesting. I didn’t know this was a thing! I can see your concerns that arise from this.
Solution suggestion: a general-purpose Conflict of Interest (CoI) statement that applies to all delegates, could be added to the ZK Nation in Delegate Standards, or working Constitution.
Does anyone else see this as an issue? If so, feel free to suggest some ideas. If not, we can leave it as is for now.
Generally speaking I am in favor, mostly because “delegate work is work” as stated before.
At the same time I am not sure we are at the point yet, where zksync delegate work is work. I understand that checking the forums and being up to date is not a thing of seconds, but at least right now the work load doesn’t seem to high. If delegates are really interested in zksync then spending on hour or two per week is something we can expect without incentives. Or in other words: People that can’t spend that time now for zksync likely won’t quit other jobs/ responsibilities just because they receive a reward X.
I think all the things to consider mentioned are very important, cause gaming the system could do more harm than good. To be honest, I don’t know about other protocols, but in zksync a lot of (locked?) insider token balances seem to be split across hundreds or even thousands of addresses (currently there are technically about 285k delegates) and self-delegated. If there is a delegate incentive program at some point, these addresses should be double checked and best case eliminated from receiving rewards.
Thanks to @tekr0x.eth for bringing this discussion on delegate incentives forward. Based on our experience with governance across multiple DAOs, we’d like to share some observations and suggestions that we feel are important to consider:
Incentivizing early stage engagement
Arbitrum’s recent v1.5 iteration on their delegate incentive program has started rewarding delegates who provide thoughtful feedback during the formative stages of proposals which helps:
Identify potential issues earlier in the proposal process creating more robust proposals
Incentivizes delegates review proposals earlier which builds consensus before formal voting
This however increases the risk of chat-gpt summary/recommendation responses. Leading into our next point…
Quality over quantity
While high participation rates are important, we should be thoughtful about what behaviors we’re incentivizing. Some key considerations:
Raw voting/comment metrics can inadvertently encourage low-quality participation as mentioned by @drnick
Recognizing delegates who identify critical issues could be implemented through retro-active rewards rather than through standard base incentives
In the spirit of thinking through the lens of a TPP, having a public dashboard such as Karma that transparently tracks onchain metrics could support part of the implementation but there would likely need to be a manual component to manage disputes. Karma also derives a forum score with weights but this can inadvertently incentivize delegates to game the metrics:
After reading through all the comments, we are pleasantly surprised by how well the discussion has gone. We’d like to summarize some of the points that we think are most important as well as some concerns that need to be worked out.
Priorities for us
Recognition of Delegate Work
Delegates invest time and effort into governance, at a baseline, we are in support of “delegate work is work” and that deserves recognition.
Balancing compensation to reward quality contributions while avoiding perverse incentives is the central challenge.
Promoting Quality Over Quantity
Incentivizing high-quality, thoughtful engagement rather than raw voting or forum activity metrics is critical to avoid spam or superficial participation.
Maybe rewarding delegates who provide actionable feedback early in the proposal process or identify critical issues during deliberation, much like Arb
Fairness and Transparency in Distribution
Aligned with several contributor’s stressing the importance of equitable and transparent systems to determine rewards, avoiding mechanisms that favor incumbents more or be gamed.
Maybe having retroactive rewards tied to proposal outcomes, ranking systems with community input, and public dashboards like Karma with adjustments to discourage metric exploitation.
Encouraging Inclusivity
There’s value in encouraging broader participation and bringing in fresh voices. In favor of active recruitment of underrepresented perspectives.
Aligning Delegates with Ecosystem Goals
Incentives should ensure that delegates prioritize the broader vision of ZKSync rather than just individual delegator interests or personal gain.
We like the ideas of rewarding delegators for thoughtful re/delegation or tying delegate incentives to the impact of their decisions
Program Design for Long-Term Success
We acknowledge and agree that poorly designed incentives can harm governance by introducing bad actors, creating echo chambers, or ossifying power structures over time.
In favor of Pilot programs with incremental development and ongoing community feedback to avoid rushing into complex solutions.
Some concerns/things to look out for
Making sure we don’t incentivize votes for “winning” proposals or using simple engagement metrics as this encourages herd behavior and gaming the system.
Compensation structures must be carefully calibrated to balance recognition with sustainability. High costs or overly complex systems could strain resources or deter participation.
Retroactive rewards encourage performance but can exclude new participants, while monthly stipends risk rewarding attendance over impact. Maybe a hybrid approach here could be good?
First of all, thank you for bringing this discussion to our attention, and apologies for the delay!
We believe that providing rewards and compensation for delegates is an essential block in the current state of many governance systems. Expecting governance and protocol experts with ‘skin in the game’ to commit significant hours each week without tangible rewards is unrealistic.
However, it’s important to note that while many governance systems require delegate incentives, not all do. This need typically arises when being a delegate becomes a part-time or full-time job or when there’s a clear need from the DAO to create this level of engagement. Below, you’ll find some thoughts and insights based on our experience:
Goal and Target Audience
The first step in designing an incentives program is to clearly define its objectives and target audience. A program aimed at professionalizing engagement among skilled, reputable delegates will differ significantly from one designed to attract new members to the DAO, or one designed only for reaching quorums. These differing goals have distinct approaches and entry barriers designs.
Variable Compensation
Compensation should reflect the delegates’ level of involvement and the quality of their contributions. It should not only be a matter of voting and reaching quorum. Setting clear expectations is key, and rewards should be thought of in terms of different tiers to encourage growth. For example, in the Arbitrum DIP, achieving full compensation is challenging, but earning meaningful rewards is feasible for those willing to engage at varying levels of commitment.
Sustainability
Rewards must be competitive yet sustainable in the long term. It’s better to start with a small pilot program and expand as needed rather than launch an overly ambitious program that might later require scaling back, potentially excluding valuable contributors.
Addressing the Principal-Agent Problem
Competitive compensation can help mitigate the principal-agent problem. Large stakeholders may delegate responsibilities, but delegates may not always represent their interests accurately -or could even serve conflicting interests-. Properly structured compensations can enable stakeholders to directly hire representatives aligned with their goals, reducing intermediaries.
Regular Payments
Monthly payments are an effective way to keep delegates engaged, allowing them to adjust their focus and efforts to perform better and aim for higher compensation tiers. While retroactive rewards can work for some DAOs, they often involve unpredictable variables tied to market conditions and internal dynamics.
Participation and Feedback
In our experience, well-designed incentives not only improve participation rates in both off-chain and on-chain voting but also enhance the quality of feedback and contributions. While introducing more variables can initially be labor-intensive for the program managers, the right tools and automation can streamline these processes over time.
Code of Conduct
A Code of Conduct or Ethics for delegates is increasingly essential. Misaligned incentives or deviations from the DAO’s ethos and goals can arise, especially when there is potential for value extraction or collusion. Mechanisms should be in place to remove delegates who fail to align with the protocol’s and DAO’s best interests.
Program Management
Program managers should have sufficient flexibility to evaluate participation, manage resources, assign compensations, and even enforce rules of engagement when necessary. This ‘discretionary’ authority must be balanced with robust accountability mechanisms and community scrutiny to ensure transparency and trust.
From SEEDGov we’re willing to participate with the crafting of the proposal and we’re open to any questions that could help this discussion. Happy to share our learnings!
I have a question (and I could have replied to any post here, just using the last one for simplicity):
Should a incentive program reward delegates differently if they are just delegate for one L2 compared to delegates that are serving in several L2 governance systems?
I am asking this cause on the one hand I am sure want experienced and knowledgable delegates. On the other hand if all delegates are working for all L2s their position could be harmed cause it’s all “same same but different”.
My gut feeling tells me we likely need a mix - but it will be hard to craft a incentive program that is attractive for both delegate profiles?
