Proposal to Change Dissolve Delay Bonus and Age Bonus Parameters

by Wenzel Bartlett (@wpb), Kyle Langham (@Kyle_Langham), Alex Mucalov (@ayjayem)

The current structure of the Internet Computer NNS entails risks that require attention by the IC community. The current NNS tokenomics parameters have failed to i) attract sufficient new participants to offset the dissolving stakes of current participants, ii) encourage current participants to remain committed long term to the IC, iii) address conflict-of-interest risks that may prevent the IC from executing its mission long term. We have published an article that explores those risks and proposes a solution with the intent of driving a community conversation and executing an NNS proposal for vote.

This article provides a detailed analysis of the problem and the proposed solution: The Current State of Governance of the Internet Computer and Proposals for Tokenomics Improvements that Incentivize Long Term Participation | by Wenzel Bartlett | Nov, 2021 | Medium

We believe the two changes identified below are worth a formal motion proposal in the NNS to enable a proper vote to see if a majority of voting power agrees. While there are many ideas that can incentivize long term staking, this proposal is focused on ideas that are actionable in an immediate time frame. The primary objective is for 90 percent of the total supply of ICP to be locked in neurons and to incentivize neuron owners to make their dissolve periods as high as possible, which are objectives identified in the tokenomics white paper published by Dfinity prior to the genesis launch.

Proposed Change #1: Increase the Maximum Dissolve Delay Bonus

The change that will be proposed is to increase the maximum dissolve delay bonus to 8x. The changes to the NNS governance that would be necessary to accomplish this task are shown below.

let d_stake = stake + ((stake * 7 * d) / (MAX_DISSOLVE_DELAY_SECONDS as u128));
assert!(d_stake <= 8 * stake);

Proposed Change #2: Increase the Maximum Age Bonus and Reduce the Time to Reach It

The change that will be proposed is to increase the maximum age bonus to 2x and to reduce the accumulation period for the age bonus to 6 months. The changes to the NNS governance that would be necessary to accomplish this task are shown below.

pub const MAX_NEURON_AGE_FOR_AGE_BONUS: u64 = 1/2*ONE_YEAR_SECONDS;
let ad_stake = d_stake + ((d_stake * a) / (MAX_NEURON_AGE_FOR_AGE_BONUS as u128));
assert!(ad_stake <= (2 * d_stake));

If this proposal passes, the voting reward yield distribution would change according to the blue trend in the graph below. The voting reward yield distribution based on the current dissolve delay bonus and age bonus parameters is shown in the red trend. The data in this chart is based on dissolve delay configuration and age of existing neurons identified by ic.rocks. A spreadsheet is linked in the references of the article that can be downloaded and reviewed. The spreadsheet is structured so each of the parameters highlighted in yellow can be modified to model alternate scenarios.

It is worth emphasizing that none of the parameter changes in this proposal would affect the total voting reward inflation schedule of ICP. The parameters in question only change the distribution of the voting rewards among the neurons in the NNS as weighted by voting power. It should also be pointed out that these proposed changes would not discriminate between genesis or non-genesis neurons or the amount of ICP staked in neurons: if a neuron owner is committed to long term participation in IC governance, then they would be rewarded in proportion. Based on current staking strategies, most neurons would have a decrease in voting reward yield if this proposal passes. Neurons with the longest dissolve delays and not dissolving would have an increase in voting reward yields. There is nothing keeping anyone from changing their staking strategy in order to achieve the greatest voting rewards if this proposal is adopted.

A time period of one week will be provided for deliberation on this proposal in the forum. The final days will be used to wait for quiet on the deliberation. If new, insightful, and actionable comments continue in those final days, then the deliberation period will be extended. Otherwise, seven days after making the forum post the formal NNS proposal will be made on this topic. The exact content of that proposal will be shaped by the ideas presented in this article as well as actionable feedback that occurs during deliberation as long as they move staking performance closer to the 90% goal and can incentivize maximization of dissolve delay.

No matter how this vote turns out, our hope is that this proposal generates a high level of participation among IC stakeholders, translating into significant active voter turnout, and ultimately resulting in the IC community collectively learning something useful about decentralized governance on the IC. Please encourage participation in this process. Please vote to Approve or Reject according to what you believe is best for long term governance of the IC. Please add your comments to this forum post so others will know your opinion. Thank you for your participation.

26 Likes

Regarding the proposal that mention the current NNS tokenomics parameters have failed to i) attract sufficient new participants to offset the dissolving stakes of current participants, ii) encourage current participants to remain committed long term to the IC, I do not see the problem is on the delay & age bonus multiplier (looks fair to me), but I do see the problem is on the requirement of minimum ICP can be spawned. (1 ICP right now).

This minimum requirement while not an issue for early contributors that have more than 10k ICP on average, but this is an issue and discourage new small stakers because it need at least 100 ICP ($4500 as current ICP price) to have a reasonable liquidity to spawn the reward at least once monthly.

To be more inclusive and attract small stakers, I think, minimum requirement for spawning need to be reduced 80-90%.

8 Likes

Hi guys, great write-up! Of course, I would be in favour however I see two issues:

  1. as you mentioned most of the neurons that would be negatively impacted hold a significant majority of the voting power…if they are currently not interested in the long-term staking would this be enough to persuade? Otherwise they are voting against there own best interests…

  2. the biggest issue I see and would like to see addressed is how to get more people to stake period…if there are people who are weary of staking for 8 years now and our currently thinking of only staking for 6 months to 2 years then this will potentially push them away.

I think it would be great to balance the proposal to try and get more people staking, even if it’s short-term, as well as rewarding long-term stakers in a more progressive and proportional way.

Virtually all other cryptos are offering at least a 5% apr for short-term limited lockup staking. Maybe this could balance the proposal. Attract short-term stakers by offering a reward/apr% on limited lockup staking for less than 6 months with no governance/voting rights. Actually I would like to see governance protected by moving voting rights to > 1year dissolve delay with a nominal apr % below that.

Just my two cents…

7 Likes

I think this is more a technical problem than something that has been decided to disadvantage smaller stakers. If you reduce that limit someone could spam the NNS with a lot of small neurons.

It would be interesting to know however why you have to spawn a new 7 day neuron instead of just instantly disbursing to an ICP address.

3 Likes

Currently, neurons get no voting rewards for the final 6 months in which they are dissolving, this means that if you stake for 1 year, you may be getting ~10% APR at first, but overall you only end up with ~5% due to only receiving rewards half of the time.

This already incentivises people to stake for long durations and deters people who want to just give it a try.

I think this proposed changed will exacerbate this problem.

Instead, we should reduce that final window during which no rewards are given to be ~2 weeks.

This way people can give staking a try for 2-3 months and get fairly decent returns which I think will encourage far more people into the system.

The rewards are already high enough for those of us who stake for 8 years, I think increasing these rewards further will do more harm than good by driving newcomers away and driving more power to those who are already heavily invested in the project.

13 Likes

Yes agree on giving staking rewards for short-term stakers but I’d be against allowing governance rights…that’s a huge risk.

8 Likes

As a fellow 8 Year Gang member I want to express my appreciation to @wpb , @Kyle_Langham , and @ayjayem for writing this article and putting together this proposal. I think we will have some great discussion spawn from this.

To that end; I wanted to summarize my takeaways from your article and share my perspective.

Facts:

  1. There is more ICP leaving the NNS than being staked
  2. A majority of neurons have a short ( < 4 years) Dissolve Delay (DD)
  3. A majority of Seed Investor & Early Contributor neurons are being dissolved

Assumptions:

  1. Neurons are being dissolved because voting rewards don’t provide enough incentive to “lock-up”
  2. Seed Investors & Early Contributor neurons being dissolved must be a sign that these individuals have lost faith in the IC and might be a security threat.

My Opinion:
There are a lot of factors that can influence a person’s decision to stake in the NNS. Voting rewards are only one of those factors. Several individuals I’ve spoken to don’t participate in NNS governance because they aren’t confident it will still be around in 8 years. The IC is considered unproven technology; if we’re being honest, that is probably justified.

I can understand why individuals who choose to stake in the NNS have chosen to dissolve their neurons. It’s another form of risk management. Sure, if you let the neuron dissolve over the next 4 years you will lose out on max rewards. But, let’s say 4 years from now the IC is on it’s last leg and because you let your neuron dissolve you can now exit your position with some amount of money. The person who locked-up is completely out of luck.

Alternatively, let’s say you get to the end of your dissolve period (or even just a year out) and you realize the IC is taking off and will likely be a success. Well, no problem, let’s just go move that DD slider to the right and fully commit this neuron to a life of luxury.

My point being that just because someone has a neuron dissolving today does not mean it will always be that way. In fact, given the volatility of the crypto industry I think it’s completely reasonable for people to manage their risk in this way. I think this is especially true for Seed Investors & Early Contributors who have already waited years (far longer than any of us) to earn their rewards.

Proposed Alternative:

  1. Remove the proposal to increase rewards for 8 year neurons. I don’t think this will be effective and it only serves a small group of dedicated IC enthusiasts who are already willing to take the risk (myself included).

  2. Keep the proposal for increasing the MAX_AGE bonus and reducing the time to reach it. I think this will be far more effective at getting people to stay locked and it is also applicable to all neurons; not just 8 year stakers.

Edit: I realize I did not address the issue of investor confidence. I don’t think that’s something we can address with a governance proposal. There is a lot of great work being done by Dfinity and the developer community. It’s my belief that we just need more time to see interest grow. As long as the core developers and app developers continue to find common ground on proposals I think we’ll be fine.

16 Likes

I agree with the general point made by both @hpeebles and @mechaquan. We should be trying to further incentivize staking across the entire DD range. (6 mos. - 8 years).

For security reasons I do not agree with reducing the minimum lockup from 6 mos. However, I would not be against a proposal to continue rewarding neurons that are dissolving during that 6 month period. A neuron that is dissolving w/ less than 6 mos. should not be allowed to vote anymore but it could still be rewarded a smaller amount until it is fully dissolved.

The question is why would we reward a neuron that isn’t voting? What are they being rewarded for?

That’s a good question and maybe we can build a condition into the system that says you can only be provided rewards if this neuron has a minimum history of voting. That way someone can’t just create a new neuron, immediately start dissolving, and then get rewards.

Would love to hear others thoughts on this.

7 Likes

I am in favor of this proposal. The most involved ones must be prioritized, for the wellness of the IC ecosystem. We can’t afford rejection of proposals like BTC Integration, everyone want it, be we could have seen this one rejected with the present tokenomics. This risk must be definitely suppressed. Let’s imagine the rejection of ETH integration. We must reform this.

2 Likes

How about if rewards stay as they are now but voting power scales linearly with lock up period?

That way we could give neurons rewards right down until they become liquid, but the overall control is still massively weighted towards those that lock up for 8 years.

I know that currently rewards == voting power, but that could be changed if needed.

6 Likes

I think it would be helpful to know how the foundation decided on the current parameters. There has never been a publication explaining how these were decided.

I don’t know why it wouldn’t scale linearly with time. Eight years is an eternity in crypto and 16x times as long as 6 months why would you only get 2x times the voting power and rewards for taking such a massive gamble? (note: I’m a bit biased because I have an 8 year neuron)

Alternatively I would also be down for a forever neuron one that you can never dissolve, but has a far higher voting power.

Also no matter what I would like this motion proposal to be made. I would be very interested in seeing how many people would vote manually for this motion proposal when it affects their own rewards.

13 Likes

Totally agree for a Forever Neuron ! But I think it implies legality problems. Currently, the forever neuron bonus is the bonus age never lost.

2 Likes

It would be a good start at least !

1 Like

The proposal is much improving the current NNS rewarding mechanism, however the updated scheme still let those early retail investors, who spent a couple of hundred per icp staked for 8 yrs in early days, they are the true icp supporters, but they still looks stupid by means tokenomics.

2 Likes

Could you please make simple Google Sheets that we can enter whatever our stakes and time? (8 years or two years locked with x amount of dissolving etc.) It can show side by side if the proposal is accepted or not. Sometimes too much information is more confusing than helping. Or post simple scenarios like eight years locked with eight years dissolving will gain or lose this if proposal accepted! If you want to keep to conversations with nerds only, great work so far.

2 Likes

I agree it would be nice if the foundation chimed in. I’ve asked many times and stopped hoping for a response. I’m not sure if I’m asking the right person, or if the right person even works at Dfinity.

2 Likes

@wpb, I agree : I think a board presentation, sendable in one picture on twitter, would be hugely efficient ! Don’t you ?

1 Like

That is crazy to think that the 8 years stakers don’t even receive 2 times the rewards of 6 months stakers. All is said. We have to solve this problem, for the wellness of the Ecosystem. Or the shorters can’t have such a power, or they can’t have such rewards.

1 Like

Does this help? Below are two examples to help you understand how to read the chart.

For an 8 year dissolve delay and 3 month neuron age, the VRY is approx 29% with the current tokenonics parameters. That same neuron would get 38% VRY immediately if the tokenomics parameters in this proposal were approved and it would grow to 51% VRY within 3 months (see Note).

For a 4 year dissolve delay and the neuron is dissolving (0 age), the VRY is approx 22% with the current tokenomics parameters. That same neuron would get 14% VRY if the tokenomics parameters in this proposal were approved. However, if they stop dissolving and leave the neuron at 4 year dissolve delay, then the VRY would grow to 28% within 6 months (see Note).

With the current tokenomics parameters (red), almost all post genesis neurons show up as noise on the lower line because the max age bonus accumulation time is so long and the max age bonus is so small. With the proposed tokenomics parameters (blue), the post genesis neurons show up as a wide VRY distribution range for each dissolve delay.

Note: Anything that is successful at increasing governance participation in significant ways will result in future rewards that are lower than the theoretical max based on governance participation today. It is always difficult to forward project the governance rewards for this reason. The further out you try to project the less accurate the projection might be. I have been staking for 6 months now and I can confirm that my governance rewards have been very consistent throughout this time period at 29%. However, this proposal is intended to move the needle by increasing governance participation.

Thank you so much but even simpler form, post into Twitter and other places, for most of us simple is better to move forward! Remember, given how advanced is the Internet Computer vs others yet the responses are nowhere near what they deserve ( Because most don’t understand the full potential)! I will vote yes!

1 Like