Tokenomics Proposal [Community Consideration]

Nice work!! Kudos to the team.

2 Likes

I almost forgot! The dashboard team also just shipped a neuron page as well. Here is an example: Internet Computer Network Status

2 Likes

This is great Diego, thanks to you and the team!

Could you explain what “undisbursed” rewards are? Are these voting rewards that were locked up in neurons or is this ICP that is just sitting in a fully dissolved neuron?

2 Likes

I believe are NNS rewards which have are either sitting on the neuron or unrealized % maturity that’s don’t exist as actual ICP yet.

@Dylan did I get that right?

6 Likes

Undisbursed rewards exist as maturity only, and have not yet been “minted” by either spawning the rewards into a new neuron, or merging that maturity back into the stake of the neuron that accrued the maturity. Put another way, undisbursed rewards have not yet increased supply, since that only happens when rewards are minted by one of the above two means.

5 Likes

Great, that helps frame it a bit. Thank you

1 Like

Do you know if this % return assumes continuous compounding or annual? I know it says annualized but I just want to confirm:

https://dashboard.internetcomputer.org/governance

2 Likes

Thanks john! Really appreciate it

@jwiegley explained it perfectly.

3 Likes

No, there is no compounding. The calculation takes the “trailing 24-hour” return (i.e., based on the most recent daily reward event) for a “neuron with an age of zero” and multiplies that by the number of days in a year to calculate an “annualized return”.

3 Likes

The 28.4% is the uncompounded return, but it assumes that a lot of factors hold constant and really only applies (and even then, just approximately) to the first year.

3 Likes

Great. So if I’m hitting “merge maturity” every day, I should in theory be earning more than 28.4% annually?

That’s correct, you should earn a little bit more in that case, my own modeling suggests around 31%.

5 Likes

Just note that if more people start to stake ICP and total voting power goes up as a result, then your slice of the rewards pie will become smaller and your returns will go down. Hence the disclaimer under the slider. Over the past few months, the estimated annualized return for an 8-year dissolve delay hasn’t gone down by much, but it’s always possible that it will go down.

6 Likes

Yes, exactly what @Dylan said; in fact, competing stakeholders is the largest factor in return, so the more popular it becomes to do so, the smaller APYs will get.

5 Likes

Supply and demand. Supply and demand.

4 Likes

Right, the benefit of people staking is that tradable supply drops. So we should expect a rough inverse relationship between returns from interest and returns from the token value.

As interest rates go down, token value should increase since when they’re staked they can’t be traded.

In this respect it’s very similar to bonds: A bond offers high rates of interest when prices are low, to attract investors; but as people bid up the price of the bond, it’s rate goes down, because it doesn’t need to incentivize anymore.

3 Likes

I’m curious to see how it will actually play out. Demand can still fluctuate wildly and, given that neuron staking funds are locked, it will leave supply relatively stable.

E.g. if the market drops again, demand for icp might drop, but the total amount of staked icp will remain roughly the same (at most, ticking down total vote power at a rate of 1 second worth of voting power per account per second if everyone starts dissolving simultaneously. That’s not real fast.) This is unlike a bond, which can be sold on the market at any time.

Does this mean we should expect fewer flash crashes out of ICP?

1 Like

@diegop @jwiegley @Dylan

Great job with the dashboard updates! I really appreciate seeing the node provider, dispersed, and undispersed rewards displayed so clearly. My hope is that the community will stop focusing so much on node provider rewards and start focusing more on inflation created by dissolving neurons as well as the unminted ICP that is sitting in neurons as earned maturity. Those are, by far, more important to the inflation / deflation story than node provider rewards.

I really hope the IC community will think more about how to incentivize staking because it would address the concerns about liquid ICP inflation much faster than burning ICP to pay for compute power. The time lapse dissolve of neurons has an incredible deflationary effect, especially long dissolve delays, because it is irreversible except for the passage of time. The #8yeargang has proven to be highly committed because over 97% of those neurons are not dissolving, which is almost almost equivalent to burning ICP. Every time that dissolve delay slider moves up, it has a deflationary effect. We spend too much time as a community thinking about ICP tokenomics in the same paradigms as other blockchains when the reality is that ICP tokenomics add completely new dynamics, especially the dissolve delay bonus and (potentially) the age bonus.

I know there are other features of the dashboard that are being worked on now, but one that I think would be helpful to the community is a breakdown of neurons by dissolve delay and dissolve status. More specifically, it would be helpful to see that breakdown in terms of both total ICP and total voting power. Taking this type of analytics a bit further, it would be very enlightening to easily see how this data changes over time. Anyway, I’m sure you have some great new features in the works and I’m looking forward to seeing them as they roll out. Once again, great job!

8 Likes