Mango token lockup and voting power

Over the last months many people (microwavedcola, Adrian, SebastianBor, Maximilian, me) have been working on getting MNGO token lockup working with DAO governance. It’ll be used voluntarily by Mango token holders and also supports vesting grants of locked tokens.

What are the benefits of voluntarily locking up Mango tokens? The first answer is: an increase in voting power. We want people who lock their tokens for significant amounts of time to have a stronger impact on DAO decisions because their incentives are more likely to be aligned with the long-term flourishing of Mango.

What I’d like DAO members to discuss and agree on are the concrete factors for the voting power increase from token lockup. To give you an idea of the available settings, here’s my starting proposal:

  • unlocked tokens keep the same voting power as before (1 token = 1 vote)
  • tokens can be locked for up to five years
  • voting power is doubled at 5 years lockup (and shorter lockups scale linearly: 1y=1.2x, 2y=1.4x etc)

For example: if you locked up 10.000 MNGO for five years, you would have the effective voting power of 20.000 unlocked MNGO.

Note that voting power is computed dynamically based on the remaining lockup duration: A five-year lockup after 4 years would have the same voting power as a fresh one-year lockup.

In a way, this is a follow up to Multicoin Proposal to Implement Governance by Token Lock and Commit-and-Reveal

6 Likes

I think 2x with 5y lockup is a reasonable trade-off between incentivizing commitment to the protocol vs too much power to parties which have excess cash to buy tokens for long term without any impact on own living circumstances (as pointed out in the previous forum post)

4 Likes

I think maximum of 2x vote multiplier and maximum 5 years of lockup seems best option to start with we can always change it later to higher or lower values.

I don’t like that my voting power declines as time goes on if I lock for 5 years. What was the reasoning for that decision?

Is it technically possible to make it so a 5 year locker will get the full power for the whole 5 years?

The idea behind decaying vote weight is that we want to scale people’s voting power based on the time horizon they’re voting for. People in the last month of a 5y lockup period don’t have the same incentives as the ones in their first month of lockup.

If you want to have maximal lockup power without having to worry about decay, we’ve implemented a mode called “constant lockup” that works differently from the normal cliff lockup: In that mode your tokens stay locked up permanently and you need to manually trigger your offramp cliff if you want to withdraw your tokens. Vote power decays only during the unlock period.

Example: Lock up 10.000 MNGO with constant lockup and 5y offramp. You’ll have 2x power forever; until you say “start my 5y unlocking period now”. Then vote power starts decaying and after 5 years you can withdraw your tokens.

2 Likes

I like this better for myself personally especially if fee distributions will be in proportion to voting power.

I agree people with only 1 month left have different incentives. This design does make sense as long as people have the option to keep being locked for longer. Consider Alice locks for 5 years and has an intention to be locked for 20 years. Bob comes in after 3 years and locks for 5 years. Under your default implementation, at year 3, Bob has more power, but seems intuitive that Alice should have more power both because she has been invested longer and probably knows more but also because she intends to be locked for longer. We’ve agreed we don’t want to give too much power to longer lockups, but people with that intention should at least get equal power.

1 Like

One could argue that if alice has an intention to lock for 20 years then why doesn’t she do that from the start?

Anyway, after 3 years, if she realises that bob has more power in the moment, and wants to have the same intensity, then she could either renew her cliff lockup to 5 years or switch her lockup to kind “constant” (from another lockup e.g. cliff, which had decaying power) whenever she wants, which gives her maximum power.

ok I was confused, I thought the constant method wasn’t in the deployment. If it is then no objections from me

Constant lockup is implemented and available in the ui, we support

  • cliff
  • constant
  • vested (monthly and daily, though daily isn’t exposed in the ui)
  • any lockup can be changed to be stricter (go from vested → cliff → constant, or to a longer duration; not exposed in ui yet)
  • technically lockups can be for longer than 5y, though there’s no vote power benefit to it (not exposed in ui)
4 Likes

Does MNGO force locked through a grant get the same extra voting power as someone who opted into locking?

Yes, voluntary locking and locking-via-grant are treated the same way right now. The only difference is that grants are eligible for clawback, while normal lockups aren’t.

1 Like