[Outdated] Hybrid $MIN Staking V2 [MIP 2.3] - Temp. Check

This proposal is no longer up to date. Please refer to the final version of the proposal Temp Check - $MIN Staking, Implementing a Future-Proof Model


Author: @CWSchub | Status: Temp. Check (indefinite duration)
Full Report : Hybrid $MIN Staking V2 [MIP 2.3]
V1/V2 Calculator: Data Driven Comparison
About the Author: CWSchub is a Minswap Kitty Farmer, an actively engaged community member since March 2022 and author of the $MIN Staking re-design proposal.

Introduction

The current Hybrid $MIN Staking Mechanism, since the implementation of Early Redemptions, has had a very positive impact on the total amount of $MIN tokens staked, with a ~27% increase from 433M tokens to 551M.

While this metric shows that the design is on the right path. The conducted research in the report shows that it has inherent flaws - in terms of fairness and redundancy - that will affect performance in the future (not if, but when).

The delivery of this proposal aims to mark the final stage in the development of $MIN Staking as a product. This decision needs to be governance-led and will mark the future of the revenue sharing model for Minswap.

Summary (High Level)

This MIP finalises Minswap’s hybrid $MIN-staking system.

  • Replaces the piece-wise γ formulation with a single exponentially-decaying function, enabling smooth behaviour in all regimes.
  • Eliminates the γ = 1 collapse (unfair) and γ ≪ 1 saturation seen in V1 (redundant).
  • Lifts Liquid-pool APR by up to +100 % in out-performance regimes and smooths Tiered APR spikes.
  • Introduces dynamic γ (ADA-delegation APR oracle) and cleans up LBE ā€œdust.ā€
  • Sets the foundations for future Liquid-Staking-Derivative (LSD) products.

Scope of Change

The changes proposed are divided into two categories:

  • Minor Changes: Quality of life changes which don’t affect the reward-splitting logic, but ensure long term adaptability and positive user experience.
  • Major Changes: Changes that affect the reward-splitting logic. They aim to prevent researched critical issues and ensure proven: resilience, fair distribution and higher APRs in a wide range of market conditions.

Minor Changes

Category V1 Proposed V2
γ source hard-coded 2.86% 30-day avg. ADA-delegation APR
LBE ā€œdustā€ persists forever convert sub-10% residue to ADA**

**the ultimate decision shall befall the Managing Members of the Minswap DAO or the relevant working group assigned to manage LBE dust.

Major Changes

Category V1 Proposed V2
γ formula γ=min(dAPR/sAPR, 1) (piece-wise) γ ∊ [0,āˆž) (no longer clipped)
Liquid APR γ Ā· tAPR (flat when γ ≤ 1) tAPR·ψ(γ, exp{-γ}) (monotone, bounded)
Tiered APR tAPR·[1+(1-γ)·r] tAPR·ψ(γ,r,exp{-γ})

Background and Motivation

The Hybrid $MIN-staking program (launched 29 Oct 2024) and has seen a 27% increase of 433M to 551M staked $MIN at its peak but exposed two design flaws:

  1. γ = 1 collapse – Tiered & Liquid APRs become identical, yet Tiered stakers still suffer an early-withdrawal fee.
    This represents an unfair distribution of rewards relative to withdrawal penalties.

  2. High-performance saturation – When MIN out-performs ADA, Liquid APR locks flat while Tiered APR skyrockets, draining liquidity to the Tiered option.
    If all the capital flows to the Tiered option, then there is an element of redundancy which needs to be addressed by making Liquid Staking more attractive.

Community feedback during the earlier MIP 2.2 temp-check highlighted similar pain-points around reward predictability and negative loops. These flaws can be explored in the following V1 Hybrid $MIN Staking Calculator.

1. Collapse of APR(LS) and APR(TS) to equivalence

As has been described earlier, the current parametrization of γ=min(dAPR/sAPR, 1) causes both options to ā€œcollapseā€ to equality when γ=1. This affects Tiered Stakers by far, as they are still subject to early withdrawal fees.

This issue becomes worrying when many different combinations of $MIN price, total $MIN Staked and daily $ADA rewards can cause this to happen:

2. High-Performance Saturation to Tiered Staking

As can be seen in the following image, when γ<<1, the Tiered Staking option exceedingly outperforms the Liquid option, due to it remaining constant as rewards increase.

Therefore, the proposed solution must:

  1. Cause APR(LS) to scale smoothly, removing the ā€œcollapseā€ when γ=1.
  2. Eliminate the behaviour of constant APR during periods of high performance.

Proposed Solution

IP disclaimer

Due to IP concerns, the full formulaic approach will not be publicly disclosed unless the proposal passes with Option B being the winner and the author will relinquish any copyright claims to the released documentation. This protects the author in case of a failed proposal.

A New Formulaic Approach

For a formal definition of V2’s formulaic approach, please see section 4 of the published report.
This new approach utilises exponentially decaying functions of γin order to overcome the forced piecewise definition used in V1. By implementing this new logic, both flaws are resolved elegantly.

The following graphical representations showcase the asymptotic convergence between Liquid Staking and Tiered Staking plotted for V2 (right charts) compared to the current, V1, plots (left charts):

Financial Impact Using a Model for Capital Flow

Section 6 delves into how $MIN holders would behave given different scenarios and how, in turn, the resulting ā€œexpectedā€ APRs for both the Liquid and Tiered Staking options would look like.
The entire model is based on utilising a ā€œminimum break-evenā€ timer Tmin where $MIN holders are able to unlock their rewards from the Tiered option while staying at least as profitable as if they had chosen the Liquid option.

The following plots represent the Expected APRs according to a defined break-even timer Tmin.


Oportunity Costs

The report analyses the financial impact V1 had on its predecessor, Boosted $MIN Staking. It identifies that implementing V1 could have saved $MIN stakers around 200k - 700k $ADA.

Analogously, considering V2 performs a change in the reward-splitting logic, the opportunity costs associated to not implementing V2 are in line with the changes it seeks to implement

Opportunity Costs of not implementing V2 for Tiered Stakers at γ>1

The following table provides an estimate Oportunity Cost on a monthly basis for $MIN Stakers who have selected the Tiered Staking option if the transition to V2 didn’t happen at current prices (1$MIN=0.0346$ADA) and total $MIN Staked: 551M $MIN.

Opportunity Costs of not implementing V2 for Liquid Stakers at γ<1

The primary motivation regarding implementing V2 for Liquid Stakers is to provide better returns when $MIN Staking enters into high performance market conditions. The following table provides an estimate Oportunity Cost on a monthly basis for $MIN Stakers who have selected the Liquid Staking option if the transition to V2 didn’t happen at current prices (1$MIN=0.0346$ADA) and total $MIN Staked: 551M $MIN.

Utilising the same method as before, but for γ=0.292.

Future Developments

V2 is proven to be resilient to all kinds of market conditions. Notice that all graphical representations of both V1 and V2 have the x-axis scaled logarithmically. Furthermore, the x-axis doesn’t represent time, but a relative performance parameter. Therefore, the entire approach in this report has evaluated market conditions two orders of magnitude apart.

Once the conditions of smooth behaviour and resilience under market conditions orders of magnitude apart have been met, it is possible to build derivative financial instruments on top.
The perfect target for such financial instruments should have an indefinite investment horizon and should be fully liquid. It is no coincidence that V2 has been developed in order to support Liquid Staking Derivatives.

Enabling Capital Efficiency Using Liquid Staking Derivatives (LSD)

Implementing an LSD system would improve capital efficiency, making the staking mechanism and the $MIN token a much more attractive investment option. This could be facilitated via two different technical implementations:

  • Fungible LSDs
  • Non-Fungible LSDs

For more information, please refer to the (slightly out-dated, but still relevant) following article by CoinGecko:

Conclusion

The current state of $MIN Staking is, unfortunately, not a sustainable one in the long term. The only real solution to offering a fair reward-splitting logic is by attacking the APR formulation.

Therefore, the author has reached the conclusion that $MIN Staking will only be sustainable in a future where:

A) There are no locking periods and there is only one Liquid $MIN Staking option.

Or:

B) A non-piecewise definition for the γ parameter is provided, enabling smooth transitions between low performance to high performance and Tiered Staking beats Liquid Staking at all times. V2 offers an elegant future-proof solution.

Or:

C) $MIN Staking is discontinued - Fee Switch redistributed back to DAO Treasury

Proposed Compensation

As per described in section 9 ā€œScope of Workā€, the proposed compensation is:

In order to establish a fair balance between an hourly rate and the value associated to the financial impact of implementing V2. The author proposes a fixed payment and a variable one based on performance after implementation of V2, to be honoured by the DAO treasury:

4.a 10000 $ADA upon approval and release of full formulaic description.

4.b image

In the event of the winning voted options are 1) or 3)

4.c 15000 $ADA upon approval so as to honour the dedicated time.

4.b describes 0.5% of total daily $ADA (excluding LBE) rewards over the course of 9 months distributed by Minswap’s MIN staking, multiplied by a performance coefficient. This coefficient is to be determined utilising the 60-day moving average of total $MIN staked to prevent malicious behaviour.

Voting Options

A) Transition to fully Liquid Staking. (4c Compensation)
B) Implement V2 hybrid $MIN Staking. (4a + 4b Compensation)
C) $MIN Staking is discontinued. (4c Compensation)

7 Likes

This is incredible work and should be implemented so as to strengthen the integrity of staking for the long term, with the caveat we actually want to keep it for the long term. There are good arguments for why staking is detrimental to the min token and the protocol in terms of forfeited volume on the min token. In this way the options between B and C create a tension and I’m not quite sure why C was tacked on at the end. So I’m going to ignore C since there is no argument for it in the proposal. Therefore, based on the argument constrained by the proposal, B should be implemented as a natural update to the initial conceit of liquid staking, making it more resilient and sustainable for the long term.

3 Likes

b option seems to be the best option

3 Likes

Thanks Contra!

The aim is to transition to a ā€œfinal formā€ for the staking mechanism. The reward-splitting logic should not be tampered with anymore.

Option B is the main focus of the proposal because proof is required to argue for it.
On the other hand, options A and C are straightforward and necessary to complement the proposal in terms of viable alternatives.

2 Likes

i got many final form feelings from the proposal lol. and for the sake of it existing in the first place it makes sense to refine it as much as possible

3 Likes

Amazing work, regarding the main points (ignoring synthetic assets since I think that is another ordeal) I believe what you are proposing is 100% a technical improvement, I however don’t dislike some of the features you are trying to solve, like the lost ā€œopportunity costā€ by early redmptions of long term stakers. I understand early redemption is costing them money, but it is helping non redeemers have a higher APR. I am okay with that. I believe that putting both things as costs in the proposal and not saying that its actually going to the rest of stakers its not fair.

So to sum up, I think its a very nice proposal, that improves the product technically, I however am not sure I like how the costs of this imporvement are beared. I am okay with long term stakers that stake for 9 months having more benefits, I think its the whole point of locking. This is however a personal opinion and I do believe its fair to ask to all min holders about it. I however think the counterpoints are not clear in the proposal, and that it could bias a vote.

My second main concern is if this is really needed right now. MEaning this implementation and whole governance process will tie up some resources, and personally I think staking has not been an area where the DAO has received any complaints at all, I understand this is an improvement, I am not sure it should be done now, since as of today, there has not been many issues with it as far as I know. I rather use the resources to complete ongoing developments like the integration with a lending platform that should ease the use of leverage (which I see as a multiplier of volume that could be extremely beneficial for minswap).

2 Likes

Thank you @calles for your thorough feedback!

I’ll start with your last comment. Yes, I agree it is currently not an urgent topic. Nonetheless, I worry the flaws I point out might catch people by surprise (something I’d rather avoid). In any case, I have no intention of going official yet.

Secondly, I see your point about early redemptions. I would agree that a 100% forfeit could benefit the rest of stakers. But then the rate of unstake before maturity would also likely drop significantly. So what’s better: one large dump of forfeited rewards or multiple smaller ones?
Additionally, ā€œliquifyingā€ MIN staking is proving to attract users who were previously hesitant. In the long run, early redemptions will be a consistent source of increased APR, rather than a sporadic event.

Lastly, I do mention some ā€œdownsidesā€ to V2 in the full report (in terms of areas: see figures 22, 23 and 24) but the ā€œpositive areasā€ in overall terms outweigh the ā€œnegative areasā€.
The most notable counterpoint is that there is a point from which V2 Liquid staking will underperform V1 (see fig below) in order to ensure Tiered Staking APR remains higher.

I used a simple numerical integration to calculate the total area of the ā€œRelative Differenceā€ curve (right graph). This yielded a 1.2408 positive/negative ratio. In other words, the good outweighs the bad.

So while there are sporadic areas where V1 might beat V2 in terms of ā€œraw APRā€; in my opinion, the overall impact of V2 far outweighs this.

3 Likes

what do you mean when you say " Non-Fungible LSDs" ?

1 Like

LSDs are Liquid Staking Derivatives. Simply put, it is a system that enables capital efficiency over a yield-generating product. So, currently, when you stake MIN, it’s there and you can’t do anything with it. But imagine if you received some sort of voucher that says ā€œthis is equal to 10.000 staked $MIN tokensā€. You can now use that voucher as collateral, borrowing against it in lending platforms.

This voucher can exist in two shapes:

  1. Fungible tokens: when staking, you mint an amount of $stMIN which can be burnt to redeem your staked $MIN. These tokens would make it easy to create lending markets on Liqwid, for example. Or you could open a CDP on Indigo with the $stMIN.

  2. Non-Fungible voucher/tokens: staking $MIN mints an NFT with the relevant metadata that keeps track of the amount of $MIN you’ve staked. In order to redeem your $MIN + rewards requires burning the NFT. This NFT can be used as collateral in lending platforms like Fluid Tokens.

4 Likes