GMX: Increasing Buyback & Distribute Fee Coverage from 27% to 90%

[gmsolq]

Background:

GMX: Transition from “Buyback ETH and Distribute ETH” to “Buyback GMX and Distribute GMX”

Background In revisiting my “Buyback and Distribute” (BB&D) proposal from last year, Buyback GMX: A Proposal for Reward Distribution Optimization - Buyback GMX: A Proposal for Reward Distribution Optimization - March 2023, and considering the recent introduction and adoption of BB&D by Gains Network, it has become clear that it is necessary to reassess and reintroduce this proposal. Additionally, feedback from the GMX community, especially the insights and discussion with Woju , and a detailed …

Over the past 30 days, GMX’s buyback has yielded impressive results:

• Total GMX Bought Back: 103,764 GMX (5% of circulating supply)
• Total Value: $3,341,200

While favorable market conditions make it challenging to attribute short-term performance solely to the buyback, it’s undeniable that a fundamental shift in supply and demand will have a significant long-term impact.

Current GMX Fee Allocation:

• Fees Generated Last 30 days: $9.432m
• Allocated for Buybacks: $3.344m
• Fee Coverage in Value Retention: 30% for GMX V1 and 27% for GMX V2

The value retention rate is estimated at 1 - 8,949 / 103,764 = 91.38%, meaning that this percentage of GMX stakers choose to hold rather than sell.

Remaining 73% of Fees Allocation in GMX V2:

• GMX Treasury: 10% (used for essential expenses like tech development, operations, audits, and bug bounties which are most settled in USDC)
• GM Pools (GM LP): 63% (auto-compounded)

Given the critical nature of the GMX Treasury’s expenses, we only focus on discussing the possibility of using the 63% of fees allocated to the GM LP for Buyback & Distribute (BB&D).

Current GMX LP Structure

• GMX V1 (GLP):
  • Rewards are obtained similarly to staked GMX.
  • A proposal was approved to convert ETH/AVAX rewards into GMX rewards.
• GMX V2 (GM LP):
  • Uses an auto-compound method.
  • Fee earnings are directly added to the GM Pool.
  • Advantages include simplicity, easy integration, and tax benefits (no income tax issues from claiming rewards).

Proposed Solution to Expand BB&D to GMX V2

• Technical Implementation Steps:
  • Reduce GM LP’s fee distribution from 63% to 0%.
  • Increase GMX Treasury’s fee distribution from 10% to 73%.
  ( Note: This is a proportional adjustment requiring minimal technical development.)
  • GMX Treasury uses the additional 63% of fees to execute buybacks through the existing buyback contract.
  • GMX Treasury airdrops the bought-back GMX to GM LPs.
  ( Note: Similar to the ARB incentives program; tools for this distribution already exist.)

This is a technical solution with minimal implementation costs for understanding purposes which does not represent the final implementation plan.

Pros

• Significant Increase in Buyback ( estimated on last 30 days fee):
  • Fee Coverage Increase: From 27% to 90% (a 3.33x increase).
  • Estimated Buyback Value: From $3.344m to $8.489m monthly.
  • Estimated Buyback Amount: From 103,764 GMX to 345,534 GMX monthly.

• Enhanced Value Retention:
  • The value retention ratio among GMX stakers is already high (91.38%).
  • Applying BB&D to GM LPs could further increase overall value retention.
  • Theoretically, it can be demonstrated that this adjustment, like BB&D, has a strictly positive impact on value retention.

• Market Stability and Risk Mitigation:
  • Continuous buybacks provide a strong safety net against large sell-offs.
  • E.g., if a top 5 whale liquidates 100,000 GMX (black swan moment), the negative impact can be hedged within a month under the current BB&D coverage but only 12 days under the optimized BB&D coverage.

• Predictable and Gradual Market Influence:
  • Buybacks offer gradual changes based on business data, unlike unpredictable whale movements.

Cons

• Tax Structure Issues:

  • Concern: Changing from auto-compound to claimable rewards may result in additional income tax liabilities for some GM LPs. E.g., a 30% income tax could reduce actual earnings.
  • Potential Solution (proposed by community contributor Jayski): Deconstruct GM into Two Tokens. GM represents assets within the GM Pool, affected by trading profits/losses and asset value changes. feeGM accumulates fee earnings over time, with an initial value of zero. This solution can help LPs avoid additional income tax.
  • Considerations: Despite a 30% tax cut for certain LPs, GMX remains one of the most competitive platforms for on-chain real yield, as evidenced by the latest screenshot showcasing its leading position among on-chain protocols. However, this proposed solution increases complexity and may delay implementation. If a simpler method cannot be identified, it may be better to proceed without this adjustment. If the impact on some LPs is unavoidable, we will further explore and address these potential effects.

• Rewards in High Volatility:

  • Concern: LPs might find earnings less attractive if rewards are paid in GMX due to its volatility.
  • Mitigation:
  For liquidity impact: GMX’s liquidity can handle the scale of LPs swapping GMX back to Pool Assets with minimal price impact. E.g., the largest LP in GMX V2 is in the Single-Sided BTC Pool, holding 10,248,906 GM, worth nearly $15 million. At an APR of 20%, the annual fee earnings are about $3 million, equating to weekly earnings of $57,534. If an LP is sensitive to fees and swaps the $57,534 from GMX back to BTC weekly.
  Volatility Impact: Given GMX’s good liquidity, its volatility isn’t excessively high. The volatility applied to $57,534 is negligible relative to the total value of $15 million.
  User Acceptance: GMX stakers, as crypto users, are highly accustomed to converting between assets like GMX, ETH, and AVAX within similar liquidity scales, supporting this approach.

• Impact on Downstream Integrations:

  • Concern: Changes might affect downstream integrations relying on the current structure.
  • Mitigation: Since the underlying structure remains the same, integrations that managed during the ARB incentives period can adapt to GMX rewards similarly.

• Other Potential Impacts on LPs:

  • Return Rates: Auto-compounding yields higher returns due to APR vs. APY differences.
  • Pool Liquidity: Auto-compounding keeps fees within the pools, enhancing liquidity.
  • Consideration: The benefits of increased BB&D coverage may outweigh these concerns, given the potential for higher value retention and market stability.

Game Theory Scenario

• Assumption

  A 30% LP withdrawal due to tax concerns.

• Analysis

  Since this affects the LP side and not the trader side, even if 30% of liquidity is withdrawn, our current utilization has enough elasticity to handle traders’ needs. With trader demand maintained, the withdrawal of 30% of existing LPs would increase APR, attracting new LPs and reaching an equilibrium. GMX also has the historical performance of GLP as evidence that BB&D is feasible for LPs.

  Using the SS ETH Pool as an example, the current TVL is $58.76m, the liquidity cap is $31.23m, and the APY is 33.82%. Additionally, the TVL is rapidly increasing, which means we can assume that a 33.82% APY is an extremely attractive real yield for ETH in the market.
  In an extreme scenario where 30% of LPs withdraw, the TVL would drop to $41.13m, the liquidity cap would remain at $31.23m, and the APY would rise to 43.97%. This would inevitably attract GM LPs to supply more ETH to the SS ETH Pool, bringing the APY back down to the minimum level of 33.82%.

  Of course, this is a dynamic process. As long as utilization is not at 100%, we have sufficient flexibility to allow GM LPs to iterate and adjust. If utilization becomes very high or even reaches 100%, capital efficiency will be pushed to the limit, and more GM LPs will flow into this specific pool.

  Through this game theory scenario, we can effectively alleviate concerns about certain LPs withdrawing liquidity. To further understand this, a counterexample could be provided: directly doubling the trading fees. Theoretically, this would double the returns, but in reality, it would result in a significant loss of traders, creating a lose-lose-lose situation for traders, LPs, and GMX alike.

Conclusion

• From the LP’s perspective, this proposal seeks to optimize the method of value retention without altering the economic model, aiming to increase the fee coverage of Buyback & Distribute (BB&D) from 27% to 90% without affecting the overall earnings of LPs.

• From the GMX holder’s perspective, a symmetrical option is also presented to the community for consideration—directly adjusting the fee distribution percentage. However, it’s important to note that any changes to the economic model could potentially have long-term impacts on GMX’s price performance. The benefits of an increased percentage and the negative effects of altering the economic model would ultimately be borne entirely by GMX holders.

• Overall, LPs and GMX holders have a mutually dependent relationship. The success of LPs and GMX holders is interconnected. If LPs consider the interests of GMX holders, better performance of GMX would result in greater benefits for LPs. Conversely, if GMX holders consider the interests of LPs, the protocol will achieve more stable growth, and traders will enjoy a better experience. If both sides refrain from making decisions solely based on maximizing their own interests, it’s very likely that a win-win balance can be achieved.

Voting Option

• Option 1: Keep the fee distribution to GMX:GM LP at 27:63 unchanged, expanding BB&D’s fee coverage from 27% to 90%.
• Option 2: Adjust the fee distribution to GMX:GM LP from 27:63 to 40:50, adjust BB&D’s fee coverage to corresponding 40%.
• Option 3: Oppose both options above.

This might be one of the most impactful and important proposals in GMX’s history.
You matter. Your vote matters. Let’s make GMX greater together!

[gmsolq]

The most common feedback about GMX is that it’s “undervalued.”

To be considered undervalued, there must first be inherent value. Imagine a protocol generating $10 million in monthly fees, $120 million annually, distributing a cumulative $350 million in fees, and achieving a TVL of $750 million, yet having a market cap of only $300 million. Clearly, this represents a misalignment with reality.

Enhancing value can be achieved through expanding markets, increasing open interest (OI) capacity, lowering trading costs, and pursuing cross-chain expansion. Collectively, these efforts can be referred to as the GMX Spear.

On the other hand, optimizing the economic model by removing MP, implementing BB&D (Buyback and Distribution), and expanding the fee coverage of BB&D improves value retention. These measures collectively form the GMX Shield.

Focusing solely on changes to the economic model as a means of improvement, without solid business development, is nothing more than self-consolation.

Conversely, if the business performs well, yet the challenges of being “undervalued” are ignored, assuming that the market is simply wrong and passively waiting for correction, such a stance is incredibly arrogant.

The market is always correct. Growing the pie and distributing it more efficiently are equally important. Any claim that prioritizing the economic model over business development is the right path, or that business development should take precedence over economic model optimization, is fundamentally flawed. We are fully capable of excelling in both areas.

Every proposal I make is aimed at strengthening both the GMX Spear and the GMX Shield, with GMX at the absolute core, to explore and achieve its greatest potential.

Undervaluation is fixable!
Glory to GMX!

[Saulius]

As DAO delegate I completely disagree with such a proposal. I think it’s dangerous even to talk about it. More comments later…

[Jonezee]

@gmsolq

• Option 1: Keep the fee distribution to GMX:GM LP at 27:63 unchanged, expanding BB&D’s fee coverage from 27% to 90%.
• Option 2: Adjust the fee distribution to GMX:GM LP from 27:63 to 40:50, keeping BB&D’s fee coverage at 27% unchanged.

I think there’s a mistake that slipped in here.

That aside: a VERY provocative proposal, which is sure to have some existing liquidity providers worried. It’s very well laid out, and I appreciate the clear list of pros and cons provided for the various options. But do not underestimate those Cons; some may be more impactful than you currently believe.

Look forward to an intense discussion.

[gmsolq]

@gmsolq

• Option 1: Keep the fee distribution to GMX:GM LP at 27:63 unchanged, expanding BB&D’s fee coverage from 27% to 90%.
• Option 2: Adjust the fee distribution to GMX:GM LP from 27:63 to 40:50, adjust BB&D’s fee coverage to corresponding 40%.
• Option 3: Oppose both options above.

Thanks for pointing out Jonezee, it’s fixed.

[Mustard]

I propose an Option 4: Allow for two types of GM pools to coexist, as discussed by Q.
One would be accumulating, and the other distributing.

The accumulating pool would start receiving a smaller percentage of fees than it currently does. For example, instead of 63%, it would receive 60%, with the remaining 3% used to buy back and distribute to the distributing pool. The distributing pool would then receive 63% + 3% from the accumulating pool, all in bought GMX.

This approach allows current LPs to continue with their existing strategies while also incentivizing people to transition from the accumulating to the distributing pool. Additionally, it encourages buybacks to take place.

The exact number of % for each type of GM pool can be discussed of course.

If only the other 3 options are present I would vote for 3.

[gmsolq]

This is a proposal, and the most important purpose is to gather opinions and feedback. We hope to hear the voices of both GMX holders and LPs—speaking from their own perspectives, from the perspective of the other party, and from the broader perspective of GMX as a whole.

A proposal does not represent the final decision, and even the voting options can be adjusted. Approval and rejection are both possible outcomes, as this is a process of democratic discussion and decision-making.

[hermit]

Agree with Saulius.
This proposal if passed could shake the foundation of gmx, in the worst case, gmx could die, because LP is the foundation of gmx, this proposal could cause us to lose a lot of LP.
In addition, we should not invest too much energy in deliberately pushing up gmx token price, and it is unrealistic to constantly push up the price by changing tokenomics.
gmx price depends on the business, and as the business grows, the price goes up, and the only important thing for us is to focus on growing the business.

[gmsolq]

@hermit

This proposal if passed could shake the foundation of gmx, in the worst case, gmx could die, because LP is the foundation of gmx, this proposal could cause us to lose a lot of LP.

For Option 1, the LP’s earnings do not actually decrease; they remain unchanged. With earnings staying the same, why do you believe this would lead to losing a significant number of LPs or even cause GMX to die?

With earnings unchanged, the concern LPs might have is the potential decrease in the quality of those earnings (i.e., the liquidity of rewards). In this scenario, even for the largest LPs, GMX has sufficient liquidity to handle these rewards, allowing LPs to convert them into BTC, ETH, USDC, or any other final output token they prefer.

@gmsolq

For liquidity impact: GMX’s liquidity can handle the scale of LPs swapping GMX back to Pool Assets with minimal price impact. E.g., the largest LP in GMX V2 is in the Single-Sided BTC Pool, holding 10,248,906 GM, worth nearly $15 million. At an APR of 20%, the annual fee earnings are about $3 million, equating to weekly earnings of $57,534. If an LP is sensitive to fees and swaps the $57,534 from GMX back to BTC weekly.

In the analysis of the game theory scenario, the potential iterations and flexibility of LPs have been thoroughly examined.

@hermit

In addition, we should not invest too much energy in deliberately pushing up gmx token price, and it is unrealistic to constantly push up the price by changing tokenomics.
gmx price depends on the business, and as the business grows, the price goes up, and the only important thing for us is to focus on growing the business.

First of all, as I mentioned, we do not disregard business growth. The solutions I have proposed to address issues related to business growth far outnumber those focused on value retention: ARB incentives, single token pool, GMX market, net OI solutions to unlock OI capacity, post-position price impact, virtual orderbook, and even the introduction of GMX-Solana, among others.

However, I absolutely disagree with the notion that simply excelling in business growth is enough. We are fully capable of advancing both aspects simultaneously. Offense and defense are equally important.

This is precisely why I have also pushed for ending MP, introducing BB&D, and now this proposal to expand BB&D fee coverage.

If, as you suggest, “GMX price depends on the business, and as the business grows, the price goes up, and the only important thing for us is to focus on growing the business,” then please explain why GMX’s business metrics are at ATH while its market cap is near ATL. How is it that a top-ten protocol in terms of fee-generating has a market cap ranking over 200?

Undervaluation is a problem—and it is a solvable problem. But waiting and fearing change are absolutely not the ways to solve it. As for price performance being determined solely by supply and demand, if a measure affects the supply-demand relationship, then it will impact price performance. The data and rationale behind why BB&D works and why expanding BB&D is effective have been detailed above.

[FallingSands]

Q,
You are an absolute champ and I’m constantly impressed with your ceaseless effort to drive the GMX protocol and token price through the gates of Valhalla.

That being said, this is too much, too fast. BB&D has been running for only 30 days and, in your own words, has yielded impressive results. We should wait at least a year to adjust this mechanism. I’m not saying the protocol should be ossifying this early, but I am saying we should strive to foster a reputation of stability and reliability. I think this change could spook a lot of LPs.

The only way I would support adding a buy-back feature to the GM pools this quickly would be if, by default, NOTHING changed for LPs:

1. fees continued to auto compound by default
2. no need for LPs to take any action to maintain the current fee mechanics
3. no new accumulate/distribute pools and complexity
4. no changes to fee proportions

BUT a switch could be added to each pool giving individual LPs the OPTION of earning their fees as GMX through a BB&D mechanism.

I think most LPs would choose not to activate this option, but if even 10% do, that’s adding nearly a third to the current BB&D volume.

Otherwise, we should continue gathering data on our current BB&D scheme and see how things stand in 12 months.

Lets not forget:
5% of circulating supply was bought back in the first 30 days, before GMSOL has even launched.

[babloo]

I oppose. Main progress from v1 to v2 is auto-compounding. This plan removes it. I think 27-63-10 ratio is good and I would prefer to keep it as it is. Maybe 40-50-10 but definitely not 90-0-10 for sure.

[gmsolq]

First, I respect the ratios and proposals you prefer.

Just to add, the most important innovations of V2 compared to V1 are achieving risk isolation, which is the foundation for supporting multiple markets; and introducing funding fee, which is the basis for maintaining OI balance and minimizing LP risk.

[tx_analysor]

Initially, I wasn’t a strong supporter of your first proposal regarding BB&D, but I must admit it has proven to be a success. However, applying the same concept to LP fees seems like a misguided approach. While it would undoubtedly drive short-term demand for GMX and potentially lead to a significant price increase, such a surge driven purely by BB&D mechanisms is unlikely to be sustainable in the long run and could ultimately backfire.

The primary concern is that this change would fundamentally alter the GM structure, particularly by disrupting the autocompounding mechanism. The temporary price pump would come at the expense of GMX’s TVL growth and compromise the simplicity of the GM and GLV pools, which are currently designed for ease of use—allowing investors to simply add capital and let it grow over time.

As you said, we need both a good economic model and effective business development, but I believe the current BB&D model is sufficient in addressing the economic model. If the concern is that GMX is undervalued, we should focus instead on developing the GMX product to attract more TVL/OI and invest in initiatives to expand its community, such as GBC and gmSOL.

In my view, the cost of this strategy outweighs its potential benefits by far.

[gmsolq]

@FallingSands

Q,
You are an absolute champ and I’m constantly impressed with your ceaseless effort to drive the GMX protocol and token price through the gates of Valhalla.

That being said, this is too much, too fast. BB&D has been running for only 30 days and, in your own words, has yielded impressive results. We should wait at least a year to adjust this mechanism. I’m not saying the protocol should be ossifying this early, but I am saying we should strive to foster a reputation of stability and reliability. I think this change could spook a lot of LPs.

The only way I would support adding a buy-back feature to the GM pools this quickly would be if, by default, NOTHING changed for LPs:

1. fees continued to auto compound by default
2. no need for LPs to take any action to maintain the current fee mechanics
3. no new accumulate/distribute pools and complexity
4. no changes to fee proportions

BUT a switch could be added to each pool giving individual LPs the OPTION of earning their fees as GMX through a BB&D mechanism.

I think most LPs would choose not to activate this option, but if even 10% do, that’s adding nearly a third to the current BB&D volume.

Otherwise, we should continue gathering data on our current BB&D scheme and see how things stand in 12 months.

Lets not forget:
5% of circulating supply was bought back in the first 30 days, before GMSOL has even launched.

Hi Sand, your recognition means a lot to me, thank you.

Your proposal is reasonable, it’s similar to BB&D providing GMX Stakers with an option to convert WETH to GMX. In the previous proposal, I also fully demonstrated the differences between the two.

[gmsolq]

@tx_analysor

Initially, I wasn’t a strong supporter of your first proposal regarding BB&D, but I must admit it has proven to be a success. However, applying the same concept to LP fees seems like a misguided approach. While it would undoubtedly drive short-term demand for GMX and potentially lead to a significant price increase, such a surge driven purely by BB&D mechanisms is unlikely to be sustainable in the long run and could ultimately backfire.

The primary concern is that this change would fundamentally alter the GM structure, particularly by disrupting the autocompounding mechanism. The temporary price pump would come at the expense of GMX’s TVL growth and compromise the simplicity of the GM and GLV pools, which are currently designed for ease of use—allowing investors to simply add capital and let it grow over time.

As you said, we need both a good economic model and effective business development, but I believe the current BB&D model is sufficient in addressing the economic model. If the concern is that GMX is undervalued, we should focus instead on developing the GMX product to attract more TVL/OI and invest in initiatives to expand its community, such as GBC and gmSOL.

In my view, the cost of this strategy outweighs its potential benefits by far.

Hey TX, thank you for recognizing the success of BB&D.

Regarding price perspectives, I have a different view. Short-term price impacts are unpredictable for anyone, but long-term price impacts are determined by the relationship between supply and demand.

Whether the growth driven by the BB&D mechanism is sustainable depends on whether GMX’s business itself is sustainable. BB&D is not a decisive factor; it is merely a neutral tool, no different from BB&Burn and others.

It’s not that I worry about GMX being undervalued, but rather that GMX is indeed undervalued. This is a problem that needs to be addressed. The solution before BB&D was simply to “wait” and “trust the market to correct its errors.” Time and again, this has proven to be wishful thinking.

Autocompound is a newly introduced feature in GMX V2, but it is not the entirety of GMX V2. As I pointed out above, in my opinion, the greatest innovations of GMX V2 are:

1. Introduction of Funding Fee: This addresses the most critical issue—OI balance—greatly reducing risks for LPs.

2. Risk Isolation: This solves the second most important issue by isolating the risks of long-tail assets. It is the foundation for supporting multiple markets rather than being limited to BTC, ETH, and SOL.

3. SS Pool: This significantly expands the limitations of the balanced pool. The flexible combination of SS BTC/ETH/SOL/Other Pools and SS USD Pools will cater to all LPs’ risk preferences, rather than forcing LPs to take on 50% exposure. This would make GMX shine in bear markets but fade in bull markets.

Lastly comes autocompound, which merely optimizes the LP experience (and in fact, I remain skeptical about whether it is truly an optimization).

Thus, overemphasizing the importance of autocompound reflects a lack of understanding of the GMX V2 mechanism. The core competitiveness of GMX V2 lies in the first three points, not in autocompound.

[babloo]

I get it, from the LP side, my plans has always been instead of selling BTC when I wanted to(if it touches 150k maybe) put it in GM-BTC. It gives me good APR, I get traders losses auto-compounded. when price of BTC goes down, my BTC position would increase even if it decreases in USD, similarly for all assets. But auto compounding is important, for maximum return. 91% is retention because GMX at 30$ looks cheap to them, at other price it decreases for sure.
This aside, I would like to have a compound button beside claim button for claiming GMX right now, which automatically stakes the GMX I get as rewards.

[cachalot]

I recommend leaving the BB&D as it is currently implemented and observing it for at least one year. After that, we can proceed in a certain direction with solid metrics.

[gmsolq]

@babloo

I get it, from the LP side, my plans has always been instead of selling BTC when I wanted to(if it touches 150k maybe) put it in GM-BTC. It gives me good APR, I get traders losses auto-compounded. when price of BTC goes down, my BTC position would increase even if it decreases in USD, similarly for all assets. But auto compounding is important, for maximum return. 91% is retention because GMX at 30$ looks cheap to them, at other price it decreases for sure.
This aside, I would like to have a compound button beside claim button for claiming GMX right now, which automatically stakes the GMX I get as rewards.

I do not deny that auto compound is key to maximizing your returns as an LP. Without auto compound, if you are particularly focused on marginal returns, you would need to frequently claim GMX, swap it into BTC, and reinvest it to achieve maximum returns (while also considering certain tax implications).

1. However, I want to emphasize that this conclusion is based on the LP’s perspective. It is a solution to maximize LP returns, but it’s important to note that these assumptions are built on the foundation of a very high 63% fee distribution ratio. Moreover, this system is not only about LPs.

   If GMX holders wanted to maximize their own returns, they could simply vote to significantly increase their fee allocation while reducing the LP fee allocation to, say, 40:50 or even higher.

   If both parties only seek to maximize their own returns, it would inevitably lead to destructive competition, resulting in a lose-lose situation.

2. But what if LPs considered the perspective of GMX holders? They might support BB&D while holding more GMX to support GMX’s development, all while maintaining the high 63% fee distribution. At the same time, there is sufficient liquidity for selling, and selling remains a free choice.

   And what if GMX holders considered the LPs’ perspective? They might agree to maintain the 63% fee distribution to LPs to encourage greater liquidity provision, thereby offering a more competitive APY.

   If both parties seek to maximize global returns, this would result in a win-win scenario.

[russiamanbit]

btw, for autocompoud context - I think with manual compounding you will catch more % of the pool if you are an active compounder and do it by hands periodically compared to the case when all the participants is equally AUTOcompaunded. Because people are lazy… and most prefer to dont care alot… and even more - there will be much more temptation to claim accomulated rewards and withdraw it for shortterm needs rather then reinvest.

[gmsolq]

GMX: Transition from “Buyback ETH and Distribute ETH” to “Buyback GMX and Distribute GMX”

We are in favor of this proposal. Buying back GMX with the fees from the protocol will help increase the price in the long run. Many users in this ecosystem are happy to see the GMX price go up, as it will help unify the community. We also need to consider the relationship between the price of GMX and the long-term success of the GMX protocol. Right now, I do not see a strong connection. GMX LPs and GMX traders do not come to GMX for the GMX tokens. If this were ten months ago, we would defini…

@Time_Research

When reviewing the previous proposal, I came across comments from Time Research (likely one of GMX’s largest LPs), which raised several points that need to be addressed:

1. “We are in favor of this proposal. Buying back GMX with the fees from the protocol will help increase the price in the long run. Many users in this ecosystem are happy to see the GMX price go up, as it will help unify the community.”

Time Research clearly highlights the significant benefits of GMX’s strong performance for its development. Increased attention on GMX will inevitably bring more opportunities to GM LPs.

2. “We also need to consider the relationship between the price of GMX and the long-term success of the GMX protocol. Right now, I do not see a strong connection. GMX LPs and GMX traders do not come to GMX for the GMX tokens.”

Time Research points out the current lack of a strong connection between GMX LPs, GMX traders, and the GMX token—a fragmented relationship. BB&D is undoubtedly the best tool to build this connection. Furthermore, we could consider introducing tiered fees based on GMX holdings to strengthen the relationship with traders.

3. “If this were ten months ago, we would definitely vote no because we held a lot of GLP at that time. Forcing GLP holders to accept GMX would not be a good idea, as it is difficult to sell $5M worth of GMX tokens in the market, while ETH does not have this issue. Now, with V1 fading out, it is acceptable for us to support this proposal.”

This point warrants further scrutiny. It is undeniable that if your earnings accumulate to $5M, selling it all at once could have a significant price impact under current circumstances.

However, the issue you described would occur if GMX tokens were minted out of thin air, without increased liquidity, resulting in $5M worth of GMX that couldn’t be converted to $5M USD. Under the BB&D scenario, $5M worth of ETH would first be used to purchase GMX, injecting liquidity into GMX, and then you would withdraw $5M worth of liquidity, minimizing the impact.

Additionally, this discussion is based on a single LP. GMX has many LPs, which introduces a time lag, effectively creating a substantial liquidity pool (as reflected by the significant increase in GMX liquidity). Thus, exits should be simpler for all LPs, with each individual exit becoming easier as a result.

Furthermore, if you are concerned about this marginal risk, you can mitigate it relatively easily through hedging or by avoiding a $5M single-sale scenario. For example, selling in $500K increments? These are all solvable issues.

4. “Lastly, when GMX holders make decisions for LP holders, they need to consider the long-term profit of the LP holders and not focus solely on the GMX price itself.”

I agree with this point—GMX holders should make decisions carefully. However, there is a balance to consider here. GMX holders own the protocol but receive only 27% of the fee distribution. GMX LPs, while potentially affected by these decisions, have high flexibility and receive 63% of the fee distribution. GMX holders bear the risk of protocol failure, while GMX LPs bear the risk of trader profits and losses. Overall, the distribution is balanced.

In my view, if GMX holders focus solely on maximizing their own interests at the expense of GMX LPs, that would be shortsighted. Similarly, if GMX LPs focus only on maximizing their marginal returns without considering GMX’s overall development, that would also be shortsighted. A lose-lose or a win-win situation—this choice shouldn’t be difficult to make.