Adjusting Arbitrum’s Gas Fees

Arbitrum Gas Fees & Sequencer Revenue Proposal

Constitutional

Abstract

This proposal seeks to increase the amount of ETH flowing to the DAO’s treasury through two milestone proposals: The first, increase the L2 minimum base fee from 0.01 gwei to either 0.03 or 0.05 gwei, and to establish a working group tasked with creating a subsequent Milestone 2 proposal that improves the DAO’s ability to act swiftly when market forces signal a need for changes to Arbitrum’s transaction fee mechanism.

Milestone 1 is a simple proposal to raise the L2 minimum base fee from 0.01 gwei to either 0.03 or 0.05 gwei. Milestone 2 will be a separate and subsequent proposal, created by the working group, that intends to create a more long-term solution. The decision to approach the subject in this manner was made on the working group call last week.

Motivation

The Arbitrum DAO has experienced a significant decline in ETH revenue, primarily due to recent transaction fee parameter changes brought about in the ArbOS Atlas upgrade. From January to March 2024, the DAO spent an average of 13.8M ARB per month, primarily on protocol/user incentives. Additionally, the number of ARB funded programs continues to grow, which ultimately poses risks to the long-term value of ARB and subsequently, the DAO’s treasury. Meanwhile the DAO’s largest source of revenue—derived from the sequencer margin—has undergone a drastic reduction. This margin, a crucial component in achieving eventual DAO sustainability, has been impacted by reductions in L1 surplus fee and the minimum L2 base fee, as implemented in the ArbOS Atlas upgrade. Although these adjustments have reduced transaction costs on Arbitrum One, they have inadvertently flattened the ETH inflow into the DAO’s treasury, as detailed in our recently published research. We highly recommend reading this post prior to diving into the rest of this proposal.

This proposal aims to optimize the DAO’s primary revenue mechanism to support the eventual long-term economic sustainability of the Arbitrum ecosystem without hindering protocol usage or growth. The DAO has missed out on substantial revenue since Atlas and this proposal looks to address that oversight. Implementation is particularly crucial during periods of aggressive user incentives, as these are times when the DAO misses out on material ETH revenue in return for the ARB it is spending.

While the DAO envisions diversifying its revenue streams through initiatives like the GCP and other capital allocation activities, licensing fees on the Arbitrum tech stack (expansion program), putting a portion of the treasury to work in RWAs and other yield-bearing strategies, the critical reliance on ETH revenue through sequencer margins remains paramount today and for the foreseeable future. This revenue is crucial not only for maintaining current operations but also for funding essential innovations that ensure Arbitrum retains its competitive edge such as the ETH denominated costs associated with BoLD. Building DAO revenue streams is of the utmost importance in order to secure the future of the ARB token and DAO.

Rationale

Ultimately, our goal is for the DAO to evolve into a sustainable entity, with robust and adaptable revenue mechanisms ensuring long-term stability and growth. For this to occur, adaptability is necessary. Dynamic adjustments to the minimum base fee variable will allow the DAO to swiftly respond to market conditions and protocol needs, ensuring that we can capitalize on opportunities and mitigate risks as they arise. A flexible approach will enable us to maintain a steady source of ETH denominated revenue, thus enhancing the DAO’s treasury diversification, reducing dependence on ARB, and bolstering overall risk management.

While the need for flexibility in fee adjustments is evident given the dynamic nature of demand for Arbitrum’s block space, properly designing such a system raises many open questions. Below are two potential paths for the DAO to consider as a part of Milestone 2, followed by a few open questions:

Option 1: Create a multisig through Hats protocol or a similar mechanism that has the ability to change the min base fee within a certain threshold, such as between 0.01 gwei and 0.1 gwei. Each signer would also have a legal agreement with the foundation that they are expected to change the min base fee variable based on a DAO snapshot vote that reaches a certain quorum (100M ARB). This mechanism maximizes community involvement in changes to the variable and reduces friction to adapt swiftly. However, we are not sure that the DAO is the best suited to make this granular level of decision in the context of the optimal minimum base fee.

Option 2: Implementing an RFP process to select service providers with optimistic control over the fee adjustments within certain thresholds. This mechanism provides a more responsive approach, leveraging decentralized governance tools like a Gnosis Safe, Hats protocol, and a X-day time lock with DAO oversight. This way, we can vote in the most suited entities to the role while minimizing their power. It is our suggested option that leads to many open questions. We dub this solution the Transaction Fee Mechanism Committee, or TFM Com.

Open Questions:

  1. In addition to regular reporting and term limits, what are other accountability methods to ensure selected service providers answer to the DAO?
  2. Negotiations and research will have to be conducted to determine a proper budget.
  3. Should selected providers be given powers over more aspects, such as target gas per block, in addition to just the L2 minimum base gas fee? This would also give the Transaction Fee Mechanism Committee (TFM Com) the ability to conduct valuable research into the economic parameters of Arbitrum and potentially change them in future (with DAO approval)

Since it is a long-term strategic necessity for Arbitrum DAO to be adaptable amid constantly evolving market conditions, these questions should begin to be addressed by the already stood up working group. However given its current missed revenue and expenditure, there still remains an immediate need for the DAO to consider increasing the minimum base fee. This proposal seeks to address both of aforementioned problems in two milestones detailed further below.

Specifications

Milestone 1

Milestone 1 will proceed to Snapshot in the coming week(s) and is a constitutional proposal to increase the minimum base fee to 0.03 or 0.05 gwei. This represents a straightforward approach and addresses the immediate missed revenue of the DAO. These numbers have been chosen as they all fall substantially below the 0.10 gwei L2 minimum base fee that the protocol implemented prior, and still allow Arbitrum to be one of the cheapest Ethereum L2s. It is clear from the analysis of the competitive landscape that in general real users are not fee sensitive at these thresholds.

Milestone 1 is a swift change, addressing the immediate and urgent revenue challenges while we fine-tune the operational details of the TFM Com. For further motivation and reasoning around Milestone 1, again, please refer to our research post. This approach ensures we stop the ongoing loss of revenue as soon as possible. Concurrently, Milestone 2 will introduce a long-term solution, establishing a sustainable framework for the future. This dual-phased strategy underscores our commitment to both immediate improvements and enduring success.

Milestone 2

As Milestone 1 is working its way through the Constitutional AIP process, working group discussions will begin for Milestone 2. The purpose of these discussions is to determine the TFM Com’s mechanisms, expectations, budget, and methods of accountability. Post working group discussions, we believe the first stage of Milestone 2 will be hosting an RFP process, in which we seek three quantitative mechanism design firms to sit on the TFM Com.

Assuming the working group is aligned with our thought process, after the three service providers have been decided, they will each be added as a signer to a ⅔ SAFE multi-sig they will create. We will then utilize a role-based access control mechanism, such as hats protocol, to give this multisig the ability to alter the minimum base fee between 0.01 and 0.1 gwei, or some other agreed upon range that is approved by the DAO. Each change must be reported to the DAO X (tbd) days before being made and go through an X (tbd) day timelock. The multisig will be created with upgradeexecutor as a module, allowing the DAO to replace members at anytime through an onchain vote.

Potential TFM Com Expectations (TBD in Milestone 2)

We envision the TFM Com being tasked with the strategic oversight and dynamic management of Arbitrum’s min base fee variable. The committee’s primary goal is to enhance the DAO’s revenue adaptability and sustainability, ensuring that the platform remains competitive and financially healthy. While many details are still to be determined by the working group, below is how we anticipate some of the responsibilities and accountability measures.

Fee Adjustment Oversight:

  • Monitor and analyze market conditions, user activity, and network demand to inform fee adjustments.
  • Manage the minimum base fee within a predetermined range (initially set between 0.01 to 0.1 gwei) to optimize network usage without compromising the DAO’s revenue streams or network stability.

Research and Reporting:

  • Conduct ongoing research into various components of the fee mechanism, including but not limited to the target gas per block and the impact of fee changes on user behavior and network health.
  • Prepare and present regular reports to the DAO on current fee structures, proposed changes, and the impact of implemented adjustments.

Community Engagement and Feedback:

  • Actively engage with the community to gather feedback on transaction fee policies and their impact on the user experience.
  • Facilitate discussions and workshops to educate the community about the fee mechanism and the rationale behind proposed changes.

Accountability:

  • The TFM Com will be held accountable to the DAO through regular updates and a robust audit process. Community feedback will be actively sought to assess the committee’s performance and alignment with the DAO’s long-term goals.
  • The TFM Com will serve a predefined term length, but members can be replaced or the program terminated by the DAO through an onchain proposal at any time.

Timeline

June 17 - Forum Post

June 24 - Milestone 1 Snapshot Vote, ranked choice, with options:

  • Do Not Change Min Base Fee
  • Min base 0.03
  • Min base 0.05

July 1 - Proceed to Tally for constitutional vote

Budget

Milestone 1 is free to implement for the DAO and the proposal process will be facilitated by Entropy Advisors.

Milestone 2 will require an additional DAO vote and a yet to be determined amount of funding, but these details fall outside of the scope of this initial proposal. The scope of the TFM Com’s work still requires more input from the wider Arbitrum community, and will heavily influence the compensation of the Com’s members.

9 Likes

Thank you @Entropy for kicking off the preliminary research and proposal on one of the most important issues for the Arbitrum DAO.

As expressed in the forum post for the original Atlas change as below, we are very interested in exploring the appropriate fee value and further considering the best way to make the DAO sustainable to balance the fee revenue mechanism.

It would be great to share, if exists, how the fee revenue would change if the minimum base fee is changed to 0.03 or 0.05 (or any other values under 0.1).

1 Like

Its grea to see that @Entropy kicked off this topic.
I absolutely agree that something has to happen and I am in favour of executing Milestone 1 asap.
Changing the base fee from 0.01 to 0.05 should not have any impact on a users decision which L2 he is going to use in the future. Its such a small impact for an individual, but a huge for the DAO and its future revenue.
Because in the end (speaking for myself) it doesn’t matter if a transaction costs me 0.05$ or 0.1$ it is still cheap compared to mainnet.

For Milestone 2 I think its crucial to work with SP that are already working with the Foundation and can be trusted. In my opinion these individuals should have KYC towards the Arbitrum foundation just like delegates that are being compensated.
Best case would be even to have an automated system, tracking other L2and thus deciding to change the base rate based on these information within a predetermined range to reduce overhead and be more flexible.
But I am not sure if its technical possible and how much time this will need to be implemented.

But it would definitely be a pro, similar to all lending markets adjusting their borrow rates for example for stablecoins.

This is taken from the original discussion post, but please note these are very rough estimates.

3 Likes

Thank you for mentioning Avantgarde during the discussion as a potential service provider for calibrating the base fee. We would agree that Option 2 utilizing optimistic control over the fee adjustment to be run by a specialist would provide a sensible balance of flexibility, expertise, and oversight.

Having run modelling work on sequencer revenues and their intersection with economics at the DAO treasury level, we are happy to be part of that discussion to help contribute to a potential Transaction Fee Mechanism Committee in a meaningful way

Thanks for the suggestion @Entropy I like the two fold approach of making a swift change and then a dynamic market based approach.

Establish clear performance metrics that service providers must meet along with regular reviews of their performance against these metrics. In addition, incentive structures where service providers are rewarded for meeting or exceeding performance targets & penalized for underperformance would be cool to have.

There are pros and cons to either approach. Allowing providers to manage multiple aspects can lead to a more holistic approach to network fee and performance optimization, on the other hand there’s a centralization risk as well as managing multiple parameters can introduce additional complexity and potential for errors.

I suggest, starting with limited powers such as target gas per block and accessing impact.

Excellent proposal, as it makes the milestone 1 easily executable and a more in-depth analysis for a later stage.

Regarding this point (for milestone 2), can we add a mechanism where the DAO can challenge the change? You guys mentioned that it will be reported X Days before being made. What happens if there is a disagreement (ex. DAO members don’t agree that is the right time to change the fee)

The following chart illustrates estimated monthly revenues for the Arbitrum DAO under different minimum base fee scenarios for the months of April, May, and June (up to June 20th). These estimates assume that all other variables remain constant. The chart compares actual revenue at a minimum base fee of 0.01 gwei with projected revenues at 0.03 gwei and 0.05 gwei.

3 Likes

Some more data that will be helpful to understand the impact of this proposal:

A normal EOA to EOA txn costs 21,000 gas
Uniswap Swaps vary depending on complexity and V2 vs V3, but 160,000 is a good average.

With a min base of 0.05 and ETH price of $10,000
0.00000000005 * 21,000 * 10000 = 1.05 cents
0.00000000005 * 160,000 * 10000 = 8 cents

3 Likes

In my opinion, the main task of Arbitrum is not to become an organization that generates income. In the long term - yes, but now there is a struggle for users. And this is the main goal.

After the update and reduction of commissions, Arbitrum still has a good competitive advantage. If the commission is increased, then other chains will take over users who want to pay less commissions.

As a result, for the sake of the short-term prospect of making more money, we will lose the opportunity to increase users and the opportunity to receive, accordingly, more commissions in the future.

The change as proposed, at 0.05 min base fee and current ETH prices, would add $0.00294 to a normal wallet to wallet transaction only in times of no congestion. We don’t believe this would impact any real user growth.

Do I understand correctly that you consider it normal that the fee for users will increase by a multiple of the increase in the base commission?
Now the swap costs $0.01-0.02

All users will go to other chains, taking into account how well and inexpensively bridges work now

Arbitrum would still be significantly cheaper than its competitive landscape. L2 min base only impacts times when there is no congestion. Arbitrum is close to being a place where it is frequently in congestion and this change then becomes irrelevant.

I’m in favor of increasing the Min Base Fee because I think it is important for the ongoing sustainability of the Arbitrum DAO to generate significant surplus fees. I don’t think this will have a meaningful impact on chain usage, because I don’t think users are price sensitive below a certain level.

1 Like

Appreciated the additional data provided for us to appropriately evaluate what potential increases make impact on future revenue fee and chain usage. We would like to express our support for the minimum base fee increase.

I’m in favor of increasing the minimum base fee to support financial sustainability. My guess is the user will have a low price elasticity of demand (:eyes: someone showing off their economics degree), and this adjustment will be a clear net positive for the DAO. This is especially true considering the sizeable current fee gap between Arbitrum and its competitors. I am excited to see the conversation continue around Milestone 2 and set up a more robust long-term solution to this important question.

Why would one change the chain because of this small amount? To be honest, I even think most don’t even realize that the price for a transaction has changed a lot. I would not for example. If the price is 0.005 or 0.009 wouldn’t care to me.

The current design and architecture underlying Arbitrum allow for the Min Base Fee to be increased in accordance with the proposal, while not meaningfully compromising the chain’s feasibility, usability, competitiveness and USPs.

Financial sustainability is an essential component of any long-term oriented organization and its prioritization can be a key differentiator for the future prospects and viability of the ecosystem.

The proposal effectively combines the two aforementioned aspects in a manner that clearly positions its implementation to be a net positive for Arbitrum.

There is indeed an unanswered question: has there been any thought on the consequeunce of raising fees?

Everybody is excited about getting more revenue for the dao; but what would the impact potentially be currently?

There is a world in which gas cost, amount of transaction and revenue could follow a supply-demand-cost model for example, and maybe increasing the fee could mean a drop in txs activity in a way that could be meaningful for specific protocols for example.

And since the proposal will go to vote soon, wanted to ask about any study on the potential consequences, for the good or the bad.

3 Likes

We are currently being spammed by a protocol called UX Link which is one of Arbitrum’s biggest txn drivers. We are making virtually no money from these transactions. One implication of this proposal is that bot volume will likely decrease, which in turn will have the negative impact of decreasing unique wallets and # of txns. However, these are vanity metrics. As described above, the impact on real users is minimal even for protocol’s with high gas costs, and it is to safe to assume that there be little to no user drop off unless you think adding pennies to txns costs will suddenly make users fee sensitive.

Arbitrum gas pricing already follows a supply-demand-cost model with congestion pricing. Min base fee is only relevant in times when congestion pricing is not kicked in.