Should Ethereum issuance be reduced by burning validator rewards as staking grows

Protocol design · 36 participants · 52 quotes · 2026-08-04 ~ 2026-09-03

The authors of EIP-8363 Tapered Issuance Burn (the proposer's tweet on the day of publication listed it as EIP-8361) proposed burning validators' consensus rewards as the staking ratio approaches 50% of the ETH supply, to remove the incentive for stake to grow beyond that. Co-author pintail said the argument that excess issuance leads to capture is not a rhetorical trick, and dapplion said that the operators who object are refusing only because they stand to lose, while this EIP benefits all holders in the long run. On the opposing side, rotki's Lefteris Karapetsas said issuance is not a problem Ethereum should be dealing with now, Aave's Stani Kulechov said Ethereum is still an unfinished product for supporting the financial system, so the protocol-level priority is privacy, and Marc Zeller of Aave Chan Initiative said the current risk-free yield gives wETH its appeal as collateral and there is nothing to replace it. ether.fi's Mike Silagadze said this change, disclosed with only 48 hours for feedback, would push out solo stakers and kill a large chunk of DeFi, and placed a public $1 million bet that the network would become more concentrated. DCinvestor said that, in a situation where staking yield has become a core reason to buy and keep holding ETH, consistency and predictability of policy matter more, and that changing issuance is itself value destroying. Micah Zoltu largely agreed with the authors' argument but opposed the change on the ground that the monetary policy of a live chain should not be changed, and Fundstrat and Lido took neither side, speaking instead about the uncertainty of the optimization curve and preparing for a low issuance regime.

Background

EIP-8363 Tapered Issuance Burn is a proposal to deduct and burn a set share of the ideal reward each epoch for each duty a validator is assigned, namely attestation, block proposal, and sync committee participation. The burn rate grows linearly with the staking ratio and reaches 100% at a fixed saturation balance, so the net staking yield falls as staking grows. The proposers argue that in the current curve the yield falls only in inverse proportion to the square root of the staking ratio and has a floor of about 1.5%, so there is no point at which the staking incentive switches off. They explain that introducing the burn lets the staking market find equilibrium where the yield meets the risk premium that stakers require, that this point lies below a 50% staking ratio, and that above it issuance no longer induces stake growth. Applying it in full at the fork would cause the yield to plunge at the current staking ratio, so BASE_REWARD_FACTOR is temporarily raised and the burn is phased in over 18 months, while the shape of the curve, which does not reward growth beyond 50%, takes effect from the first day of activation. By the figures quoted on Twitter, the net consensus yield falls from 2.6% to 1.2% at current staking levels, and the first-day reduction is 13%. The authors are pintail, Jérôme de Tychey, dapplion, pa7x1, Ladislaus von Daniels, and Justin Drake. The core principle rests on pa7x1's prior research, and the refinement of per-duty burning rests on research by Anders Elowsson. According to the EIP document the date of writing is July 14, 2026, and it was submitted to ethereum/EIPs on August 4, 2026, which opened the Magicians thread, and its status is Draft. The proposer's tweet on the day of submission listed it as EIP-8361, and it was later cleaned up to 8363. The submission came 48 hours before the PFI deadline for the Hegotá fork, and in the thread gregk pointed out that there was not enough time to review a monetary policy change of this size, and that on the strawmap, issuance changes were expected to be handled in an I* fork. De Tychey replied that PFI is not scheduled inclusion but the start of feedback, and that if the validator entry queue stays full at its limit, more than 70 million ETH will be staked by January 1, 2028, so waiting for I* is effectively the same as giving up on the change. Because it is a consensus layer change, a hard fork is required. On September 2, a separate research proposal thread on how validator control changes if EIP-8363 works was also posted on Magicians.

Why it matters

The issuance curve is a parameter that sets staking operators' margins, DeFi collateral yield, and non-stakers' dilution rate all at once, so touching it once moves the profit and loss of the entire application layer together. In this debate, the main thrust of the opposing arguments was not the issuance figure itself but two things. One is that staking yield has served as both a reason to hold ETH and the benchmark rate for wETH collateral, and the other is whether researchers and core developers can change the monetary policy of a live chain without community consent. This second question is why a structure formed in which even those who agree with the authors' argument oppose the act of changing itself, and any future issuance change will have to pass through the same gate. The proposal remains in Draft, so nothing changes immediately, but as long as the staking ratio keeps rising, the same question will come back at every fork. Points to watch are the progress of the discussion in the Magicians thread and whether it is included in Hegotá, the actual response of operators such as Lido that have begun adjusting operating costs and revenue sources on the premise of a low issuance regime, and the possibility of liquid staking withdrawal that ether.fi has signaled.

Key points

• Is placing a de facto cap on the staking ratio a problem the protocol should solve now, or should unfinished work such as privacy, scaling, and UX come first
• When the net consensus yield falls from 2.6% to 1.2%, is it only validator income that shrinks, or also the reason to hold ETH, the demand for wETH collateral, and the load on DeFi built on liquid staking tokens (LSTs)
• Does the proposers' assumption hold that once staking exceeds 50%, remaining holders have a strong incentive to avoid dilution and will crowd into large LSTs and centralized entities? The opposing side says this has not happened on other proof-of-stake chains and that 0.2% annual dilution is no incentive, and some point out that Solana has already exceeded 50%
• What happens to solo stakers? The authors say that at a low staking ratio the yield is a multiple of dilution, so they are not harmed, while the opposing side says solo stakers, who have no fees to cut, are pushed out first. Within the opposing camp, the diagnosis that it is already over is split from the argument that they must not be allowed to be pushed out. Projections that validator costs approach zero in Lean Ethereum are also set against demands for a proper cost model
• Can researchers and core developers change the monetary policy of a live chain without a 48-hour notice and community consent? Even those who agree with the argument oppose the act of changing itself, and the risk of a contentious hard fork and a disconnect between researchers and the community were raised

For — The incentive for staking to grow beyond 50% must be cut off now

pintail, Jerome de Tychey 🦇🔊, Lion ⟠ dapplion .eth, Justin Drake, Dankrad Feist, FlowState , Merlijn The Trader, HuntingIsland — ΞTH 🐬

• This is a minimal market-driven amendment to Ethereum's issuance policy, and its purpose is to remove the incentive for staking to grow beyond 50% of the ETH supply.
• The argument that over-issuance leads to capture is not a rhetorical trick, so it should be engaged with rather than dismissed. At low staking ratios the yield is a multiple of dilution, so even the after-tax return is far larger than the dilution, and the only thing that cannot be guaranteed is a specific amount of income. As we can see now, stake keeps flowing in until issuance is offset by dilution.
• Lido, ether.fi and Aave reject it because it hurts them, but this EIP benefits every holder in the long run.
• Many people support it, but those opposed are louder and have more to lose. The claim that everyone agreed issuance does not need to be touched cannot be true while there are people who believe it should be changed.
• I do not accept the framing of a reward cut. Lower issuance tightens supply and strengthens the asset that underpins DeFi and stablecoin activity, and lower issuance makes ETH scarcer, which is positive for the bull thesis.

Against — Issuance is not a problem to solve now, and the monetary policy of a live chain must not be changed without consensus

Stani, Lefteris Karapetsas, Mike Silagadze🛡, DCinvestor, Marc Zeller, Micah Zoltu, Péter Szilágyi, _gabrielShapir0, Emiliano Bonassi, scoopy.sockpupp.eth, Mr O’Moduluszk, ManLy, Xerberus, 初七, hanniabu.eth (Ξ, α), UnHΞdged, Nomad, Aibra

• Ethereum has serious problems to address, both as a protocol and as an ecosystem, and issuance is not one of them. Ethereum is still an unfinished product in that it serves the financial system, so privacy is the protocol-level priority, and finite energy should go to what really matters, such as scaling and UX improvements.
• The current risk-free rate is what makes wETH attractive as collateral, and nothing replaces it. If leveraged staking, the driver of large app-layer revenue and AUM, takes a hit and app leaders lose TVL and revenue, that destruction of value is not offset by a small reduction in inflation. A dependency analysis of about 12,200 DeFi components also found that 5 of the top 11 system tokens were ETH LSTs, with no spare backbone behind those hubs.
• ETH issuance is already lower than Bitcoin's and lower than the new gold supply added each year, so it is not a main driver of price. Like it or not, staking yield has become a core reason people buy and keep holding ETH. Policy consistency and predictability matter more to market confidence than perfect optimization, and the very idea that a committee that thinks it knows the answer can repeatedly change parameters on a live network is offensive to large holders, so the act of changing them destroys value.
• Even if one largely agrees with the authors' arguments and the proposed schedule is better than the current one, that does not mean it is acceptable to change the monetary policy of a live chain. A proposal that cannot pass the gate of consent from stakers, validators, users and ecosystem developers only raises the risk of a contentious hard fork. Past issuance changes gave a lot to many other groups in exchange for causing pain to a powerless group, but this proposal only takes and gives nothing. The valid sub-items should first win community consent, and then a proposal everyone can support should be built.
• It was published with 48 hours for feedback, and no builder was asked for input beforehand. Remarks like 'Ethereum is not a democracy' and 'the math says you are wrong' show that researchers and core developers have become completely disconnected from the community. The defense that it was already decided in a closed discussion years ago and you were simply not there only confirms the rotten parts of ivory-tower technocracy and the power structure.

Neutral — Withholds a stance for or against and examines the shape of the curve, the cost data and a low-issuance regime

Zach Pandl, Lido, Thomas (Tom) Lee (not drummer) FundstratDirect.com, materkel.gwei 🦇🔊, donnoh.gwei 💗, oisin.eth | Obol, La-Li-Lu-Le-Lo, jasperthefriendlyghost.eth

• It must first be confirmed whether the opposing camp could also support an issuance curve that flattens but never reaches zero.
• The issuance question has been discussed in several proposals on Ethereum Magicians since 2024, so this EIP did not come out of nowhere and is most likely the result of discussion that has passed back and forth among several proposals over the past few years.
• Validator operating costs are likely to approach zero in Lean Ethereum, but since we do not know when stateless ZK validators will arrive or what effect they will have, the debate is currently being held on outdated data. On the other hand, real-time proving costs over $100,000 a year in hardware depreciation alone, and a proving-only client is more than 5 times as expensive as a large exchange, so solo staking is still at a disadvantage compared with delegation. An EIP like this needs a proper cost model so that we can work from data rather than vibes.
• These days an LST is effectively a home staker, as Rocket Pool, Obol and Lido each support hundreds of home stakers. There is little evidence that any chain inevitably heads to a 100% staking ratio, but Solana has already passed 50%.
• Whether we are overpaying for security and placing an unfair burden on those who do not stake is a fair question, but the optimal curve is unknown, and there are also unintended consequences and volatility risks. Even without issuance revenue, Ethereum still has many ways to generate revenue.

Other — The source text alone does not establish which side this is on

Etc., pratas.eth

• Etc. replied to a post organized into five items, saying that items 1 and 2 are not drivers, items 3 and 4 are underrated, and the focus should be on item 5. The numbering matches the five-item list that DCinvestor posted the same day, but the quoted text alone cannot establish which post this was a reply to.
• pratas.eth said that more than 90% of ETH is no longer held by solo stakers, so people should stop pretending to know. The source text alone does not establish which side of the argument, among those citing solo stakers, this was aimed at.

← Debates