EIP-8363: Should ETH Be Sound Money or a Productive Asset?

EIP-8363 proposes burning a rising share of validator rewards as the staking ratio climbs, tapering issuance toward zero at approximately 50 percent ETH staked. The authors argue the current issuance curve has no mechanism to switch off staking incentives, meaning non-staking holders are continuously diluted to subsidize stakers. Research cited by co-author Jerome de Tichet suggests minimum viable issuance for Ethereum economic security is around 0.5 percent annually, and he projects the staking ratio will reach 50 to 60 percent by 2028 under the current regime.

The proposal targets an ideal staking ratio of 20 to 30 percent and reaches zero issuance at around 50 percent staked. De Tichet estimates the post-EIP equilibrium at roughly 40 to 45 percent staked, where validator yield would fall to approximately 0.3 percent APR. Oshin Kain estimates the equilibrium slightly higher at 45 to 49 percent with a yield near 0.4 percent APR. The current curve floors at approximately 1.5 percent yield, which creates constant upward pressure on the staking ratio. A transition period of roughly 18 months after Pectra goes live is included to avoid a sudden yield shock, and the curve shape, maximum issuance level, and transition timing are all described as open to adjustment.

Kain's primary objection is that the curve goes all the way to zero rather than stopping at 0.5 percent, which he considers the point where most negative effects arise. He distinguishes between finality security, addressed by slashing, and censorship resistance, for which there is no slashing penalty if 51 percent or more of stakers coordinate to exclude transactions. He estimates real daily security cost under the current regime at approximately 5 million dollars per day and warns it could fall to 1 million dollars per day or less under the proposed equilibrium, which he questions as sufficient for what he describes as a trillion-dollar computer. He also warns that reduced yield could drive stakers toward ETFs and custodians, worsening the Nakamoto coefficient, and that an entity holding 40 to 60 percent of ETH could drive issuance to zero under the new curve, causing rational actors to exit and handing that entity majority stake control.

De Tichet responds that EIP-8363 focuses on maximizing the cost to revert finality rather than the cost to censor, and argues DVT and FOCIL are better tools for censorship resistance than issuance changes. He contends that looping stablecoins to borrow ETH and earn staking yield is subsidized by dilution of all ETH holders without buying additional security. On solo stakers, he notes that France applies a 30 percent flat tax reducing a 2.5 percent gross yield to roughly 0.7 percent effective, and argues the current curve is already pushing solo stakers out faster and with higher dilution for everyone. He identifies MEV burn and anti-correlation penalties as the mechanisms most likely to genuinely help solo stakers, and notes that if all solo stakers exit under EIP-8363, issuance would still be lower than under the current regime.

The illiquidity of native staking is identified as a structural driver pushing stakers toward liquid staking tokens and centralized exchanges. In normal conditions withdrawal takes roughly four days and in stressed conditions could take months. One proposed solution is allowing stakers to burn ETH to exit immediately, with the cost rising during stress periods. Without a burn mechanism accompanying EIP-8363, larger operators can exploit timing games and capture more MEV, reinforcing consolidation incentives. One speaker agreed with the view that MevBurn should ship first and EIP-8363 be considered afterward with more data from real-time proving and CKVM developments.

Kevin frames the debate as a political economy problem rather than an engineering problem, noting the two camps have not moved toward each other in approximately two years. He observes that the All Core Devs process has never passed something with this level of contentiousness and that ACD is unlikely to push EIP-8363 through with active dissent without amendments. All Core Devs meeting 185 is scheduled for August 20 at 1400 UTC and is described as a key moment. October 26 is the proposed-for-inclusion deadline at which core developers will decide whether EIP-8363 and the anti-correlation EIP are ready to trigger testnets. If not included in the fork deploying later this year, the decision defers to the next fork expected in early to mid 2027.

The underlying tension is a zero-sum trade-off between ETH as ultrasound money with a monetary premium and ETH as a productive yield-bearing asset for DeFi and institutions. Jerome rejected the framing that EIP-8363 is bad for institutional adoption, noting Grayscale published a measured, welcoming response to the issuance discussion, and said some ETFs and institutional buyers have told him they have a narrative problem explaining why ETH will hold its value. He warned that if stake concentration continues under linear issuance, security assumptions will need to be reconsidered within three to four years, and that most mature proof-of-stake chains without slashing sit at 52 to 55 percent staking ratios with some near 70 percent.