Iâll speak mostly to first and second order economics rather than the mechanism design, which others have covered better than I could.
Firstly there has been huge miscommunication over the ownership of the proposal. Authors should clearly write that this is their independent work, not the public stance of the Ethereum Foundation. The result was chaos, confusion, focus drain, and real costs for teams across the ecosystem that had to respond.
On first-order effects, I believe the honest description is:
-
Home stakers who exist today mostly stay, but new ones stop coming. ROI on a home setup moves toward ~10 years to pay for a machine that will likely never pay for itself.
-
On the rationale presented for this: small operators run on ideology, not money, so theyâll be the last to go. It describes a cohort that only shrinks. No new home stakers join, existing ones age out. A policy that âprotectsâ solo staking by freezing the current holders in place while making the next one irrational is not protecting solo staking. It is managing its decline.
-
Small professional operators close. Big ones consolidate to cut cost, and cost-cutting extends to operational personnel. Fewer people across the board, which means higher probability of slashing events, key management failures, and everything that follows.
-
Public good operators (client teams) lose their life support. Today, LST operator sets are what keep minority-client and public-good operators alive.
-
LSTs (including LRTs) run on extremely thin margins. There is no long-term survival for them under this regime. They can cash-cow until the first big fork, and then you get maintenance mode: fewer audits, weaker monitoring, slower incident response, less security spend across the board. That degrades the operational credibility of the network, quietly.
The response to this in the thread is the wedge analysis. I think this analysis misses its most important term. CEXs do not need staking revenue. Staking is a product line for them, a retention feature that cross-subsidizes everything else. In short, they will stake at zero, or at a loss, indefinitely just to provide complementary service. The highest-wedge participant is not the one who exits first when that participant doesnât need the yield at all. And once the field narrows, you have functionally monopolized the creation of new validators: the only entities that will turn on new validators are the ones that are already profitable at the new equilibrium or structurally able to operate at a loss. That is not a distributed validator set. That is a cartel of incumbents with a product line.
This is the part of the proposal I would most like the authors to answer directly: the mechanism is described as size-progressive, but larger stakers overwhelmingly stake on behalf of others and will not voluntarily stop growing to avoid the curve. Who is left standing is determined by who can operate at the thinnest margin for the longest, and the answer is entities for whom staking yield was never the business model.
On LSTs, they are the only mechanism Ethereum has ever had that distributes stake: to minor operators, to minority clients, to public-good operators and are soft capping the large operatorsâ max share within the protocol. The only pushback Ethereum has against large operators is bottom-up social consensus, and LSTs are currently the instrument through which that distribution is actually executed. Remove them, and stake dies toward whoever already has scale, distribution, and a balance sheet. It is also worth saying plainly that the slashing-risk picture changes qualitatively: it is a very different network when one operator with 30% of stake (running mostly a single client) blows up, versus a market where LSTs actively curate distribution and keep minority-client and public-good operators solvent. Client diversity doesnât trend on timelines, but it is part of the security impact this proposal touches.
On second order effects, quickly, because they stack:
LST/LRT business models collapse â CEXs, who donât need staking revenue because itâs a product line, inherit staking TVL. Winners: Binance, Coinbase, a handful of ETF issuers, DATs. This is the validator cartel outcome, delivered by an anti-cartel proposal.
Lending: Aave loses the bulk of its activity (correlated-asset looping is the majority of its flow), Morphoâs path to fees gets set back, smaller markets close entirely as cost of minimal maintenance exceeds revenue.
DEXs: onchain arbitrage between LSTs and ETH disappears; that flow is a meaningful share of volumes. Aggregators lose pairings; users eat worse routing on what remains. (offchain-onchain arbitrage also takes a hit.)
Earn products: this is my day job, so let me be concrete. For an ETH-denominated yield product it is yield or sunset. Before sunsetting, you fight to extract with higher risk. And here the proposalâs DeFi logic collapses in on itself: the motivation says âLSTs and other staking derivatives displace raw ETH as the ecosystemâs working moneyâ and that high staking yield made other DeFi yields unviable. But capital seeks yield regardless. If the yield is not available onchain, it does not sit idle in raw ETH. It moves into centralised and riskier solutions. The proposalâs own logic therefore predicts the outcome it claims to prevent: more ETH custodied, less self-custodied, in wrappers with worse risk disclosure. Lending markets will list RWA substitutes at scale to replace the lost ETH-correlated flow, because the revenue hole is huge. Iâve watched genuinely solid RWA products fail risk diligence, and under this regime they get listed anyway. End state: users pushed into materially riskier products, for materially higher risk:reward ratio, concentrated in fewer, bigger, more custodial wrappers.
And then the largest point: a validator set dominated by a handful of operators in one or two jurisdictions is a regulatory attack surface, not a neutral one. A single monetary policy decision that ends up consolidating stake into regulated custodians/operators hands governmental leverage over the network. That is not hypothetical game theory; it is just how coercive pressure works on KYCâd entities with licenses to protect.
There is also a consistency problem in the motivation itself. This ecosystem deliberately chose to subsidize data availability for L2s to a near-zero price. A policy decision that collapsed the fee burn, the single largest deflation engine ETH had. Whether or not cheap blockspace eventually pays for itself in demand over long timeframes, the timing mismatch is real and it was our choice. To now declare net inflation intolerable and reach for the issuance that affects security raises questions about the sequencing of economic policy before changes are enacted.
On process and scope: the urgency claim is asserted, not evidenced. Authors should focus on providing specific outcomes with proof, not theoretical subjective opinion which is being provided as a reply to any post until now. The queue projection is a straight-line extrapolation of current inflows through a period the authors themselves describe as yield and sentiment driven. We know itâs conviction driven by Bitmine as theyâre in process of acquiring 5% of total supply and staking it. Congestion comes from chosen technical setup of the network, not of broad demand.
To me this proposal does not make sense, it is rushed (a slow implementation does not fix a fast decision), and it does not meaningfully incorporate the counterarguments raised when this was last pushed. The draft reads like the same thesis re-presented at a friendlier moment, not a revised position. That is not what should be expected from prominent researchers. In other words what we expect and what this process needs is that authors become more receptive to feedback and to iterate proposals in response to it. Nobody is 100% correct at all times but we can jointly build better models and outcomes for Ethereum.
So, the concrete asks toward the authors:
-
Withdraw the EIP from HegotĂĄ PFI consideration. A monetary redesign of this magnitude, surfaced 48 hours before the deadline, cannot legitimately enter a fork pipeline before the evidence work below exists. PFI may âopenâ feedback in procedure, but it sets a default path in practice. Meanwhile a large majority of the community, builders, and network participants responding publicly are against the proposal as-is (LobsterDAO TG, Twitter and even this thread as public verifiable sources).
-
Engage independent financial experts to validate the economic effects: demand elasticity, operator cohort margins, DeFi revenue impact and network utilisation, including CEX cross-subsidy behavior, rather than resting the case on the authorsâ own equilibrium modeling. Academic models of a steady state are not validation of what happens to operating businesses on the way there.
-
Publicly engage the counter arguments with research, not opinionated rebuttal. Publish work that engages those findings on their own terms, and where the evidence is genuinely split, the design must reflect the split rather than assert through it.
-
Commission the DeFi cascade analysis (lending market viability, DEX/aggregator flow, staking-adjacent product risk migration) with the teams that operate those markets, before SFI, not after.
-
If supply growth limitation is the true target, start where the ledger was deliberately zeroed out: the fee and blob-pricing side, and evaluate the non-issuance levers for the concentration concern (correlated-failure penalties, further consolidation) before touching the security incentives the whole economy prices off.
Iâll close with the institutional angle, because itâs my honest read after years of these conversations: what institutions bought on Ethereum is the guarantee that no other institution or individual can overtake the chain underneath them. Business continuity. Credibility that the settlement layer they build on for decades doesnât restructure its security economics underneath them. This proposal trades that for a hypothetical ETH price effect. The new equilibriumâs winners are loss-tolerant actors, and loss-tolerant actors are not ETH holders, current stakers, and they are not Ethereum.
Best,
M