Hello @EthWarrior, and thank you for taking the time to write all of this. It is genuinely useful feedback, and apologies for the time it took to get back to you. I’ll focus on your first section (Zero-yield staking selects for institutional staking) and a bit of the fifth for now.
The proposal’s primary goal is to deter the deposit size from crossing the 50% mark. Several security considerations sit behind that objective; the most important one, for me, is keeping a large reserve of unstaked ETH as the last credible counterweight the social layer holds in defence of neutrality.
Under the current curve the dynamic snowballs the other way: the higher the ratio, the higher the dilution, and the higher the dilution, the stronger every holder’s incentive to stake merely to avoid it. It is fair to assume most of that snowballing stake accrues to large actors, because they are perceived as carrying the least operational and counterparty risk.
Agreed on who sits near the front of the exit queue: a 32-ETH home validator with a $500/yr bill (your own, rather fair, assumptions) needs 0.82% all-in before any risk premium, which the taper’s yield path crosses at roughly 48.5M staked. But fixed costs bind at the floor of the solo distribution, not across it: the same $500 is 0.82% of stake at 32 ETH, 0.21% at 128 ETH, and 0.013% at a consolidated 2,048-ETH validator.
Which is why the size distribution of solo-staked ETH is, to my knowledge, the single most decision-relevant number neither side of this thread has. I think EthStaker has made attempts in the past, but nothing conclusive came to my knowledge and I highly doubt we can ever get good estimation. If solo stake is concentrated at 32 ETH, your objection binds on the whole cohort; if it is consolidating post-Pectra, it binds only at its floor.
When you ask “who is still staking at 0.7%?” this assumes that 0.7% is where the market rests. It rests there only if 0.7% is the marginal staker’s premium, in which case the ratio has already stopped near 42% of supply, comfortably below saturation, and the zero-yield regime never arrives.
I want to take this head-on, because for me this is the strongest sentence in your post. I think it proves more than you intend. Cost-per-validator and MEV-capture advantages are relative: they favour the same operators at 3% yield as at 0.5%, under the current curve as under the tapered burn.
The revenue level neither creates nor removes them. So if cutting revenue cannot fix a composition problem, it follows equally that keeping revenue high is not protecting composition either. I half agree with you composition needs composition tools like MEV burn (which this proposal considers an important followup), correlated-failure penalties (maybe @OisinKyne?), inclusion lists, etc. under whichever curve. Then the issuance level should be set by the things the protocol does control: dilution, the existence of an equilibrium, and whether growth keeps paying.
For the sake of the argument, your points in the first section needs institutions to be yield-insensitive (“will stake at 0.13% if the alternative is 0.00%”; “stakes because it is a product line”) while section 5 needs the same institutions to be yield-sensitive (“that demand exists because ETH became the first major crypto asset with a regulated, distributable, forecastable yield”). If they stake regardless of yield, the ETF demand channel survives the tapered burn and so the demand-destructive case kinda falls. If instutions respond to yield, they exit long before saturation and are likely to seek better risk/reward options (including just holding or lending). Do you agree?
The survivors you refer to (“entities that stake for structural, regulatory, or product reasons, which describes exactly the KYC’d, jurisdiction-bound, coercible operators the proposal fears”) do also have exit points, because “structural, regulatory, or product reasons” all cash out as somebody’s economics eventually. For example, an exchange taking ~25% of rewards is running a product line on a quarter of approximately nothing. A treasury booking 0.13% gross against custody and audit costs is booking a loss. An ETF whose distribution yield nets to roughly zero is marketing a staking feature that no longer sells (on the same screen as the Solana product you cite, by the way there’s an ongoing issuance reduction vote in Solana land). I wouldn’t consider them less yield-elastic per se but still, sooner or later they exit too.
Now onto what you are asking for:
My short answers will be:
- Why not, we could do that and/or also try to rally a meaningful group of solo stakers endorsing this proposal.
- I feel that this is out of our scope and I’m not sure how tax opinions will be considered or challenged in this context. The burn is applied before the rewards ever reach the accessible balance of the user. The “mint less” approach creates difficulties on the microincentives beyond 50% but could nevertheless be an option.
3. 4. Why not. Do you or AAVE want to partner on this ? Based on recent discussions with Mike from ether.fi they may be willing to help too. I’m mixing 3 and 4 since hypothesis from 4 should be used in 3.
- This is good feedback and we can consider it in multiple ways. Do note that to our core objective is deter the deposit size from crossing the 50% mark so this non-zero floor has to be low enough or progressively tend towards zero over time (which can be done from a transistion period tweaking).