This is the most measured and fair explanation of the failings of the issuance debate to date. If you want to see what's yet to be addressed, see how this thread evolves (and ask the cap authors to address them asap).
These concerns have been raised consistently for 2+ years, and they are still yet to be addressed, we are not ready to move forward with a proposal to change issuance without a proper data driven deep dive into these challenges. Forcing a change through without measuring the problem and discerning whether its real or theoretic, and without ensuring the cure isn't worse than the disease is a recipe for disaster.
Minimum viable security that doesn't mean 'anti-fragile as fuck' won't cut it. Social slashing is not a realistic threat right now and won't protect us from a centralised validator set. Ethereum is valuable because its a permissionless, decentralised, world computer, not because of its issuance rate. Sacrificing the former for the latter, will dramatically devalue its value proposition in my eyes.
I also don't believe the current curve is sacrosanct, but I think a constrictive curve is more dangerous and worse than the status quo. I think the burden of proof to say otherwise is high, and I don't believe that work is yet done, nor will it get done in time for a decision on H-star. I'm happy to share what I know about costs and staking economics, to help us pathfind to a future with a long term resilient and secure Ethereum, and right now I think that means choosing CROPS over hard money for the network.
I think that EIP 8361 tries to do too many things (increasing the moneyness of ETH, pre-empting remotely possible future decreases security due to overstake, protecting solo stakers, etc) at once and in my estimation will mostly do the opposite. I honestly believe that issuance change is too complicated to treat with this urgency, that the far-future concerns are drastically overblown, and it's very bad timing to introduce such a divisive proposal, especially in such a way. It has also been pushed last minute in an attempt to squeeze into Hegota (still subject to the EIP Inclusion process), even though potential issuance changes had been communicated as per the current strawmap to be slated for I* at the earliest. IMO, this proposal lays Ethereum's hard-fought uniqueness (real decentralization and neutrality) at the sacrificial altar of ETH as money, camouflaged against an insufficiently scrutinized concept of "minimally viable" security spend. Unfortunately, it also doesn't come to terms with the fact the moral hazard of a "too big to fail" event is not necessarily mitigated by limiting staking adoption, only displaced, as ETH seeking yield will mostly just go to more risky and likely custodial solutions, in my estimation thereby increasing the likelihood of such an event.
Apart from misgivings around the governance and consensus (or lack thereof, given its wide-reaching scope and implications) process around this proposal, I find that the selective supporting research is too theoretical and is lacking from a macro, micro and behavioral economic modeling perspective, especially with regards to 2nd and 3rd order effects. This applies to especially to two aspects:
1) one of the central theses around the urgent concern about the stake rate (which is lower than most other PoS networks which are mostly fine) that also hinges on the idea that just because issuance is positive at 100% stake then there's necessarily incentive to do so (ignoring years of economic history and literature which illustrate that demand is actually expressed as a complex function of utility, and not merely purely theoretic expected value, and strong contra examples such as USD as cash/savings vs T-Bills); and
2) on the effect of drastically reduced issuance (and real validator rewards) on the resulting validator set, and the negative externalities that will be born out of this material change in monetary policy (and possible further changes which might be required for correction).
Due to the new proposed curve a very real possibility of a sustained equilibrium at or near 50% of Ethereum staked with 0 nominal (nevermind real) yield is now possible, which would essentially be a death-knell for the security of the network, as node operators and staking protocols which prioritize expertise, decentralization, and expensive know-how are priced out and large, centralized, minimal-cost parties take over the operation of the overwhelming majority of the network's validators. Unfortunately, at current ETH/USD prices, such a scenario isn't even necessary for this proposed issuance change to lead to a large consolidation of the operator set, which would drastically lower the network's Nakamoto coefficient and thus its resilience.
In an eventuality where few actors are allowed to control one of the fundamental levers of the network -- determining its security (i.e. control of fork choice) -- by simply being able to offer staking at break-even or at a loss for long enough to price most other actors out of the market, what will be left of Ethereum's neutrality when the only recourse (now made ever-more necessary) will be a substantial and messy social slashing? Not much.
I'm not against an issuance reduction in general; I am against an issuance reduction without centering on the core questions, and giving the community adequate time, tools, and analysis to be able to consider the bevy of implications. To me these core questions are:
* What is the right amount of economic security given that we want Ethereum to be the centerpiece of onchain finance? A minimum viable security approach here doesn't cut it -- nation states don't think this way and the largest financial network in the world shouldn't either
* What is the effect of the market structures generated by any issuance curve, the relevant expected prices of operation (eg in ETH/fiat terms), and the effect of issuance changes -- and their consequences -- on demand for the asset, especially on the underlying validator and node operator set?
* How does the network retain its neutrality, utility, and demand, if social slashing is utilized not in the same way that nuclear arms are (as a last-measure deterrent) but rather as as the go-to solution for problems created by the network's own security paradigm?
* Why do we honestly believe that the terminal staking rate is in the high double digits? What practical economic modelling or historical indicators are there to point towards this?
* Are we okay with a hyper-centralized validator set just because if they misbehave we will socially slash them? Why have PoS at all then and not just go to PoA or PoG?
더 보기