EIP-XXXX: Tapered Issuance Burn

EIPs EIPs core
1

Discussion topic for EIP-XXXX: Tapered Issuance Burn

This EIP implements a modification to the ETH issuance curve by way of a partial burn of validator rewards.

References

This core principle is in large part based on prior work by pa7x1. A key refinement - the per-duty version of the burn, is due to Anders Elowsson (key insight contained in this post, see also his issuance FAQ).

Additional prior research on this topic over the last few years is indexed at issuance.wtf.

Update Log

2 Likes
2

After taking an initial look I have several comments, but before diving into specifics let me raise a point on the process.

This post comes 48 hours before the Hegota PFI deadline. Is the intention to propose it for inclusion with less than 48 hours available for feedback?

This clearly doesn’t leave adequate time for community review of a monetary policy change of this magnitude. Note also that according to the strawmap, the expectation was that an issuance update would be considered for I*, so several participants may be preparing according to that schedule. Moreover, legitimate arguments have been raised in various prior discussions about consequences of an issuance update that need to be considered but don’t appear to be taken into account in this proposal — more time for open public discussion is clearly needed before this enters consideration for any specific hard fork

3 Likes
3

Ethereum direct economic security (stake under slashing) is very strong for a very long time. Strong enough that any potentially successful attack on Ethereum will have to be indirect. It’s going to be through identity hijacking (e.g. stealing validator keys), custodial intermediaries compromise, government coercion, supply chain attacks etc. Threat model is not a malicious open market ETH buyer. A strong proposal to tweak Ethereum security has to quantify that in some way; otherwise, the change will about as arbitrary as original curve or mechanism design. The proposal doesn’t make an attempt to do that.

One reason I’m raising this question is that I can tell from the get go that practical security of Ethereum consensus against indirect attacks is strained at this moment. Ethereum first line of defense historically is diversity and decentralization of node operators. Many professional node operators (who run most of Ethereum stake) are barely breaking even. Cyber security weather is the worst ever, we’re in the slow crash era of LLM-driven cyberattacks. And ETH price dollars, which you need to spend to pay for expertise, is not feeling so good.

Making economic conditions worse will forcefully transform the security model from decentralization-driven to concentrated professionalism-driven, switching the market model to a much smaller consolidated set of operators. It is certainly a choice one can make. But I think that without understanding of first and second order effects on practical Ethereum consensus security and brand value it’s a bit of a yolo one. Tight timing on decision making and closed doors proposal development doesn’t help either.

For the most technical aspects of proposal:

  • don’t get why it’s designed as “mint and burn” instead of “mint less”, especially that rationale for the change is partially driven by the tax law issues. Why not go for less tax ambiguous option?
  • I like that it’s got a slow adjustment period. Gives an opportunity to roll back if it turns out that validator set we’re getting is not secure enough.
3 Likes
4

Thanks for the feedback and taking the time to read through it!
Note that PFI does not mean inclusion. Being proposed for inclusion is what opens the floor for feedback, not what closes it. But yes, this EIP does aim for Hegotá, for two reasons.

First, the topic is not new, and there is plenty of time to discuss it. Discussions and feedback will have several months to take place (starting now :tada: ). It’s worth mentioning that we discussed a similar proposal back in 2024, and the wider debate has been running since 2023 and much of the literature is collected at issuance.wtf.

Second, the clock is ticking. Waiting for I* or later effectively means abandoning the change. The primary goal of this EIP is security and capture-resistance, which is met by the burn fully cancelling issuance at 50%: issuance no longer incentivises staking beyond that ratio. The arithmetic argues strongly against delaying, even under conservative projections. Validator entry is rate-limited by the protocol, and the activation queue has been running at essentially that maximum for a while already. If entry stays saturated every epoch and few stakers exit, by January 1, 2028 more than 70M ETH will be staked (north of 55% of the supply).

Acting now means the market settles into an equilibrium below 50% but acting after the overshoot means correcting a much larger imbalance, with more stake forced to exit and more disruption for every participant. The gentle path is only available now.

It is also worth noting that this king of reform gets harder every month. A larger staked base means a larger constituency earning fees on the status quo. The previous issuance debates already showed how effectively that pressure can stall a proposal.

Proposing the change now means we have enough time to review it, and that the correction stays soft: the reduction phases in over 18 months, plus the usual lead time from scheduling, giving participants on the order of two years to adjust.

5

As written, this EIP damages Ethereum’s long-term staked economic security, hurts its defi ecosystem, potentially destroys its decentralized validation model, and leaves a whole host of questions unanalyzed and unanswered. Some of these important questions are:

  1. What is the appropriate level of staked economic security to secure the anticipated multitrillion dollar onchain economy?

  2. What happens to solo stakers and Ethereum’s decentralization model as yields tend towards zero?

  3. What happens to LSTs (an important defi money lego) as yields tend towards zero?

  4. What is the anticipated continued centralization of stake to the operators with the lowest marginal costs of operation as yields tend towards zero?

  5. How does the narrative of ETH being a kind of “bond of the internet” change as yields tend towards zero? (BTC is already a proverbial “pet rock”)

Answers to the above are not much to ask for, but a well-researched attempt at addressing even #1 above would be a good start.

2 Likes
6

Thanks for this.

Long-time staker, never contributed to discussions.

This proposal:

:check_box_with_check: a. has a mechanism to reduce staking incentive above 50% staked; cool.

:cross_mark_button: b. claims to be in the benefit of solo stakers; ??

One section argues that the current issuance curve is hostile to solo stakers:

“Solo stakers are forced out. Dilution erodes everyone’s real return as the ratio climbs, but solo stakers… cross into negative dilution-adjusted returns well before large operators…”

Most would agree.

However, later the proposal analyzes operator incentives primarily in terms of consensus issuance revenue:

“An operator’s income from consensus issuance is its share of the stake multiplied by the total amount issued.”

That seems to reduce to:

Operator consensus revenue = market share Ă— total issuance

My confusion is that this is a statement about revenue, whereas operators ultimately optimize for profit.

The protocol treats every validator the same, but the economics are not the same.

A solo staker may have materially higher effective costs due to hardware, operations, downtime risk, and (depending on jurisdiction) taxation. A large staking provider can spread many of those costs across thousands of validators and may have more favorable operational or tax structures.

If consensus issuance is reduced across the board, it seems plausible that the higher-cost participants become uneconomic before the lower-cost participants.

So the question I’m struggling with is:

What evidence or model supports the claim that lowering staking APR improves the composition of the validator set, rather than simply reducing the staking ratio while increasing concentration among the remaining validators?

I understand the argument that reducing issuance can reduce the equilibrium staking ratio. What I don’t yet understand is why that mechanism preferentially disadvantages large operators rather than solo stakers.

I’m asking this genuinely. It feels like there is either:

  1. an assumption I’m missing,

  2. a complementary mechanism elsewhere in the proposal, or

  3. empirical evidence that institutional staking demand is more yield-sensitive than solo staking demand.

If it’s (3), I’d be very interested in seeing that evidence, because it’s the step in the argument that I’m currently having trouble following.

1 Like
7

There is a reasonable mid ground in these discussions that centers around the actual liquidity within the various DeFi protocols, and how it relates to staked ETH.

To put into a few bullets, I think the fears of stakers and wider ecosystem participants can be summarized by @CelticWarrior’s post above, with the core of it being:

  • Will yield actually trend towards zero?
  • How will that impact DeFi?

The reality is that all of this is impacted by the same question – who is the most yield sensitive staker, and what are the limiting factors of their participation? Out of the major participants of solo stakers, institutions, CEXs, and LSTs, the clear yield sensitive participant is LSTs and looping given the liquid nature of the instrument.

Next is then identifying the bound to LST and looper participation. That being said, if you want to retain these participants and keep the most major DeFi markets afloat, you need to ensure that the staking yield is always marginally higher than the lower bound of borrow costs. E.g. As a mid-large size LST looper, my target is roughly ~.3-.5% higher staking yield over borrow cost, giving me a 5-8% APY when levered ~12x at the current ~2.5% staking APY.

The market will roughly ensure this is the case (yield wont trend towards zero, and a margin between staking and borrow costs will keep DeFi alive), however it may be prudent to reduce the rate of this impact or to cap the lower end of the emission tapering, leaving most participants happy.

TL;DR: Fears of DeFi dying are overblown, but loopers will be impacted most. If the EF wants to keep native DeFi as a priority the suggestion would be to implement a much slower time factor to the tapering and/or a higher lower end cap of ~1-1.5% rather than a trend towards actual 0.

Edit: Also apologies for joining the issuance discussion so late

Edit 2: I would also suggest that lending markets can accommodate changes by adjusting their utilization curves for staked ETH

1 Like