Practical endgame on issuance policy

[aelowsson]

@keyneom

I don’t know that I think my points are being fully addressed (I might not have explained them well enough–like the point of a billion Bitcoin miners participating in consensus at once). Here are probably the two major points I’d contend with from your reply. Apologies that I’m not addressing all your points either.

It is quite natural to not perceive all points as fully addressed from the first response in a discussion. This often requires further back and forth.

@keyneom

@aelowsson

This is literally described in the post:

Isn’t this agnostic to market cap? If eth market cap is $100 vs $1 trillion that significantly impacts the security of the network in being able to withstand a nation state level attack. One requires almost no effort to buy enough supply to control the network and the other is much more difficult to do so. Limiting the amount staked makes this even more impactful. That’s my point. It is totally subjective what is “enough” economic security.

The focus of my response was to the following statement: “If there are 100MM eth and 50% is staked that’s very different than only 50MM eth and 50% staked.” Under the current issuance policy, Ethereum is incapable of handling this scenario correctly, because the offered yield would differ since it varies with deposit size and not deposit ratio. So I explained the plan to rework our issuance policy to handle that. This has all been worked out but will be phased in over several hard forks. In essence, we will track the circulating supply and swap out D for d in the equation of the reward curve, also including the circulating supply S.

When it comes to the influence of market cap, this was described in the first paragraph of the link I provided as the start of my answer. If you could find the time to review these links, it would be very beneficial, since it takes up a lot of time answering these questions.

@aelowsson

Ethereum’s economic security will in the long term inherently be linked to the ability of ETH to retain its value. A holistic perspective is therefore important also when considering economic security. This is underscored by reflecting on the early days of Ethereum, when the ETH token was much less valuable. Eight years ago, In May 2016, Vitalik deliberated on the value that Ethereum can and cannot secure at a stake participation of 30%. At the time, the market cap of the ETH token was roughly 500 times lower than today, and the economic security that Ethereum could offer was therefore limited. Increasing stake participation from 30% to 60% would only increase the value of 1/3 of the stake (the threshold for the ability to delay finality) from $70M to $140M. This highlights that once the proportion staked has risen above insignificant levels, it will ultimately be Ethereum’s role in the world economy, and the ether’s role in the Ethereum economy, that determines the economic security.

The contention is thus that we cannot focus too closely on market cap when specifying the reward curve.

@keyneom

@aelowsson

Why do you feel the need to try to coerce users to transact when they do not want to?

No one is coercing anyone. I could ask you the same question. Why do you feel the need to try to discourage users from transacting when they otherwise would?

Clearly, you consider issuance policy as an instrument for affecting transaction demand. You discuss inflation and deflation, sharing opinions on how it will affect user behavior, with quotes such as “the incentive people have to spend immediately if they otherwise get diluted”, etc.

This is however not the purpose of Ethereum’s issuance policy. We do not and shall not rely on issuance policy to affect transaction demand, but solely to affect the willingness to supply stake. Whether ETH deflates or inflates does not factor into the equation. Thus, I do not “feel the need to discourage users from transacting”, because I know that it is a fruitless endeavor. I am strictly focused on security and composition of the staking set.

@keyneom

The only way of avoiding this is net 0% deflation/inflation. I already mentioned that would be ideal but I don’t think it is long term feasible to have a fixed supply. Issuance paired with the burn is a great approach but will almost guarantee that one is stronger than the other. I’d rather lean towards slight inflation than deflation.

There is nothing to avoid, because we do not try to influence transaction demand with issuance policy. Slight inflation implies that Ethereum is incapable of securing its network when relying on current revenue. For this reason, deflation is a healthy sign for any cryptocurrency.

@keyneom

…once you have your fixed costs addressed the marginal costs of adding another validator are almost nothing which can make it very hard for a new LST upstart to compete. In the end I just don’t see us significantly preventing LST dominance much more effectively than we already do as a result of modifying the issuance curve.

An LST that provides a five times lower staking yield to its customers will have a very hard time to compete. But this is not integral to the discussion, let us move on.

@keyneom

You haven’t even demonstrated an actual need for MVI as I indicated in my second post. If people aren’t agreeing at least on that high level point first you will never have success convincing them of some random other arbitrary curve that feels nice to you.

The primary motivation for a reduction in issuance is the reduction in costs, leading to a welfare gain. I have linked the relevant write-up and figure on this several times, but will present it here also, if you perhaps missed it.

Two reward curves are shown in the figure, both under the present level of MEV (300k ETH/year) and the hypothetical supply curve used in the post. The supply curve indicates the amount of stake deposited D at various staking yields y. It captures the implied marginal cost of staking. What that means is that every ETH holder is positioned along the supply curve according to how high cost they assign to staking, with their required yield implying that cost. Relevant costs include hardware and other resources, upkeep, the acquisition of technical knowledge, illiquidity, trust in third parties and other factors increasing the risk premium, various opportunity costs, taxes, etc. The area above the supply curve indicates the stakers’ surplus (what they actually gain) and the area below the supply curve the costs assigned to staking (the marginal staker would not stake at a yield below the supply curve).

By maintaining the current reward curve in black, Ethereum compels users to incur higher costs than necessary for securing the network. Adopting the red cubed reward curve from the post eliminates implied costs represented by the dark blue area, thus improving welfare. These implied costs amount to around 344 000 ETH, corresponding to 859 million dollars.

The issued ETH covered for hardware expenses, taxes, reduced liquidity and risks that users would choose to sidestep under a lower yield. With the red curve, they can, and the benefits are shared by everyone (including remaining stakers), creating value for all token holders through a reduction in newly minted ETH. It may seem strange that it matters how many new ETH that are created. But imagine if every ETH was converted to 10 ETH tomorrow so that there were 1.2 billion instead of 120 million ETH in total. Then every ETH would be around 10 times less valuable. What matters is the proportion of all ETH that you hold. There is also a surplus shift from stakers to everyone from a change in issuance policy, indicated by the darker grey area. That ETH indeed previously benefited stakers, since it was taken from everyone (in the form of newly minted ETH), and then given specifically to them.

It is important to understand that the cost reduction corresponding to 859 million dollars is the yearly savings, not a one-off event. The implied cost reduction applies each year, benefiting Ethereum’s users on an ongoing basis.