This is not directly related to the content of the post but is about Ethereum. I have been putting this hypothesis out there on twitter for a while now, looking for someone to refute me, but no one has yet. This seems like a place where I might be able to find someone who understand Ethereum enough tell my grasp of the protocol is wrong (or right):
Ethereum is going deflationary. Not disinflationary like Bitcoin, but actually deflationary, with token burn. Whether you think deflationary money is bad or good is not germane to the problem I'm about to explain, as I see it.
Ethereum makes a claim at decentralization. However, by moving to a Proof of Stake system, they have set a fixed amount to be staked (32 ETH in this case, though the number itself does not matter). With the token going deflationary, that means that the opportunity cost to stake will always be increasing, mathematically, in ETH terms. The staking requirement does not adjust to changing token supply.
For argument's sake, if there are 100 ETH on the network, and you stake 1 ETH, you effectively need to lock away 1/100 of all wealth on the network. However, if 10 tokens get burned, now you lock away 1/90 wealth on the network. Staking that 1 ETH just got 10% more expensive because the opportunity cost of staking that ETH increased relative to spending it.
Economically, this means that the system is going to drive out marginal stakers, by design. The only response I have gotten from ETH holders is about staking pools, but staking pools don't fix the underlying economic incentive brought about by deflation: increasingly holders will find it less worthwhile to stake their ETH.
It seems to me that ETH is headed right for some kind of centralization death spiral, as only increasingly wealthier holders will be able to justify staking.
> Not disinflationary like Bitcoin, but actually deflationary, with token burn
It's possible there's some recent development I'm not aware of so please fill in any gap here, but my understanding: Token burn is already in place since EIP1559 (base gas fees get burned now, instead of getting paid out to miners/validators. There is a separate priority fee for incentivizing inclusion in times of congestion). It will work similarly on ETH2. As laid out in EIP1559, this means that the ETH supply may be inflationary or deflationary at different times, depending on if total base fees are greater or smaller than miner/validator rewards. So it's not a given. However I could agree that given continued growth of Ethereum adoption, the deflationary hypothesis is not unlikely.
> Staking that 1 ETH just got 10% more expensive because the opportunity cost of staking that ETH increased relative to spending it.
Can you explain how this follows?
Shouldn't it be the opposite; as long as you don't require the immediate liquidity, deflation should mean that it's actually preferable to hodl, and staking rewards compound with that.
Aside from that, due to MEV, being a validator means you get reserved slots where you can [extract MEV yourself/safely submit transactions of your own without being front-run]. This is something that is today primarily taken advantage of by bigger or more proficient actors - meaning yes, this is a force towards validator consolidation. On the other hand I believe the "flashbots market" will likely become more efficient and accessible to the point where commonly used validator clients will have built-in interfaces to exploit this through an open market for privately broadcasted transactions (as mentioned in references this already is the case for ETH1). This means that over time the playing field gets leveled as getting a cut from MEV gets accessible for smaller non-professional stakers as well.
Looking at the current direction, I would agree in the sense that we're on the track for consolidation of staking power and resulting compounding centralization of ETH wealth, but not for the reasons you're stating, and I would not agree that smaller stakers are driven out by design - it works out that way despite the small-staker-incentive situation, not because of it.
Future EIPs may change the dynamics further in either direction. I don't think it seems obvious either way.
Bitcoin's supply is better described as deflationary since it's capped and Bitcoin experiences ongoing losses. Eventually there will be fewer and fewer bitcoin in circulation. This will happen way before Bitcoin emission ends in 2140, since in 2061 for instance, only 320 BTC will be mined, and yearly losses will easily exceed that.
Staking pays all stakers out proportional to their stake, which is as fair as is possible, and more fair than PoW. Distributed staking pools lower the barrier to entry to 0.01 ETH.
Ethereum will still be printing ETH in perpetuity; it only becomes deflationary whenever more is burned than printed.
> Staking pays all stakers out proportional to their stake, which is as fair as is possible, and more fair than PoW. Distributed staking pools lower the barrier to entry to 0.01 ETH.
This doesn't change the economics of what I said, which are about opportunity cost of staking vs. spending.
> Ethereum will still be printing ETH in perpetuity; it only becomes deflationary whenever more is burned than printed.
I understand this. But what I've been hearing from ETH holders forever now, is that ETH will be going deflationary. As in, burning more ETH than printing is something will definitely be happening. It's the bedrock of the flipping narrative.
Based on current and projected usage vs staking rewards, it looks inevitable. However the rules that Ethereum operates by are not immutable. They have changed, and they will change again. So just because it looks like it will be deflationary this year, doesn't mean it will always be deflationary.
> Staking that 1 ETH just got 10% more expensive because the opportunity cost of staking that ETH increased relative to spending it. Economically, this means that the system is going to drive out marginal stakers, by design.
I don't really get it. It's more expensive because you're making more money? Why wouldn't everyone just stake until they feel like spending? How does this drive out marginal stakers?
> But what I've been hearing from ETH holders forever
These are just people pumping their bags. No one actually knows how it's going down. Anyone who holds such alpha isn't posting it on Twitter.
Eth will hit a stable point of total supply, with both inflationary and deflationary mechanisms mentioned in other comments finding an equilibrium according to market forces. It won't deflate forever, so there won't be a centralization death spiral according to your logic.
There is no magic in the mechanism behind ETH's inflation/deflation. If an equilibrium is found, it will be because the appropriate levers are pulled until the desired ratio is reached.
It's also worth keeping in mind that the levers needed may not exist within the system. ETH is still a massively ambitious and risky experiment. Anyone who says otherwise is naive or trying to sell you something. That being said, ETH is absolutely the most interesting thing happening in both tech and finance. I hope it works out.
What you seem to be afraid of is something that might or not happen in a distant future. At that point they just change the rules again. Ethereum has done it several times. If they don't like the direction things are going they just change it. Some people get mad about it but it just keeps chugging along and these people go out and possibly create a fork that goes nowhere.
I still think crypto is high priced poop people currently view as valuable. But I don't see how your point goes anywhere to prove it.
You assume everyone always chooses the option of highest return, however not everything in life is about opportunity cost.
I personally know two people running test stake nodes. They've spent the time setting them up, documenting, and really getting to know all the ins and outs of it because they find it interesting. They will run production nodes. You only need some X amount of people to do the same to avoid overt centralization. Then it just comes down to an argument of what is the minimum number for X where things feel properly decentralized.
The reason they require 32 ETH to stake is that there's a scaling limit on number of stakers. They're working on improving that, thus decreasing the staking requirement.
It's very likely to stay a power of 2, so just a single reduction down to 16 ETH would make up for 34 years of deflation at 2%.
1. Interesting question, and I haven't seen an answer that refutes your point, except in disputing that the supply of ETH will keep shrinking.
2. I'd like to distinguish an increase of the money supply from inflation. They're not the same. You're talking about a decrease in the supply of ETH, here, not deflation.
3. Related: While the supply of BTC is currently growing around 1.7% p.a., I wonder how much is lost to lost keys. Has that been estimated somewhere?
the premise that it will be deflationary forever isn't right. It can't be and it will find an equilibrium or even become inflationary again as dynamics around the burn and staking play out.
Ethereum is going deflationary. Not disinflationary like Bitcoin, but actually deflationary, with token burn. Whether you think deflationary money is bad or good is not germane to the problem I'm about to explain, as I see it.
Ethereum makes a claim at decentralization. However, by moving to a Proof of Stake system, they have set a fixed amount to be staked (32 ETH in this case, though the number itself does not matter). With the token going deflationary, that means that the opportunity cost to stake will always be increasing, mathematically, in ETH terms. The staking requirement does not adjust to changing token supply.
For argument's sake, if there are 100 ETH on the network, and you stake 1 ETH, you effectively need to lock away 1/100 of all wealth on the network. However, if 10 tokens get burned, now you lock away 1/90 wealth on the network. Staking that 1 ETH just got 10% more expensive because the opportunity cost of staking that ETH increased relative to spending it.
Economically, this means that the system is going to drive out marginal stakers, by design. The only response I have gotten from ETH holders is about staking pools, but staking pools don't fix the underlying economic incentive brought about by deflation: increasingly holders will find it less worthwhile to stake their ETH.
It seems to me that ETH is headed right for some kind of centralization death spiral, as only increasingly wealthier holders will be able to justify staking.