Eth 2.0 Staking Returns

staking
economics
#1

I recently wrote a post which walks through the basics of Ethereum 2.0 economics. On top of that Ivan Martinez, software engineer at Prysmatic Labs put out a tweet thread which argues for the community to come together and discuss the staking economics behind Eth 2.0.

I’d like to start to gather thoughts from people on the current staking returns. At a total network stake of 10,000,000 (which appears to be a nice target for now), we’re looking at 2.5% return to stakers. That’s much more competitive than any DeFi option for lending ETH at the moment (usually ranged between 0.1%-0.5%) but it’s barely profitable giving validator costs at current price.

My hunch is we need to hike the rates around 25-35% versus in the current spec but interested in everyone’s thoughts!

2 Likes
#2

I agree with you about the necessity to hike the rates. High ratess allow to contain the strong price fluctuations and allow the effective participation of small stakers.

#3

I’ve been a vocal advocate for a higher issuance rate during phase 0 for a while now so I’ll give some my thoughts here.

Given the fact that the beacon chain will be launching as a stand-alone entity (and will remain that way for 1+ year), I think we should experiment with very high rates (as high as 50%) to monitor the behaviour of participants/validators/stakers in the network. This issuance rate will adjust according to how much ETH is staked especially since that during phase 0 the bridge is currently unidirectional (from eth1.0 to the beacon chain) so we may see a lower staking participation than we would if the bridge was bidirectional.

The table here shows what a variable issuance rate might look like.

We can of course experiment with this once we have a proper cross-client testnet available but I don’t think we’ll get a great idea of how stakers will behave until it’s on mainnet and there is real money at stake (pun not intended).

3 Likes
#4

I am still not happy with the choice to make the ETH —> BETH transfer unidirectional. This choice alone requires enormously high staking returns (lack of liquidity & utility, uncertainty about bugs & adoption).

One way to combat this would be paying out the staking returns on the ETH 1.0 chain.

1 Like
#5

Can you make a separate thread for this? It’s a valid concern and something that should have its own discussion area.

1 Like
#6

I agree that the staking economics are critically important. I am not sure where I sit on the sufficient size of returns yet. Some related thoughts:

10,000,000 ETH is about 9.5% of total supply, which might be locked up in a relatively short span of time. This will likely drive the current price of ETH up, particularly due to the abruptness of the reduction in supply. It will be interesting if this starts a feedback loop in which the more ETH that is staked the lower the supply, thus the higher the price, and thus the more incentive to stake ETH for returns in an appreciating asset. Of course, eventually stability will be found, but we might be surprised by how much ETH is staked. I think we need to consider rate economics at $150, $1500, and yes, even $15000 ETH. What I mean to say, is that short-term return economics might not reflect successful long-term economics.

IMO rates on other DeFi options like Compound will probably adjust towards whatever the staking rate is, as a move towards staking will reduce their supply.

1 Like
#7

For me 2.5% isn’t a great incentive in the current economy where I can realize similar returns from t-note yields - I’d say staking during phase 0 is higher risk. Under these conditions maybe we’d see less staked supply and thus a better return, I need to gander at this somewhere other than my phone, but assuming the network is reasonably secure at 5mm staked then this may be fine, if not we should definitely reassess inflation.

#8

One thing with this though is after inflation is basically 0 return. I think we need to compare to other ETH lending alternatives like defi for competitors. That’s of course assuming the stalker already has ETH. If not and is being considered as a new investment vehicle, that’s a different story.

This is something I’d love to discuss with the researchers. What is their ideal target?

#9

I should have some free time tomorrow, I’ll see about doing some quick calculations about the cost of compromising the network at various levels of staking participation :grin:

1 Like
#10

Decentralised Staking Networks. Mainly the current project ‘rocket pool’ are going to play a large roll in proof of stake, and will effect its return profile. I think this is massively under looked. Stakers will be able to increase there returns by staking with rocket pool. Inexperienced users will be able to stake without much knowledge.

1 Like
#11

If I’m reading this correctly, you’re saying that if I already have ETH and plan on holding it, then staking at 2.5% may be OK (which makes sense). I’ve been accumulating ETH with the hope that the staking rate would be comparable to other traditional investments, but 2.5% is not going to entice me to stick around. Yes, I’ll keep some to see how it appreciates, but I’d rather put funds into a tax free bond, or something comparable. I’m also likely older than most folks here and that makes a difference too.

#12

One thing to consider here is that a tax free bond is barely profitable given USD inflation. Once we are on Eth 2.0, network inflation rates should be <0.5%.

I think most holders are in it to speculate on ETH so the added staking interest on top is just a bit of icing on the cake. I really doubt we’ll see too many external investors seek out ETH just for the staking returns.

I hear you overall though and why I’m advocate for boosting them some.

#13

Really great discussion so far! I have a few thoughts:

  1. The starting rate should be low - the first folks to transfer ETH will be very long term holders and are already anticipating holding their ETH for years whether the rate of return is at 2% or 5%. As Eric mentioned - we’re not expecting external speculators to enter. We just need to ensure we have enough staked to launch the beacon chain and an 8% return at 1m ETH staked is sufficient motivation.
  2. Long term, I expect ETH validator returns to be roughly 4-6%/annum. This is in line with the security spend (block reward + fees) for Bitcoin, Ethereum, Monero, and Tezos today. As such, I expect the inflation rate to increase from current targets.
  3. Regarding profitability, most folks will be running many validators simultaneously and will have no trouble making a nice return. However, I disagree that this will lead to a proliferation/dominance of staking services. We at Bison Trails will support ETH 2.0 and make it really easy for anyone to spin up their own beacon nodes and validators (entirely non-custodial). Whales will run their own infrastructure if they have strong engineering teams or use our service to participate in the network while smaller folks will pool on staking services.
  4. We don’t need to perfectly solve this problem today. If we’re launching shards next year and have too little ETH staked, we can increase the reward rate to get to the desired validator #.
  5. If we accept we’ll be off, I’d rather come in too low and adjust up than the inverse to minimize dilution of existing token holders.
#14

I’ll speak as someone who wants to do staking, but doesn’t want to operate their own node. Slashing scared me away from that, which is why I like the idea of RocketPool. Now, I know this goes against what ETH 2.0 is supposed to be about, but I would rather avoid slashing and put my 32 ETH into a pool. As far as staking returns is concerned, I believe the initial returns do need to be large in order to provide an early incentive for protecting the network.

1 Like
#15

Agreed. Mithril Ore does this in a different fashion. The more parties doing it the less any one can be nefarious — regarding ethereum POS security.

Mithril Ore token = the title to the ethereum being staked.

1 Like
#16

Ethereum initial proof of stake - is more risky than a bank certificate of deposit (CD) but akin to locking up capital in exchange for a percentage gain over time.

Current bank CD rates for 1 to 5 years are 2.8% - 3.1%. Then — plus the risk of slashing - assuming 95% uptime - would say that the number of 2.5% is far too low given this concept. I would need some help with calculating slash risk but at least 3.1% ROI- (if not 5% given risk).

Also, many thanks to Eric Conner for all of the proof of stake work/organization he is doing!

2 Likes
#17

Is there any benefit in giving validators a premium for uptime, or runs of uninterrupted uptime? Not sure if it is doable, and perhaps unnecessarily complex.

Basic example: if 2.5% is the base rate at the current participation, a validator might get something like +0.02% for % uptime greater than 75%. Thus, a validator with 80% uptime would be earning 2.6%, a validator with 100% uptime would be earning 3.0%.

If slashing has a non-linear downside for downtime (when blocks aren’t finalizing) should rewards have non-linear upside for uptime?

#18

That makes sense, but to me, it’s a bit unfortunate. Staking will therefore do little, or nothing, to increase adoption, which is what I was hoping to see. It will be great for folks already holding ETH though.

#19

I think for now it will do very little to bring in others. Volatility will always attract outsiders over small interest gains.

However, imagine a scenario where Ether has settled into being the global money of the internet and is stable. At that point, I could envision people seeing it as an investment when it comes to only staking.

#20

I think we underestimate the amount of eth hodlers that will stake less because of economic considerations but more due to their belief system and ethical responsibility to Ethereum. Yes, I do not think we should depend on this longer term, but it is worth thinking about.

2 Likes