But couldn’t you just have whatever $ amount that’s staked

in RPL for this purpose just be ETH but kept as liquid/insurance and not actually staked? Like there’s no actually special properties about RPL itself right, you just need liquid collateral as a deposit to discourage bad behavior?

I mean I guess you could. That’s just a different design. And if you care about capital efficiency, the model you’re describing is pretty inefficient. You never want idle eth just sitting there

Right now when a user comes to Rocket Pool to become a node operator, do they get rETH for their 8 ETH collateral?

Eh I mean I’m not sure there’s a meaningful distinction. I can either put eth up directly or swap my eth for RPL, like in that 2nd scenario there’s not eth sitting idle with Rocketpool but now there’s more eth sitting idle in the LP contract. Like I get why from Rocketpool’s POV they’d like to drive demand for the token but not clear to me this keeps more ETH non-idle or has an additional safety properties. Seems the core point is you just need liquid collateral that can be slashed on demand.

Again, I’m not on the rocket pool team so I can’t answer for them. But at a certain point you either believe that tokens have utility (in this case, the utility of RPL is insurance on rETH and leveraged staking returns on staked ETH for node operators). You could easily characterize that as just a bullshit wrapper around eth but then you kinda got to wonder which tokens are not that

I’ll just say I don’t think requiring RPL is necessarily a bad or illegitimate design. I think it’s fine for protocols to try to figure out ways to drive organic demand for the token aside from governance or direct rev share. Just saying I don’t see any safety properties unique to RPL, seems like it works but just because it has the property of liquid collateral which would also work for any other sort of liquid collateral.

I agree. There's nothing wrong with using RPL. In fact, it brings extra utility to it and aligns incentives of node operators with RPL holders. It just makes the loan costlier. But for that cost, you get more incentive alignment and RPL demand in return. A fair and proportional trade. It just lowers the LTV of the operator loan.

Yeah I get the same idea from reading the above. The argument that it couldn't be ETH as people would want to have that in validators insteas seems a bit moot. As any collateral that is posted could be sold for ETH in principle.

But why RPL? Could be anything right?

It could be anything yes. But let's be real, outside of ETH or RPL there is no reason to make it any other token from Rocket Pool's perspective. It would be silly to be anything else.

Of course, thats why Im questioning it. As there seems no (current) specific reason to exist now. Maybe more alligned utility will be addec in the future. I mean everything is very early. But I miss the inherent real economics now

Went to bed last night but saw the RPL discussion kept going so I’ll address it. Insurance is needed but it should just be in ETH. Yes RPL helps align node operators and creates short-term demand-based value accrual. But the demand it is based on is primarily NEW node operators entering the system and posting RPL collateral. New node operator growth is much slower than rETH demand which is why Rocket Pool reduced requirements from 16ETH to 8ETH to create more borrowing amongst existing operators (but many of these have already seen their RPL double so don’t need to buy more RPL ie diminishing value accrual). Meanwhile a new NO with 8ETH is required to speculate a greater amount on RPL as “insurance”, hurting adoption. That is the issue.

Rocket pool has mechanism that try to balance node operators and rETH demand. If either grow too large, revenue/value gets tipped to the other. If you are interested in the mechanism, go to their discord. For the record, there is currently SIGNIFICANTLY more node operators waiting for ETH than there is demand for rETH at this particular moment.

I understand the mechanism. I poured through the docs/forums/chats over the past several weeks. The system is always imbalanced because node operator growth is so much slower than rETH growth. rETH was bottlenecked for a long time due to lack of node operator growth. Atlas solved this by going from 16ETH to 8ETH effectively tripling capacity by allowing existing node runners to go from one 16 minipool borrowing 16, to two 8 mini pools borrowing 48. But this is not sustainable for RPL value accrual.

Ya this is sort of what I was saying that it does create demand for RPL but that demand is sort of capped and will actually decay. There's no recurring demand from existing NOs and it also can't scale with additional NOs coming into the system since ETH is deflationary so there's basically a fixed limit to how many NOs there can ever be. Like it can keep growing while ETH staking is in this expansionary phase but eventually that has to slow and more or less stop.

But isn’t that true for all staking in general?

yes growth of all LSDs will slow. the issue is that RPL is pumped by growth and will crash when growth slows. and by that I mean new node operator (who have to buy RPL) growth, which I expect to flatline within the year. rETH can continue to grow as they increase borrowing amongst existing node operators LEB16->LEB8->LEB4. they can change their tokenomics but the longer they wait the harder it gets and the more node operators get hurt