Community discussion: while creating incentives for pNetwork nodes, what fee model should pNetwork apply on cross-chain operations?

[bertani]

Community discussion: while creating incentives for pNetwork nodes, what fee model should pNetwork apply on cross-chain operations?

The pNetwork is the increasingly decentralised layer powering and governing the cross-chain pTokens solution as well as other features (example: pNetwork Portals). As part of the project’s progressive decentralisation roadmap, a growing number of components are being introduced to the system.

Node operators are a critical component for the pNetwork - they are responsible for securing cross-chain operations and increasing the overall security of the protocol (check out the state of the network at nodes.p.network).

For the ecosystem to experience healthy growth, it is essential to align incentives for all participants. Introducing a fee on cross-chain operations processed via pNetwork creates an incentive for entities to spin up a pNetwork node and support the network. From the opposite point of view, a network of nodes contributes to a more stable and higher quality service, making pNetwork more attractive for new users.

The PNT token represents a key element of the system as it aligns incentives for all participants. In fact, PNT is leveraged internally by the pNetwork system to enable operations for both node operators and DAO members. The flywheel effect is triggered by the need for node operators and DAO members to hold and stake PNT tokens in order for them to perform their respective roles.
Prospective node operators need to stake a minimum amount of PNT tokens (currently set at 200,000 PNT), which is then used to show their commitment and serves as a bond.

When operating a pNetwork node, node operators are eligible for DAO voting rewards (currently set at 42% APY, rewards distributed at the end of every epoch). However, a long-term incentive is needed for the node operation to be sustainable since DAO voting rewards are planned to stop at the end of the second year.
This is a proposal to introduce an additional incentive for node operators to reflect their key role within the ecosystem.

This proposal is about a fee structure that goes in the direction of incentivising the incoming traffic, therefore setting asymmetric fees on peg-in and peg-out operations. Further fees are added when leveraging additional features such as pNetwork Portals, enabling a variety of more advanced use-cases.

Below, a breakdown of the proposed fee structure:

The proposal sees a 0% fee for peg-in operations and a 0.25% fee for peg-out operations. Further fees are added when leveraging additional features - specifically, the proposal sees an additional 0.05% fee when transferring metadata across blockchains (i.e. this additional fee would be applied when using pNetwork Portals. An example could be the wrapping and un-wrapping of NFTs across chains), with a minimum fee. Should other features be incorporated within the pNetwork system in the future, similar logic would apply.

Example: smart contract interaction (metadata required in both directions) between Eos and Ethereum with attached a cross-chain transfer of $1,000,000 which goes one way and back.

The payment for the fees happens in terms of the asset users transfer cross-chain. As an example, while processing the tokenisation of Bitcoin on Ethereum, a peg-in fee would be collected by the system in BTC and distributed to pNetwork nodes.

Below, a chart showcasing the US Dollar countervalue of fees distributed to pNetwork nodes based on the cross-chain transaction volumes processed.

This post serves as the starting point for community discussions. Once a consensus on the above proposal or an iteration of it is reached on this board, a formal proposal will be opened on the pNetwork DAO for PNT holders and DAO stakers to vote on it.

A minimum of 7 days will be given for the discussion to happen, yet the discussion may remain open for a longer period of time if deemed necessary. The main goal of this post is to gather feedback and to gauge community interest on the proposed fee structure or on an iteration of it. The DAO vote will last 3 days as usual - just once the DAO proposal is accepted, it will enter into effect.

[JAK]

I think for now incentivizing inbound with zero fees makes sense, but may not be sustainable in the long run if the peg-in rate is disproportionately higher than peg-outs. The fee amount itself seems fair enough at 0.25%. As for the portal fees, there is very little to no data around the wrapping of NFTs. I wonder how you came up with a minimum of $5 and if it is feasible? Maybe a quick outline of the backend process and typical use case that makes sense at the current rate can help clarify?

[alicecorsini]

Alice from pNetwork here.

Agreed, the fee structure should follow the evolution of the project itself. Right now, we are in the early days and so in my opinion it makes sense to favour inbound transactions and encourage people to try out the system. In the long run, I agree such a structure may need revisions.
Should there be a need, the pNetwork DAO can propose and approve a new fee structure.

As for pNetwork Portals, you are correct that the data we have are limited as of now. Metadata support enables a variety of use-cases, including but not limited to cross-chain movement of NFTs. In general, pNetwork Portals power interoperability for smart contracts across blockchains (more details here: Announcing pNetwork Portals, a new way to enable cross-chain smart contracts | by pNetwork Team | pNetwork | Medium). Example use-cases could be:

• cross-chain interactions between DEXs on multiple blockchains

• interaction with pNetwork DAO from a non-Ethereum blockchain

• interaction with steroids.finance from a non-Ethereum blockchain

• cross-chain movement of NFTs

Again, the fee structure here could evolve in the future as per pNetwork DAO decision. The current proposal sees additional 5 bps on cross-chain transactions that make use of pNetwork Portals, to be added on top of the simple cross-chain asset transfer. The fee applied to simple asset transfers is competitive with other alternatives - it makes, therefore, sense to add additional pricing on top since an additional feature is being added as an expansion of the simple cross-chain asset transfer. The proposal includes variable fees dependent on volumes rather than flat fees with the rationale that a higher incentive for higher volumes makes it more attractive for prospective pNetwork nodes to operate. Such a model takes inspiration from the fee model applied by exchanges as well as decentralized liquidity pools (i.e. Uniswap). The minimum of $5.00 is mostly aimed to prevent abuses of the network as well as guaranteeing a minimum positive stream for nodes on low-volume transactions. The specific proposal of 5 sounded reasonable as a non-negligible and non-expensive amount, but happy to discuss alternative proposals.

I hope the above makes sense. Happy to discuss if it does not!

[JAK]

I guess I was just thinking on the end-user level, but I see now the structure is built for partnerships. It will be interesting to see some use cases this year and collect more data on cross-chain interactions. I’d still like to see what others think, but it seems to make sense for now.

[gmork]

Ren has a burning fee of 0,1%, They are discussing about moving it to 0,2%. Their minting is charged 0,25%. My opinion is that, right now, a pegout fee of 0,3% to 0,35% can work for pNetwork - considering that fees to swap from PBTC to RenBTC should be at least 0,25%.
0,25% pegout it’s quite “psycological” (REN is 0,1+0,25 while pNetwork is 0+0,25) so it’s OK.

For future proposals, I would consider differentiating fees among different BCs, the costs for the nodes are different and this should be taken into account.

[alicecorsini]

These are all good points!

We looked into fees applied by competing solutions and we tried to align the pricing. As you mention, Ren applies 0.25% mint fee and 0.1% burn fee. Others apply symmetrical fees on peg-in/peg-out (0.25% for WBTC, 0.3% for imBTC).

Because the visibility of the project benefits from a larger Total Value Locked in the system (that currently serves as a vanity metric), we thought an asymmetrical fee model that favours peg-ins would make sense for the current stage of the pNetwork.
You are right that the peg-out fee could be slightly increased to 0.3% or 0.35% to compensate for the 0% peg-in fee. That’s a possibility. My personal opinion on it is that moving the percentage of peg-outs higher than alternative solutions would be a disadvantage for us at the current stage (since our 0.3% or 0.35% would need to compete with alternatives that offer 0.25% and 0.1%). That said, I totally understand your thinking. Would be great to hear other people’s opinion on this!

Worth noting is the fact that we do have the additional metadata feature (pNetwork Portals) that none of the alternative solutions discussed above offers. This feature (that basically enables cross-chain transfer of data and can be used for example for moving NFTs, for cross-chain connecting liquidity on multiple DEXs, etc). We would be applying additional 5 bps in each direction for the usage of that feature.

Happy to hear additional thoughts on these points

[VVV]

Hi everyone,

First I’d like to say that I’m glad to see that you have gone for very competitive fees when when combined. Low fees should be great for token velocity and liquidity. Also being cheaper than competition should be even more reason for users to use pNetwork over the other more cumbersome and expensive options. I hope that this therefore means more fees in total to the nodes than they would get with higher fees.

Particularly with high ETH fees, and the validators being compensated with the 42% staking rewards anyway, low fees are a must at the moment.

WRT the asymmetric 0/0.25% fee - I get where you’re coming from - encourage peg ins and discourage peg outs to maintain a higher TVL but I have a few comments:

1. I think that if I want to peg in, or as a lot of your users might just be doing this through the Eidoo wallet - the difference between 0 and 0.1% is really insignificant. People are used to paying this on exchanges, people even pay 1-3% fees on Coinbase or wherever. Arbitrage traders will also probably not care, and for them the OVERALL fee is what matters, not one way, as they’ll likely trade both ways.

2. The 0.25% fee is the opposite - it’s higher than exchanges and just feels like quite a hit. I’ve had this feeling trying to cash in/out of wBTC who have a 0.3% fee I think - 1BTC and thats a $140 fee - really makes you think twice about what you’re doing, and I ended up not using them, went to an exchange instead to pay a 0.1% conversion fee and a small spread.

Since I am proposing to raise the peg in fee, I therefore propose to lower the peg out fee to 0.15% and the balance will remain the same. The dynamic is still as you intended but again, I feel like these fees will actually increase the amount of trade happening.

3. I know this is probably a bit more complex to implement but my final suggestion is that instead of the 0.15% peg-out fee going to the nodes, 0.1% goes to the nodes and 0.05% is converted into PNT and burnt. I don’t know how to implement it, but could be that the fees are pooled and traded into PNT on uniswap at the end of the week.

Reasoning - drives the value of PNT which is in everyones benefit. Particularly users who aren’t validators, would have an incentive to tell people to peg in/out. Validators are big PNT holders so it benefits them just as much. Sort of a global fee for using the pNetwork.

I’d say ideally these are things to test, but I guess it would take too long to test so the best we can do is speculate on user behaviour/psychology.

Comments welcome!

[marcodalponte]

Hi VVV, I understand the reasoning on points 1 and 2, plus there is a risk that peg-out commissions act as a psychological barrier and limit peg-ins despite being commission-free.
But IMHO we should test the 0% peg-in and 0.25% peg-out structure to see how users react and if it helps the TVL increase, if this model doesn’t work we might propose to change it.

On the third point, I completely agree with you, burn PNT with a portion of fees collected would be great for all the holders, not only for node operators.
Complexity could be shifted towards node operators by burning directly their staking amount (definitely uncomfortable but effective).

[gmork]

Hi,
in my opinion, the value of PNT is more strictly linked to the profitability of nodes rather than a deflation mechanism. As I don’t see other reasons for a deflation machanism than the research of an increase in value of PNT, I believe that the first and more relevant focus is to make the node profitable. Once they are, the DAO can think about destinating additional fees to deflation. We need to be sure the nodes start operating decentralizing the PNetwork as first goal!

[alicecorsini]

Thank you all for your thoughts, that’s very useful!

Just wanted to clarify that burning is already used as a mechanism within the PNT tokenomics. Specifically, the PNT token is automatically burnt in relation to services offered via the Eidoo app. For example, 20 PNT are burnt every time an EidooCard BASIC is ordered. A similar approach is followed for the EidooSwap.
This is a continuous burning mechanism that happens on top of manual token burnings. Coinciding with the launch of the new tokenomics (summer 2020), for example, over 28 Million tokens were burnt (reference: https://en.cryptonomist.ch/blog/eidoo/the-edo-token-is-being-upgraded/).

[VVV]

Agree with the potential psychological aspect of it where a higher fee on the way out might make you feel trapped. Equal on both sides feels more like paying for a valuable service.

Considering that we have 0% peg in/out fees right now, I don’t think that the proposed structure will be a great test. I agree with testing. Would be great to test the difference between multiple fee structures but I doubt we’d get the proper data quickly and easily enough.

[VVV]

Sure, but I think that the fees are a small part of the overall profitability until the DAO payments stop so no need to worry about it.

It’ll cost what $30 a month to run a node? Getting paid 42% (which you can get without running a node) but then also the pegging in/out fees on top… Don’t think 0.05% will make a huge difference in their willingness to operate a node.

[gmork]

Running a pNetwork node costs 200$ a month considering the required security requirements (it runs on AWS nitro enclave). The DAO rewards are not an incentive to run a node, they are recognized to any staker. I just don’t see any relevant effect addressing the 0,05% to deflation. At this stage I would rather discount this percentage from the fees trying to make them more attractive or increase the rewards for node operator. Once the node business become really attractive and many of them are active, I would reconsider my position. A lot of node operators means a lot of deflation as bags of tokens need to be staked. And node profitability is at the core of the Project’s sustainability. So, until the business is strong, I wouldn’t address elsewhere any part of the fees. My PoV.

[alicecorsini]

The reasoning behind the asymmetric fee structure is focused on the overall benefit each operation brings to the pNetwork ecosystem. Peg-in operations create an inherent higher benefit to the pNetwork as they contribute to increasing the TVL (total value locked). On the contrary, peg-out operations create an economic incentive for nodes (when fees are applied), but they actually contribute negatively to the TVL.
For the reason explained above, the asymmetric fee structure can balance the two operations by introducing an incentive for nodes while facilitating users/liquidity onboarding. This model sort of follows the same logic applied by exchanges, where “maker” types (those who create liquidity) receive a more favourable treatment compared to “taker” types (those who remove liquidity).

[alicecorsini]

I personally agree that node profitability is a key element for the project’s sustainability and that we should focus on that. Benefits for PNT holders (non-nodes) are already being implemented in various forms, including DAO voting rewards, rewards for liquidity providers in pNetwork-related pools (Uniswap, Curve among others), burning mechanism for Eidoo-related services. Plus, the more nodes we have the more tokens are at stake aka lower circulating supply.

[alicecorsini]

Note for everyone: We are aiming to move up this discussion at a DAO level with the start of the next epoch. Thank you all for your thoughts and contributions! If you have other comments, please keep sharing