HIP-1: Setting 5% minimum fee on network level

Governance Validator DAO
41

I generally agree with you. When I saw what was going on in the delegator/validator scene, setting a minimum fee came to my mind as well.

I would argue that delegators take far bigger risks than validators, it’s not so easy money at all. There needs to be harmony between the two parties. What’s a fair fee going to be?

I think one solution would be lowering the barrier of entry to becoming an elected validator. Then the real “lazy” can delegate and the rest could just run their own node. The fair base fee of any % would be closer to the equilibrium. People would have a real choice. Can we?

42

“Problem” with this is we cannot lower the barrier… that’s what EPoS is about… The “barrier” to enter as validator is set by the EMS nothing to do about that…!

1 Like
43

Edit: I see he has joined and had his say :slight_smile:

44

How come you think delegators take big risks? Their one and only risk is delegating tokens with a validator they would trust, not even to optimise gains, but to be sure the validator wont act maliciously. Other than that, thats it, validators dont really have any control or access on delegator’s tokens, only the delegator that own those tokens have.

At the same time validators need to harden their VPS to make sure people dont hack them, they have to make sure theyre online and signing blocks (which is not as trivial as it sounds), and they have to play the EPoS game each day.

All in all, im a delegator in many other networks, and while delegating is sometimes the easiest way (or the safest if youre not tech savvy), there is a big gap in risk for validators and delegators, not just with Harmony but with every other chains too.

On the matter of price rising and fees → if you check everyone else, no one is lowering their fees, in any network. You would ask why, since the price is higher so more gains in USD? Theres a few reasons, one is taht most of validators went through the bear market with the same fee settings, and a lot of them were not even having positive balances due to the low price back then. It only makes sense they can finally profit a bit from all their hard work. I 100% support that in Harmony and every other chain. You would probably feel the same if you ran your own validator, its easy to say things if you dont know how it looks when you have to actually pay attention and work each day. The second reason is, a lot of these validators dont really care about fiat value of the token, they just want more tokens. Granted this is a much smaller pool size but even these exist, and for them, the amount of tokens they earn now is the same, as back then.

Regarding lowering the barrier - yes we are doing that but in stages and are not rushing, to make sure the network is safe and secured. Once most of the seats are given out, its only on us external validators to make the network work, which can be risky at this point as any real issues would require all validators to restart the chain at the same time - not an easy feat.

That said, one of the proposals to lower the barrier is that EPoS upper bound suggestion (the other proposal here in the forum), which is about the only thing that can be done right now, due to the nature of EPoS. Those bounds are actually a security feature, so they cant be streched much unfortunately.

1 Like
45

In addition to HW costs and time/effort, there’s also a substantial regulatory/legal burden placed on valdiators in form of tax reporting, insurance, business organization/formation costs, website/email costs, potential staffing for 24/7 monitoring, etc. I don’t think Harmony wants a network full of 0% validators running on bare minimum.

If validators are earning 5% fee, but aren’t proving their worth, delegators can stake with someone else. With 5% fee paid to validators, you’re likely to see those fees reinvested back into securing the network and helping the community rather than sold on an exchange.

46

Any ETA on when this proposal could go live on governance.harmony.one? Curious on next steps for planning purposes.

47

I don’t really see the benefit of a forced 5% minimum fee. I think there is an argument to be made that it actually contributes to stake centralization:

Small Validators/Newcomers:
5% might help to cover the cost at current prices, but it robs them of a powerful marketing tool. The smaller stake is already making it hard to keep the APY competitive (EPOS optimization), now they wouldn’t even be able to offset it with a lower fee.

Big Validators
They have a base income that could get quite substantial at higher price points and cannot be undercut anymore. With better EPOS optimization & therefore APY, there’s not much reason for delegators not to stake with them → stake centralization.

Self Validator
As a former validator I also don’t see it incentivizing me to spin up my node again. I happily pay the 5% of a larger one, as the fee is usually lower than the rewards I was losing because of inflexibility due to a smaller stake.

Since I’m paying it anyway the forced fee wouldn’t affect me, but I don’t think it solves the problems that were highlighted. In a way, it might reinforce them.

48

Implementing this scheme won’t be too complicated but the problem is how to decide on these brackets fairly. Once implemented and launched, it shouldn’t be changed too quickly again and it should be working for all the situations in the foreseeable future.

49

For small features, implementation complexity is generally not the main problem (it usually won’t take more than 1 quarter to finish). The more important question is whether the proposed feature is well thought of and really solves problems. It’s better to write things up in a formal proposal with detailed reasoning and analysis first if we want to consider any new features.

50

Yes i feel like setting the levels right needs a lot of thought put into it, probably some calculations. Otherwise i like it and should be started separately.

One thing i fear though is that it will be confusing to delegators, since they will have no idea why a certain validator they delegate to went from 3% to 4% (or whatever values we decide) after they delegated. And those would change quite often, as delegators will be redelegating from one to another. I feel like a smaller number of brackets is needed, to really catch the few key sizes ( 0 - 5m, 5-20m, 20-50m, 50-200m, 200m+), and maybe even that is too much brackets.

On another note, we should still prevent the race to the bottom scenarios so the fees for the levels have to be really thought of correctly.

1 Like
51

It’s an interesting suggestion but agree with RJ that the brackets are hard to set. We may just need two brackets. One is the newly elected validators with some ramp-up time like a few weeks to attract new delegators. Another bracket is the floor fee for everyone.

52

yeah i like this suggestion… it helps new validators bootstrap while the “old” ones are restricted to some minimum.

It can be lower than 5% though if majority wants less, but i def wouldnt go below 3%

Awesome suggestion @leo !!!

53

Ok everyone, i think this is the best solution so far. So newly elected validators get a temporary boost and can have 0% fee if they want, but its limited to X weeks. I would suggest 5 -10 weeks. This is still to be discussed so i wont open a poll for this one.

Below is the poll to vote what would be the minimum fee for existing validators, please vote for what you feel is the best:

What should be the minimum network fee for existing validators?
  • 1%
  • 2%
  • 3%
  • 4%
  • 5%

0 voters

2 Likes
54

Hey everyone! Delegator here. Aaaaand, here’s my two cents.

I think it’s important for our validators to feel incentivized. I want our validators to be rewarded for the good work they do.

I saw mentioned above that a 0% fee could be offered to new validators to help get elected. I’m not sure how difficult that would be too implement, but could be a good idea.

Most of the validators I delegate with have a 5% fee anyways, so I can’t say I’m against it.

If the minimum 5% fee is moved forward I hope there is a strong community push and campaign to educate the delegators about the benefits of there being a minimum 5% fee, as to not discourage anybody from not staking their ONE. Doubtful that would happen, but just a thought. The community is already pretty strong, so I don’t think a little education on the benefits would be too difficult.

Edit: doubtful that people would stop staking their ONE, not doubtful there would be a strong community push to educate. Lol. Just clarification.

Another edit (sorry lol): An example of the community push to educate delegators literally just happened on Twitter. A delegator was reluctant and hesitant to give up the 0% fees (he has a voice, but not a choice lol) because he believed the minimum 5% fee would SUBTRACT “5” from the 10-12% APY. Leading him to believe that, as a delegator, he would only receive 6-7% APY. He was quickly brought up to speed that his math wasn’t correct and immediately shifted his response being very positive and receptive of the 5% minimus.

2 Likes
55

Hehe yeah, i had to reply to it, although its clear that the commission is always taken from the rewards. The % might have skewed it here.

Eg: if you earn 1000 ONE and validator commission is 5%, you will be left with 950 ONE. If you imagine these 1000 ONE is 12% APY, then it becomes more clear :slight_smile: But yes good point on educating delegators about how this works, although in this particular case, it was just a bit of misunderstanding!

1 Like
56

Absolutely. I’ll be there to help spread the word! :slight_smile:

1 Like
57

This is the misunderstanding I see re: validator fees while moderating the Reddit community, and I can understand their confusion. The concern/misunderstanding from delegates is that a 5% will reduce a potential 11% APY down to 6%. I think some light community education will be needed if this vote passes successfully to help soften the blow and to simply help delegates understand the rewards/fee system. We, myself included, can coordinate an announcement on social media platforms not only for the protocol change, but to also include some examples of how it truly impacts a delegate’s returns.

Here’s an example post I put together a couple weeks ago to help educate the community on validator fees after seeing some concerns posted by Redditors:

This ^ but boiled down to three sentences attached to an announcement would go a long way.

EDIT: concern > misunderstanding

2 Likes
58

Adding to this, I would much prefer a solution similar to what Ogre proposed where newly elected validators have the option to set a 0% fee for a short period of time (if possible). I think we’re not giving enough credit to how 0% fees attract delegates for smaller validators. Sure delegates may leave the validator once fees are increased, but a number of them won’t, and it’s likely these delegates who stay would have never staked with the validator who while unelected if their fee was anything above 0%.

It’s hard enough convincing delegates to stake with unelected validators with 0% fees. Try getting the common delegate to stake with an unelected validator set to 5% fees. It won’t be easy.

2 Likes
59

Agree with everything you’ve said RockTheBlockchain. I don’t think a little light community education will be hard to pull off, but it will have a big impact.

To justify your point that 0% fees will attract delegators, but then will not leave once fees are raised, I first started delegating with RockTheBlockchain because of the 0% fees, but after watching my rewards accumulate with RockTheBlockchain validator I will never leave lol.

60

hey, as you can see in the post with the poll above, i mentioned that one is already decided upon. So newly elected validators will be able to set 0% fee, limited to a short period (which we should still discuss). So yeah this proposal is already getting upgraded and finalised, the question is now only which minimum fee for existing validators (looks like 5%) and what the period for newly elected ones would be. After that, we will put it on Governance app for voting!

Re: the period for newly elected, i feel like 10 weeks is ok, since its not that easy to attract new delegators so i feel something like 5 weeks seems a bit too little. Since we use epochs we can also suggest this to be in epochs, which will probably be the way the network will check it anyways. In that case i think 50 epochs (~50 days) would be good enough?

1 Like