Adding utility to mkr token (discount stability fees for vault owners who hold MKR)

Makerdao is the most useful lending platform on Ethereum with the benefit that it can generate billions of liquidity in DAI and this puts maker in a position of being a central bank in defi that can provide people the best rates. However, I don’t think that MKR has enough utility. I know rune suggested ideas like locking your MKR for years to get free loans but those are not things that people in Defi tend to do, specially with MKR performance during this bull-run being quite boring. I think we can say you get x% discount based on your MKR balance during the time you are borrowing money and when you want to repay the loan the weighted average of the discounts is reduced from the interest.

something like this can be good:

1 MKR → 5% discount
5 MKR → 10%
10 MKR → 15%

5000 MKR → 50%

You get the idea. In the future when asset managers like yearn finance want to borrow DAI to put it in use in big volumes they will be incentivized to own MKR.
I think this idea solely can have enough benefits for MKR holders that maker burning wouldn’t be much needed (although right now it doesn’t seem to be burning either, less than 1% of the supply has been burned since the beginning). This way, vault owners will be ingrained in the community too.

Another idea that I have is allowing big MKR holders to run their own debt pools. They can put up MKR as collateral so if they don’t manage the pool well their MKR is auctioned. The pool managers can allow people to mint DAI with riskier assets like LP positions or new defi coins.
This is kind of like Rari-Capital and can put maker in a position that abracadabra.money can be implemented as a single debt pool on maker.

If enough pool managers exist the competition and risk tolerance can make makerdao move faster, for example to list an asset like OHM maybe all the community doesn’t agree with the safety but an individual decides the potential benefits outweigh risks and lists this asset in their own pool. Of course most of the fees generated by these pools can be kept for their managers and some of it given to maker itself so there are huge incentives to run these pools if you already have large amounts of MKR. There are many things that need to be tuned for this idea though, like deciding who can run pools well by the community etc… But I think it’s doable and can reduce the pressure of reaching consensus among all community for each single change.

In my opinion a Maker 2.0 is needed to allow us to benefit from all new defi developments and increase maker’s dominance in the lending market so we can all brainstorm about new features that can make makerdao more interesting to use.

15 Likes

Hey! This is a very simple yet seemingly clever idea!

I don’t think it has been discussed earlier and I’m looking forward reading some feedbacks from the community and Risk teams.

I have just read this comment in reddit (here)

And indeed it’s true that there are a lot of Vault users who don’t even see the point of MKR.

Offering them a discount on the SF would be a strong incentive to hold MKR and, as far as I can see, it also does not have direct tax implications.

4 Likes

One alternative:

the SF can be seen as providing at each block:
A) money to run the protocol, and
B) dividends to MKR holders (currently delivered by burning mechanism).

This means that each MKR holder, at each block, should get a dividend proportional to (B) above, depending on how many MKR they hold vs total MKR in circulation.

Then I propose: the discount on the SF of a vault should just be equal to this dividend (in that specific block).

Motivation: It just makes sense to directly discount the SF of a Vault of a MKR-holder from the amount of Money that they are rightfully expected to get as dividend from the SF on their Vault.

Example: Suppose I have 1k MKR, and that there are 1m MKR in circulation. Suppose I have a vault and the SF accrued in the next block is 1 DAI. Then I get a discount of 1k/1m = 0.001 DAI on the SF of that block. So I just get to pay 0.999 DAI, instead of 1DAI.

1 Like

This model can be good too! it’s hard for me to compare them tho. To me the simple discount model is easier for people to understand and calculate exactly how much benefit they get by holding MKR. Dydx uses this model too.

1 Like

Yes I agree. What I proposed above should imho be kind of uncontroversial: It just makes sense to discount the SF of MRK holders from by the rewards they are expected to get.

We can be more aggressive/simple/attractive etc, with other schemes (such as the one you originally proposed) but that of course requires discussion and careful analysis.

1 Like

Thanks a lot for your feedback! I’m curious what you think about my second suggestion: allowing large MKR holders to run their own debt pools.

I am not sure I have 100% understood it yet, so I’ll wait. But it seems on a different level of complexity (and thus, time required to approve it and/or implement it) than the first.

2 Likes

This strikes me as a very interesting idea. I would love for members of the @Risk-Core-Unit to comment on possible unforeseen issues with this kind of program.

For example, how might this effect parameter and SF proposals? How might Vault users behave with MKR? Will there be a lot more MKR velocity/volume as a result of this? How does this affect MKR’s risk profile, etc.

Fantastic proposal idea, thank you for sharing it!

5 Likes

I really like these ideas. Is it possible to integrate this with the 3x vote power for lockup concept proposed by Rune in the SE proposal? As stated, this could incentive major protocol users to buy and lock MKR, as well as subsidize and foster their participation in governance. It could even push passive MKR holders to mint some dai to get their full MKRs worth of benefits.

I feel like a lot of concern was shared about the proposed mechanisms in the Sagittarius Engine proposal, and this could be a more sustainable alternative for providing concrete rewards to MKR holders.

Should MKR holders need to have and lock other collateral, and take out a loan with interest, in order to benefit from their MKR? What if they can’t come up with collateral, or don’t want/need a loan?

1 Like

@someone
A fair concern. I think the originator of this proposal was lockup skeptical if I recall correctly.

Still, an argument can be made that a large share of tokens being in required or voluntarily user sustained long term lockup can have similar effects to a buyback and burn by removing mkr from circulation.

I think adding real utility to mkr regardless of requiring lockups will also reward passive holders by driving up demand for the token

Am I mistaken, or is most of this discussion around turning Maker into a credit union or (maybe more accurately) a coop?

Just trying to understand some of what is being thrown around here. Interesting discussion, if nothing else

2 Likes

Yeah, a credit union would be good way to describe it! It’s more a way to optimize the utility for the current holders and incentivize vault owners to hold MKR. I guess if you have millions of dollars in DAI loans it’s very logical to own $MKR if you are offered discounts on your interests, right now that ETH is in up only mode you can just never repay the loans! specially for people who borrowed very early on this makes sense.
I guess if you offer discounts this way this will result in people borrowing dai and converting some of it to MKR as it will help them to reduce their costs a lot!

Vault holder here and wanted to say this would be really awesome for both sides.

3 Likes

good idea , good for both MKR holder and MKR user

2 Likes

The proposal reminds of utility tokens that never really took off, because people usually don’t want to be paying for services in volatile tokens. Here the MKR represents a discount though, which one could argue is not the same thing. I think Nexo uses this model and I am unsure of its actual utilization. To me it feels projects have been implementing this primarily for regulatory reasons - again, utility tokens hypothesis…

I do support implementation of mechanisms where borrowers get certain benefits, but in return they should provide something that lowers risk of their exposure. For instance, I’d certainly like more to be rewarding (or offering discounts) for high collateralization ratio vaults, or vaults subscribed to protection services.

This though wouldn’t solve the “lack of MKR utility” addressed in this post. I do think if people locked MKR/DAI LPs or MKR/ETH LPs to get SF benefits, this should have positive effects in a situation where we’d have to potentially cover loss with MKR dilution, because MKR liquidity would be more robust in tail scenarios if part of MKR LP liquidity gets locked. It is basically the same model that Aave employs, but in this proposed case you would be rewarding vault users and not only MKR token holders. So it is not a bad idea from a risk point of view (adjusted for locking in MKR LP liquidity instead of only MKR), just not sure of its actual utilization in practice. It can also get very complex when trying to price this, because it is hard to measure risk benefits.

6 Likes

No vault owner seems to like holding MKR. Why? Because it has zero benefits for them. By doing this you will incentivize them to hold MKR because even if you have a 1M$ Dai loan which you have to pay 20K interest for it, you can just buy 1 MKR and pay 19K interest per year which makes buying that MKR very attractive and basically free if you want to hold that position for more than 2 years. I’m sure there are larger vaults who would be willing to buy much more MKR.
This will make vault owners own mkr and become part of the DAO too which is very beneficial long term.
By the way $dydx has really took off and the only functionality of their token is this. You can’t simply throw in a bad example and say the model is bad, at least consider the benefits and then provide valid counter-arguments.

3 Likes

I am not sure I agree. MKR remains first of all the “governance” token of MakerDAO, allowing you to profit from the protocol’s income and, also, to carry some of its risk.

If we are adding some utility alongside the above, this doesn’t turn MKR into a utility token used to pay for services. Also because, as you say:

MKR would not be needed to open vaults, nor would have to be “used” to pay for anything. It would just act like a “VIP card” that allows for better rates.

1 Like

I’m open to being convinced otherwise, but this sounds like added complexity that would either 1) turn off users who felt excluded without MKR or frustrated by extra steps, or 2) be gamed/cannibalize earnings

2 Likes

how? Honest question.

From what I see, this is a discount on the SF. Worst (and predictable with accuracy) thing that can happen is that MakerDAO’s income goes down and the risk profile is changes (as @Primoz said, but again this can be calculated with accuracy as it is predictable).

Also

giving the MKR holder their “cut” immediately (as a discount), rather than paying the SF first and then eventually be rewarded by the buy-and-burn mechanism, seems very direct to me and the opposite of:

1 Like