Proposal to change USDi rewards

General Discussions
#1

I think the protocol should reassess the existing liquidity incentives setup. As it currently stands, we have borrow rewards to reduce the cost of borrowing and to distribute IPT to the protocol’s users. Additionally, we have the ETH-USDi pool to supplement the need for USDi redemptions. The initial thought behind the borrower rewards was that by increasing the borrowing rate, the supply rate would also increase to sufficiently attract lenders. Since the new interest rate curve was introduced in proposal 12, the protocol’s deposit rate has been trending down and currently sits at 5.25% with a 7-day average of 5.10%.

I think we need to find ways to increase the available liquidity in Interest Protocol, and one of the best ways to do so would be to incentivize converting USDC to USDi and holding that USDi . If the protocol were to end its ETH-USDi incentivizes, it could repurpose the 42,734 IPT/week toward USDi holders. If we use the current price of $0.14, then that would be about $300k annualized. If we target a 10% yield rate with 4% already covered, then it should incentivize about $5m of additional USDC in the system.

However, some borrowers hold their borrowed USDi, so they could double dip on the incentives. While the immediate thought could be not to include those addresses from the count, that isn’t a feasible solution. Instead, I propose we lower the borrower rewards by 1/3 from 85,469 IPT to 56,979 IPT and increase the supply rewards from 42,734 IPT to 71,223 IPT. The 25% higher supply rewards versus borrow rewards should shift the protocol’s liquidity profile to be more beneficial for both lenders and borrowers.

I would love to hear what other IPT holders think of this proposed revamp of IP rewards.

1 Like
Proposal to Pause IPT Liquidity Mining
Open up discussion regarding changing IPT rewards
#2

Stated goal:

What does this mean exactly?

Like describe to me a vault, USDi, USDC IP state that signals success?

  • Does IP want more USDC as backing liquidity (and hence more USDi)?
  • Does it want more USDi borrowing?

I personally don’t think this is going to do much of anything and begs some questions.

The questions:

  • How is USDi in liquidity contracts going to be compensated in this scheme?
  • Why is it better to incentivize USDi holders than USDi borrowers. Frankly the borrowers are taking the risks and paying the interest. USDi holders are getting ‘some’ return. Why skew the rewards towards USDi holders than borrowers?

Lets describe more fully the goals, and methods here before we just start moving significant incentives around. If we can’t reward USDi in LP like we do USDi in wallets I think this is a huge problem since it will mean USDi will sit in wallets vs. making 1 or more markets with LP.

1 Like
#3

“Available liquidity” is in reference to USDi/USDC that can be borrowed from the protocol. I think borrowers have an appetite that the protocol is not fully serving because it has a generally small amount of available borrowable capital.

I think it would be good if the protocol could maintain a reserve ratio of 30%-40% while scaling the USDi in circulation and the USDi borrows. Currently, the protocol is experiencing a classic chicken-and-egg problem where in borrowers can’t borrow a worthwhile number of funds, and depositors cannot deposit meaningful funds without significantly lowering the deposit rate.

Depending on where USDi is pooled, the script could be modified to calculate and assign rewards USDi holders in 3rd party contracts. We would need to set a threshold at which this would occur, but it should be fairly possible.

I wouldn’t characterize the proposed incentives as favoring lenders over borrowers; however, to date, the protocol has not had direct incentives for lenders while they play an important role in the protocol. Generally speaking, lenders are taking on more protocol risk than borrowers because they are lending directly to the protocol, and borrowers are isolated from other users.

#4

I much prefer diverting incentive to USDi holders. As you alluded to before, demand for leverage is very low during bear markets. All we can do now is add support for new tokens while we wait for the next bull run. So, there’s no point in incentivizing borrowing for now. If someone wants to take out a loan, it’s probably because we are the only venue to do so, not because we have the cheapest rates (but because the protocol is so small, no guarantee rates will still be low a few days from now). So I would only re-incentivize borrowing once we start to transition out of this lull to accelerate borrowing interests.

Theoretically, if the USDC reserve is high, borrow rate will be low already, so there will be no need for additional incentive. Why don’t we direct all current IPT reward to USDI and USDI-ETH pool holders? Could do 50%/50% or 75%/25%. This will increase USDC liquidity for the protocol and decrease borrowing rates.

And is there a way to vest the IPT rewards? Could do a year or longer. That would create more long-term alignment. I personally have USDC that I would convert to USDi if there’s liquidity incentive.

1 Like
#5

In short, we should strive for maximum USDC liquidity within the protocol to create low interest environments for when organic demand for leverage returns.

And in the long run, we need to find actual utility for holding USDi (maybe Poppie card?) for when supply incentives are turned off.

1 Like
#6

Another point is as a USDC holder, I rather get paid in IPT rewards (something that has the chance to grow in value should the protocol grow market share) as opposed to USDi that’s static. Should the smart contracts be secure and I don’t need the USDC right away, this can be a long term strategy.

On the other spectrum, if I have a CDP, the IPT reward is nice and offsets interest rate, but I am ultimately more concerned about my liquidation level and what I will do with the additional liquidity that will generate a higher beta. So the IPT is not a huge point of consideration. If the USDC reserve is not stable, rates will vary greatly, which is another point of consternation for borrowers. I will feel much more confident in taking out a loan if I see the protocol has million and millions of USDC in reserve.

So bottom line is I think we should change direction and try to incentivize USDi holders more, as we’ve already seen a relatively juice (5%+) rate isn’t interesting enough to onboard more USDC liquidity.

1 Like
#7

Thank you, @spicetrader, for your feedback. Generally, I agree with your thoughts. I tried to address two of your questions below.

I’m apprehensive about entirely stopping borrowing rewards because it will change behavior more than ending incentives for ETH-USDi LPs. I’d rather make the proposed change and revisit the topic after a month.

Anything is possible, but in my experience, vesting rewards isn’t super useful because users will price it into their behavior.

#8

Sure, I agree the more prudent way to slowly make adjustments rather than something so drastic :+1:

1 Like
#9

Hi @Getty,

With the current liquidity for USDi located on Uniswap v2, I believe it makes more sense to have USDC liquidity on either Uni v3, Curve or Balancer v2. I personally think there is a great opportunity for having liquidity on Balancer v2.

I view IPT incentives as one two or three types of incentives Interest Protocol can offer early adopter of USDi. I suggest Interest Protocol creates a Factory with Linear Pool to create bb-ip-USDC. This Linear Pool can then be paired with wrapped USDi. The Balancer v2 pool becomes:

BPT = bb-ip-USDC / wUSDi

The USDC linear pool would deposit 70%-80% of the USDC into IP and Liquidity Providers on Balancer. This pool would qualify as a Core Pool and Balancer would receive 50% of the yield generated by USDC deposits. Of the 50% that goes to Balancer, Balancer keeps 25% as revenue and uses 75% for bribes on Hidden Hand.

If IP offers IPT rewards on USDC deposits, these will accrue to the address that deposits USDC into IP which is the linear pool address. Sometimes, Balancer claims these rewards as revenue, but it is likely that the rewards could be used for bribes on Hidden Hand or they can claimed and distributed through the Balancer Gauge to Liquidity Providers. who stake the BPT.

In addition, Aura Finance can create a gauge for the BPT and users who stake the BPT with Aura Finance, would receive BAL, IPT and AURA rewards. This has the potential to drive more USDC into IP. It also creates an incentive for users to borrow USDi and deposit it into the wUSDi pool. The key takeaway is the Balancer and Aura Finance ecosystem will provide additional yield to liquidity providers which also contributes TVL to IP at the same time.

Regarding the split of IPT across USDC and USDi, I tend to favour more rewards to USDi over USDC. However, I think with the potential of bb-ip-USDC / wUSDi, I am inclined to favour the proposed split and having an agreement with Balancer how the IPT rewards from the Linear Pool are distributed.

There is always the upside the BPT can be listed as collateral on other markets and if USDi was also listed, it would enable a nice recursive loop that leads to greater liquidity for USDi and USDC deposits into IP.

We can also consider migrating wstETH / USDi liquidity to Balancer which would also meet the Core Pool requirements. Core Pools typically get decent support from veBAL holders which is always an added benefit.

The above extends upon what was initially published in the IPT rewards distribution and technically, it could be considered separately / complimentary. Overall, I think this is a sounds direction to head in.

1 Like
#10

That is an interesting idea. Generally, I’ve been apprehensive about supporting USDC/USDi pools because USDC should be focused into the protocol. However, the concern is satisfied if the pool can deposit most of its USDC into the protocol. Though, in this instance, it isn’t clear to me who would utilize a USDC/wUSDi pool. If the pool doesn’t generate fees for Balancer, then I don’t know how likely the protocol would accrue notable rewards. Alternatively, the protocol could do something similar with the USDi Curve pool.

If the protocol could support Curve or Balancer tokens as collateral, that would add another interesting dimension to this conversation.

1 Like
#11

Pending additional feedback, I’d like to propose the end of the ETH-USDi Uniswap v2 rewards and a cut to borrow rewards of 1/3 to be redesignated to USDi holders on this upcoming Wednesday.

Additionally, I want to address the question of IPT rewards going to USDi holders in secondary contracts, such as the ETH-USDi pool or the USDi Curve but not limited to them. Any secondary contract (if it is made public) with at least 100k USDi can qualify to have the IPT rewards flow through. A participant must make a forum post requesting it be added with an explanation/example of how the accounting should be handled.

2 Likes
#12

Let’s see if I have this.

IPT rewards right now are:

1/3 to ETH-USDi v2 LP
2/3 to USDi borrowers

You want to change this to
1/3 to USDi borrowers
2/3 to USDI holders (only contract that may have more than 100K is the ETH-USDi)

That sound right?

So what this will do is basically pay people to drop USDC into the protocol to mint USDi to hold.

I would offer a counter proposal.

1/3 to borrowers
1/3 to USDi holders (including USDi held in liquidity pools)
1/3 to USDi that IS in liquidity pools.

While it may make sense to toss 2/3 of the rewards at USDi sitting anywhere I think it is unfair to characterize USDi paired with other liquidity in a pool as equivalent to USDi sitting pretty much dead to markets in wallets.

Points to above are that.

  1. You get access to 1/3 IPT to borrow some USDI.
  2. If you hold your USDI you get access to another 1/3
  3. If you hold AND pair with other LP in a protocol you get access to the remaining 1/3 IPT rewards.

I think all three are equally important and should make enough of an incentive shift to bring in some USDC liquidity to reduce borrowing rates.

Final NOTE regarding (3) above. There has to be a distinct difference between LP with significant IL vs. one without. USDC/USDi (say in a balancer pool) has pretty much zero IL, where as USDi/ETH does. Initially I think we don’t have to distinguish because whole point here is represent with (3) the idea that USDi is paired with some other value and the hope here is that it is at least 50:50 USDi:OtherStuff to justify the additional 1/3 IPT reward rate based strictly on the USDi in the LP avg’d over time.

One additional point. I think some rewards for IPT provided as liquidity would be useful as well. Even if 10% of the IPT rewards (about $1K/week at current price) were driven to v2 and/or v3 LPs this gives $100K of LP (which is about what we have) something like IPT APY of 52% at current prices, $1M worth (which is about 10% of the float) would earn about 5% in IPT rewards.

#13

Hi @Getty,

Within the bb-ip-USDC / USDi pool, Balancer earns 50% of the USDC yield, 50% of the swap fees as Revenue and 50% of the rebasing in USDi.

Of the 50% that goes to Balancer, 25% as retained and 75% is used for bribes on Hidden Hand.

An example:

For just bb-ip-USDC generates $8 of yield, $4 goes to LPs, $3 to hidden hand in bribes and $1 is retained by Balancer.

The same logic applies to $8 generated by swaps fees.

For wrapped USDi, assume USDi rebases and quantity increases from 100 USDi to 108 USDi, then similar to above, 1 USDi is retained by Balancer, 3 USDi is used for bribes on hidden hand and 104 goes to LPs.

Balancers revenue model is different to Curve Finance. Balancer enables 50% of yield to be directed away from LPs to Balancer, whereas Curve offers either 100% or 0 % to LPs.

Hopefully this clarifies how Balancer benefits from the pool and why it would be well supported by the broader Balancer and Aura community.