KIP-022: Incentivize the Youves Flat Curve AMM

Governance Kolibri Improvement Proposals
#1

Author:
Hover Labs [email protected]

Created:
1/30/2022

Status:
DRAFT

Background

Existing Farms

The Kolibri protocol currently maintains 3 farms which emit kDAO. These farms incentivize activities that are advantageous to the protocol. Specifically:

1. kUSD Farm: Incentivizes holding kUSD. This is important because it provides a reason for users to hold kUSD to balance out sell side demand for kUSD before use cases develop.
2. QLkUSD Farm: Incentives participating in the Kolibri Liquidity Pool. The Liquidity Pool provides stability to the protocol by providing liquidity to liquidate ovens (Read More)
3. Quipuswap kUSD/XTZ LP Token Farm: Incentivizes providing liquidity for the kUSD/XTZ pair on Quipuswap. This allows price discovery and stability of kUSD. Quipuswap is the oldest and most liquid DEX in the Tezos ecosystem.

These farms all run for 1,048,320 blocks (1 year @ 30 second blocks). The kUSD farm and the QLkUSD farm will emit 75,000 kDAO, and the Quipuswap farm will emit 150,000 kDAO. This means that the Quipuswap pool emits 2x the kDAO per block as the others. This increased emission rate is to compensate for the fact that participants in an XTZ/kUSD Liquidity Pool on a DEX may experience impermanent loss.

DEXes and AMM Algorithms

The Quipuswap DEX is an automatic market maker (AMM) that uses a constant product formula to calculate a price by looking at the ratio of assets in the pool. This means that each trade on the pool can change the price, and that large trades may move the price significantly. This type of AMM is well suited for assets that may not be strongly correlated in value.

A second class of AMMs exist which use a different curve, which is “flatter”. This curve is useful because it provides low slippage around the initial rate. These types of curves are well suited for assets that should be correllated in value. A reference implementation on Tezos exists, which includes documentation about how this curve works.

The Flat Curve

The Youves team issues an algorithmic stablecoin on Tezos called uUSD. They have recently deployed an DEX which uses a flat curve for correlated assets. This DEX offers a kUSD/uUSD pair, which should be highly correlated.

Providing convertibility between stable value assets that should be pegged to each other creates arbitrage opportunities for trades who can help keep the price on peg. This is valuable for algorithmic stablecoins, such as kUSD, since arbitrage is an important mechanism for maintaining price stability.

Incentivizing Liquidity in the Youves DEX

The Youves DEX is the first DEX to make use of a flat curve on Tezos, and providing liquidity on the kUSD/uUSD pair seems likely to help stabilize the peg of Kolibri. Therefore, it seems advantageous to provide a kDAO farm to incentivize liquidity on this DEX.

Impermanent loss only occurs when assets diverge in price. Therefore, on a flat curve DEX with correlated assets, impermanent loss is very unlikely to occur. To this end, this proposal suggests creating a new farm for kUSD/uUSD holders with the same parameters as the kUSD and QLkUSD farms. Specifically:

  • 75,000 kDAO allocated to the farm
  • 1,048,320 blocks run time (1 year @ 30 second blocks)

Costs

The Kolibri Community Fund contains 350,000 kDAO. Kolibri Governance Proposal #17 allocates 741 kDAO for a hackathon prize. KIP-021 suggests allocating a 56,260 kDAO to a new farm. Assuming that both of these proposals optimistically pass, the Community Fund would contain 293,009 kDAO.

This proposal would allocate an additional 75,000 kDAO to a new farm.

which would leave 218,009 in the Community Fund.

Costs

The Kolibri Community fund contains 350,000 kDAO. Kolibri Governance Proposal #17 allocates 741 kDAO for a hackathon prize. Optimistically assuming the proposal passes, there is 349,259 kDAO remaining in the community fund.

Sunsetting the kUSD farm early recovers 18,750 kDAO, and 75,000 kDAO will be allocated to the new farm. Effectively, the cost of the new farm is 56,250 kDAO.

After this proposal, assuming Kolibri Governance Proposal #17 passes and KIP-021 is implemented, the community fund will have 293,009 kDAO remaining.

Implementation

Farming Contracts can be deployed outside of the DAO. These contracts will not have any kDAO in them and will therefore not emit until the DAO proposal passes.

A lambda can be submitted to the DAO that will call the sendTokens(75,000, <new farm address>) entrypoint on the Community Fund Contract, to transfer 75,000 kDAO to the new farm contract.

2 Likes
#2

I’m always in favor of incentivizing liquidity but IMO uUSD isn’t an optimal choice

it’s preferable for us to subsidize a pair that has a direct way to access liquidity, given the state of the ecosystem wUSDC is the only option that is easily bridged to ethereum
I believe with the recent announcement of plenty and wrap about the future of the Wrap protocol it’s safe to say wUSDC is here to stay and also the bridge is probably gonna be expanded to avalanche which will drastically reduce the cost of bridging
Besides IIRC uUSD had been around 10% offpeg for a long period of time before the flat curve was deployed.

Since the code is opensourced we could deploy our own flat curve kUSD/wUSDC cfmm and a farm for its LP.
I can think of a few problems/tasks that it raises:
Who’s going to deploy it? Deploying an opensourced piece of code sounds like a trivial task that I could do
Who’s going to create a UI for the swap page? We can outsource it

Regarding KIP21 I’d support the idea of sunsetting kUSD farm without creating a KSR farm which I believe to be a little excessive, the KSR is not as important for the protocol as liquidity and in fact is self-sustainable.
An additional buying pressure is only needed when kUSD is going below USD, during these times the stability fee goes up providing the KSR participants a better return

#3

Issue with this statement: “It’s preferable for us to subsidize a pair that has a direct way to access liquidity, given the state of the ecosystem wUSDC is the only option that is easily bridged to ethereum”

Although this is currently the only automated cross-chain wrapping solution, and despite my own admiration for the Bender Labs team and for the technology that they build, wUSDC cannot scale nor gain institutional grade utility. Besides the fact that automated wrapping solutions should be under the microscope now because of the Solana-Ethereum wormhole hack, USDC circumvents all of the KYC required to mint/redeem USDC. USDC Is what it is because of that. Without it, and by circumventing it, wUSDC becomes a target for money laundering. That’s the tradeoff, and that’s why institutions are staying away from it.

Conversely, OneOf and Dogami already said USDtz will be their primary stablecoin. More such announcements will follow from commercial institutions.

Very easy to say “of course you’d say that Kevin, aren’t you involved with USDtz.” Yes I am. We do not have any reserve lending, nor do we have a governance token a this point. There is no business model to which I would profit form the scaling of USDtz. We’re focused on just making it the best coin for everyone. USDC can be used as collateral, as is GUSD, BUSD, PAX, TUSD (dollar-backed compliant stablecoins; not USDT, not DAI)

As USDtz was the first dollar-backed stablecoin on Tezos, we designed it in manner that would be

  • Compliant
  • Scalable
  • Dapp Programmable
  • Tezos-native

Really the only friction (for some) is buying at scale without going through a minter or minting directly. As liquidity has increased the need to do that has significantly declined as more and more people are able to buy and sell USDtz on the markets themselves.

Furthermore, we’re building a FIAT on/off ramp at this very moment :slight_smile: which would further streamline the process.