PSP-IPΔXX: PSP Liquidity Balancing Strategy

PS Research
, ,
1

PSP-IPΔXX: PSP Liquidity Balancing Strategy

Keywords

PSP Liquidity, Balancer

Simple Summary

To migrate liquidity from the outdated Bancor pools and add new liquidity onto UniSwap V3 to balance the sePSP2 skewed pool liquidity.

Context and Goals

The proposal seeks to address two issues that the redactor has noticed from the current liquidity situation of PSP: The skewed liquidity depth, and Bancor V2 liquidity. To best approach the issues and solutions required, this proposal can be roughly divided in two sections:

1. Skewed Liquidity Depth

Since the release of PSP 2.0, liquidity for PSP on Ethereum has seen a significant boost thanks to the sePSP2 token. Its TVL has grown by ~650% from a value of $1M USD to $6.5M USD. This has allowed for a significant amount of new stakers being able to participate in the ecosystem and reduced the pressure for other sources of liquidity provision, such as liquidity mining, which were leading to unsustainable mercenary liquidity.

To reduce the amount of potential impermanent loss experienced by sePSP2 stakers, the DAO opted for sePSP2 to be composed of a balancer pool token (BPT) made of 80% PSP and 20% ETH. However, while this has led to stakers experiencing more exposure to PSP and reduced IL, the dominance of the Balancer Pool has led to a very skewed liquidity depth. Seen below is an example of the consequence of this: while a 20 ETH > PSP tx only has a 1.7% impact, the opposite scenario has a 10% impact.

369×511 21.5 KB

To mitigate this imbalance, a skewed pool is going to be needed aimed at providing liquidity on the PSP > ETH side.

2. PSP <> BNT liquidity

Currently, around 1.5 Million PSP is parked on Bancor V2. This version of the protocol is currently deprecated, and for this reason, this proposal seeks to also withdraw this PSP and place it into the newly proposed solutions for liquidity described in the next section.

Means and Metrics

To resolve the aforementioned issues, the proposal suggests for the following actions to be taken:

  • Unstaking of the PSP on Bancor - this unstaked PSP will be used as a source of liquidity in the new liquidity positions being opened. In total, there are approx. 1.5 M PSP staked in Bancor V2.
  • Creation of a 1% fee PSP / USDC Uniswap V3 position -
    • For the first liquidity proposal, we propose to deploy 10M PSP ranging from spot price from a +5% of the time of management to ~ $0.5 PSP.
  • Creation of a 1% fee 20% PSP / 80% ETH Uniswap V3 position
    • For this first ETH liquidity position, we propose a 15 ETH deployment from the DAO treasury, with additional ETH being deployed if this liquidity position is found to be effective. We find this amount to be a good stating amount, as it constitutes a significant position while not being a significant amount of the DAO treasury (less than 30%). The remaining PSP will be dependent on the price ratio at the time of the opening of the position.
  • Permission for the Governance committee to manage the positions as needed as long as they are within range.
    • Fee claiming can be done without communicating, and will be part of the DAO treasury automatically.
    • If necessary, position adjustment can be done by GovCo, but a clear communication will be needed following the potential alteration of a position
    • Finally, if the Governance Committee deems it necessary, an active liquidity manager could be appointed in the future to maximise the capital efficiency. Depending on the demands of the manager and the budget required for this, this last part would have to be discussed through govenance channels first.

The easiest way to measure the growth of this proposal is by assessing the liquidity depth of PSP after the implementation of all pools.

Forward-thinking considerations

This proposal was written with the current price and liquidity of PSP in mind. In the future, some of these positions might become out of range and might need to be either re-assessed or funded with additional treasury funds.

It is also important to note that , due to the nature of sePSP2, if we see a significant growth of amounts staked in sePSP2 the aforementioned proposal might not be enough , and the divergence between upwards and downwards liquidity could re-emerge.

Voting options

  • For
  • Against
  • Abstain
5 Likes
2

It seems obvious to remove the unnecessary liquidity of Bancor effectively.

It is also obvious to create a stable and permanent liquidity, and especially a liquidity that will not be held by the majority of stakers.

If I understand the idea correctly, you are proposing the creation of 2 additional pools via V3 of uniswap?

The first 80% ETH and 20% PSP which would counterbalance the Balancer pool to limit slippage in the direction PSP->ETH.
This seems to me a nice idea and a nice first use of the ETH of the DAO.

The second would be a pool fed with 100% PSP (10m PSP) which would be 5% above the market price with a 1% fee on trades.
This pool would no longer operate above 0.5c per PSP (see V3 uniswap pool operation).
This would allow certain arbitrage possibilities but also the diversification of the DAO’s funds while creating a new source of liquidity?
Eventually, we will have to choose between rebalancing the pool by injecting new PSP or letting it fill up with usdc.

I update my word.
Maybe the pool 80ETH/20PSP will be too small against our big balancer Pool and in definitive not usefull.

Furthermore, as @enerow found. There is a PSP/USDC pool on V3, like the one you suggest.
The fees are 0.3% with an initial liquidity of roughly 9m PSP.

So to be competitive with it, you’d have to either reduce the fees or have a much higher liquidity?

1 Like
4

You’re facing the same bug I reported few months ago to the team (to Younes on 7th of Feb 2023).

Capture d’écran 2023-04-05 à 14.05.43
Capture d’écran 2023-04-05 à 14.05.431320×168 15.8 KB

If you focus on the “Receive” form field you will get a bugged price & price impact.
This issue has been there for as long as I remember using ParaSwap.

IMAGE 2023-04-05 13:34:31
IMAGE 2023-04-05 13:34:31881×1280 52.4 KB

Here I focus on the “Receive” field - 11% price impact for 843K PSP / 20 ETH

IMAGE 2023-04-05 13:35:03
IMAGE 2023-04-05 13:35:03824×1280 50.2 KB

Here I focus on the “Pay” field - 1.95% price impact for 843K PSP / 22.0333 ETH

For 20ETH, the true values are :

IMAGE 2023-04-05 13:43:53
IMAGE 2023-04-05 13:43:53839×1280 49.8 KB

Price impact 1.62% selling PSP to ETH

IMAGE 2023-04-05 13:44:04
IMAGE 2023-04-05 13:44:04831×1280 48.3 KB

Price impact 1.53% buying PSP from ETH

Yes, makes sense.

Correct me if I’m wrong, but that’s like selling $500K worth of PSP to the market.
I think this part should be another proposal with many questions to answer :

  • Do we all agree to sell DAOs treasury PSP now?
  • At what price?
  • For what purpose?
  • How? Why UNI V3 100% pool and not OTC deals? What are the other options?
  • For what asset in exchange? Why USDC and not ETH or other assets?

I don’t think we should do that, for those reasons :

  • The sell/buy imbalance is really small, for 20ETH 1.62% vs 1.53%
  • In the event $PSP outperform $ETH the Impermanent loss will be HUGE.
  • We could better use the DAOs treasury.

Nevertheless, having a 1% fees UNI v3 50% PSP / 50% USDC PoL of a significant size could make sense.

This would help us generate trading volume on $PSP, even when the price of $PSP is not moving at all, when $ETH is moving it constantly create arbitrage opportunities between the PSP/ETH and PSP/USDC pools.

For exemple to get a ChainLink price oracle we need to constantly make $3M daily volume for 30 days.

1 Like
5

Hi there, I’m Brian from Gamma Strategies, which is active liquidity manager on top of Uni v3.

Seems to me that the main crux of the issue currently is that there’s more price impact on PSP sells than buys due to the 80/20 liquidity structure of Balancer. There’s less liquidity to sell into and additionally the liquidity is on a full range AMM, further compounding the price impact on sells.

Just wanted to share some resources with regards to a 80% ETH / 20% PSP position on the 1% ETH/PSP pool on Uniswap.

Please see this price impact simulation here for the ETH/PSP 1% pool on Uniswap:

In this price impact calculator here, these are the following assumptions:

  1. $200K ETH & $200K PSP are placed in a dual-sided LP (Base Position) from 50% of current price to 200% of current price of PSP/WETH

  2. $600K ETH is placed in a single-sided position right above the current price (Limit Position) to 50% above current price of PSP/WETH

  3. Assumes no other liquidity is in existence (therefore, this is a worst case situation. The presence of other liquidity will diminish price impact even further)

Results:

  1. The price impact for a 10 ETH sell is 0.95% on only $1M of liquidity, of which $800K is WETH liquidity and $200K is PSP.

  2. Assuming that trades go thru both the Balancer and Uniswap pools, the price impact on the sell side would likely be materially lower.

  3. Assuming the same amount of liquidity (assuming 200K WETH / 800K PSP), the Balancer pool would suffer 9.92% price impact vs. 0.95% price impact on Uniswap. The Balancer 80/20 position is a significant decrease in capital efficiency due to the fact that is a full range position with less ETH allocation.

Happy to answer any questions from the community here about the liquidity situation!

2 Likes