Allow Bonded Positions to convert to Staked Positions

This proposal is for a feature to allow users to seamlessly and instantly transition their bonded liquidity from all pool types into a fully staked OSMO position, bypassing any lock period present.

Background

Osmosis has historically implemented a 14-day lock period for staking and liquidity provision.

For staking, this gives the chain the ability to slash malicious actions if they are discovered after the fact by other validators.

For liquidity provision, the bonding period was initially intended to improve liquidity stickiness during volatile periods, which has been unpopular with liquidity providers and has resulted in emergency unbonding proposals such as Proposal 226 during the UST depeg event and drafts being raised during subsequent USDC and WBTC minor depegs.

With the introduction and rollout of Supercharged Pools, the bonding period has been removed for standard liquidity provision in these pools. The lock period is still in place for Superfluid staked positions due to the same requirement as Staked OSMO.

Users who wish to transition their liquidity from a Superfluid staked pair, whether in a Classic pool or a Supercharged pool, to full OSMO Staking must first unbond their liquidity, wait for the 14-day lock period to expire while not receiving staking rewards, swap to OSMO, and then stake.

Proposal

This proposal asks that a feature be implemented that allows a bonded position to be migrated to a fully staked OSMO position without undergoing an unbonding period.

This feature would take the form of an on-chain message that breaks any current bonds (including superfluid), swaps all assets involved to OSMO, and stakes this. As the same bonding period would exist in both the start and end states of this message, there would be no way for staked value to become fully unbonded by a bad actor through this to avoid slashing.

Enabling this message would reduce friction for the transition of liquidity to a staking position by minimizing the time taken for this user flow and maintaining staking rewards during the process.

This user flow results in more OSMO being staked that otherwise may only be staked at a superfluid discount which improves the security of the chain.

As well as the specific user flow this smooths, this mechanism allows users to instantly freeze impermanent loss on a bonded liquidity position by moving entirely into an OSMO staking position. This ability may make liquidity provision on non-Supercharged pools, which may be more sustainable than Supercharged pools for highly volatile tail assets or have long-term external incentives already in place, more attractive. This also reduces the impermanent loss risk still present in full range, Superfluid staked, Supercharged pools, which now consist of the only bonded liquidity in Supercharged pools by allowing these to switch to the single asset staked position at will.

This feature would be implemented during a future software upgrade proposal.

Target On-Chain Date: 30th July 2023

2 Likes

Great change and will be an awesome addition to osmosis.

I’m a bit wondering for how many users we would develop such a feature.

People who are on a 14-day lockup in a LP and using SFS are apparently;

  • already ok with impermanent loss
  • already ok with having only half of their assets get staking rewards
  • already ok with 14-day unbonding periods

Is this mainly to counter the advantage LP-ers from Supercharged pools who did not bond for SFS? Because they can swap at any given moment, without the 14-day unbonding period?
Or is it because the incentive models changes for Supercharged pools where long-range positions get less LP rewards?

The big downside if people decide to switch is that the overall liquidity drops. This might be less of an issue in Supercharged pools with a high density around the current price, but for non-Supercharged pools this will make the assets only more volatile which may not be such a good choice imo.

For me this question really is about how many users would actually use this… and I am a bit afraid that the answer is that it only targets a very very small amount of users. Which makes me wonder if it is even worth spending the time to develop with all the addtional risks and measures which have to be taken to make it secure.

For the SFS users in Supercharged pools who will be the only people still bonded in those pools. This also helps with liquidity breadth which is something that concentrated liquidity pools often struggle with.
The work is apparently relatively minor - to the point that this will likely be in the next upgrade.

1 Like

The fact that an emergency unbonding has only been needed to address the collapse of UST and that the emergency unbonding proposals that people began to draft to address the minor USDC and wBTC depegs were not would seem to suggest that the bonding period has be an effective policy.

Additionally, emergency UST unbonding was provided for 10 pools - #560 (UST/OSMO), #562 (UST/LUNA), #567 (UST/EEUR), #578 (UST/XKI), #592 (UST/BTSG), #610 (UST/CMDX), #612 (UST/XPRT), #615 (UST/LUM), #642 (UST/UMEE), and #679 (4Pool)- and while I did not see any draft of an emergency unbonding proposals related to the minor wBTC and USDC depeg events, I only recall a couple comment on CommonWealth and Reddit panicking about the USDC depeg event and calling for emergency unbonding and at least a dozen expressing that the market was over reacting, which it was.

As such, even if the liquidity bonding period is unpopular if it is any effective policy, why is there a need to provide people with the opportunity to circumvent it?

The difference in the percent of bonded liquidity between SFS pools and non-SFS pools (SFS pools have a higher percent of bonded liquidity than non-SFS pools) would also suggest that the bonding period is unpopular amongst liquidity providers that provide liquidity to non-SFS pools.

If SFS was enabled for more pools, would there be less complaints? This would also increase security as well.

Also, are there any safeguards in place? Should roll out be staggered?

1 Like