Re-evaluation of Liquid Staked OSMO held by the Community Pool

Proposal Discussion
1

This set of three proposals would diversify the OSMO currently used to support stOSMO liquidity in the community pool by:

  • Withdrawing 50% of liquidity from the stOSMO/OSMO Stableswap pool
  • Redeploying this into a more efficient stOSMO/OSMO Supercharged position
  • Using the OSMO saved from this capital efficiency improvement to establish Supercharged positions of qOSMO and stkOSMO.

This proposal approves the withdrawal and redeployment of stOSMO/OSMO liquidity as well as performing the community pool spend.

or

This proposal approves the use of 1.25M OSMO from the capital efficiency gains of concentrating the stOSMO/OSMO position to establish a Supercharged position of qOSMO. If the stOSMO concentration proposal fails this proposal is automatically void. If this proposal fails the 1.25M OSMO surplus will be returned to the Community Pool.

or

This proposal approves the use of 1.25M OSMO from the capital efficiency gains of concentrating the stOSMO/OSMO position to establish a Supercharged position of stkOSMO, if the stOSMO concentration proposal fails this proposal is automatically void. If this proposal fails the 1.25M OSMO surplus will be returned to the Community Pool.

Background

stOSMO deployment

20M OSMO was allocated to the stOSMO/OSMO Stableswap pool in Proposal 641. This liquidity was instrumental in the adoption of stOSMO across the Osmosis ecosystem by providing robust underlying OSMO backing for the redemption rate of stOSMO.

The efficiency of this liquidity can be vastly improved.

The Stableswap pool is approximately 32M USD and typically sees around 1M USD a week.

The Supercharged pool has approximately 680K USD of liquidity, mainly in a peg-supporting position. This pool typically sees around 200k USD weekly volume, demonstrating far higher capital efficiency than the Stableswap pool.

1512Ă—585 51.4 KB

Volume per week in stOSMO/OSMO Stableswap Pool

1520Ă—596 49.7 KB

Volume per week in stOSMO/OSMO Supercharged Pool

1015Ă—427 56.4 KB

stOSMO/OSMO Supercharged pool liquidity distribution

Osmosis has previously supported alternative Liquid Staked Tokens (LSTs) ampOSMO and bOSMO through the funding of SAIL DAO in Proposal 708. With ampOSMO’s bribe system for delegations and pending deployment of slow burn arbitrage vaults, and bOSMO’s recent successful LAB streamswap, there has begun to be a variety of LSTs being bootstrapped by Community Pool liquidity.

qOSMO

Developed by Quicksilver, qOSMO differentiates itself in the current liquid staking landscape through features including an intent system where users determine the underlying staked OSMO validator set and fees of only 3.5% of staking rewards.

More details on qOSMO can be found in this blog post.

stkOSMO

Developed by Persistence, stkOSMO utilizes an automated rating system to distribute the underlying staked OSMO to 75+ Osmosis validators and charges fees of 5% on staking rewards.

More details on stkOSMO can be found here.

Importance of Liquid Staking Diversity

Liquid Staking is a unique challenge for Proof of Stake chains, which require a bonding period. To encourage sufficient security to be provided through tokens, these often receive staking rewards in the form of inflationary emissions, transaction fees, or yield mechanisms.

This lack of flexibility is binary by default. Users can lock tokens to receive these rewards or attempt to beat the opportunity cost of those methods by using the unstaked token for alternative use cases such as liquidity provision or active trading.

Liquid-staked tokens provide additional options by removing the locking period in exchange for a fee from the staking rewards. The functionality of this lock removal relies on underlying liquidity for the LST to be sold into, allowing long-term restoration of any slippage incurred by the demand for this instant unstaking by arbitrageurs performing the unbonding period instead.

Liquid Staking centralization has been a contentious topic on Ethereum during the relatively recent move to Proof of Stake. Ethereum’s model requires a minimum staking deposit of 32 ETH, with each key needing to run its own Validation node. For the first two years of the Proof of Stake network, there was also no ability to unstake. These factors all led to a dominant market for custodial staking via liquid staking, which circumvented the minimum staking requirement, the technical barrier to participate, and the only trustless way for a staked position to be sold.

Liquid Staking providers require trust of custody and substantial liquidity to ensure minimal slippage when exiting the liquid staking. Ethereum’s leading liquid staking providers quickly became popular Centralized Exchanges such as Coinbase, Kraken, and Binance, with the tokenized variety being dominated by Lido’s wstETH.

863Ă—447 45.9 KB

Source: https://dune.com/hildobby/eth2-staking

Lido has been close to controlling 33% of all Ethereum stake, allowing one provider to control which transactions are processed by the chain through obstruction. This centralization is the subject of much discussion on Ethereum, as Lido comprises 31 validators. Lido may not be viewed as a single entity for all actions as these actors may be split on a decision.

This alliance of validators is similar to the model of Liquid Staking appchains or DAOs typical in Cosmos. Here, the provider has checks and balances through decentralization, and there would need to be multiple actors with different goals to abuse this minority share. However, all validators running a specific liquid staking protocol naturally favor that protocol’s success as a whole, and most entities are likely to vote selfishly. This results in a soft bias toward decisions that are good or neutral for both groupings rather than independently good for the primary blockchain and potentially negative for the Liquid Staking Protocol.

Cosmos SDK chains, such as Osmosis, have a far lower barrier to non-custodial staking through the validator system. Validators perform the technical task of consensus on behalf of users and governance tasks alongside users. Users can also stake any quantity of the governance token by default rather than requiring a minimum amount.

This results in an initially lowered demand for liquid staking on Cosmos chains, only overcoming the opportunity cost for the unstaking period and the hurdle rate for unstaked assets. These may eventually lead to increased usage of LSTs, especially with the comparatively higher APR for staking on most Cosmos chains against the ~3.4% on Ethereum.

Liquid staking protocols add a layer of stake on top of validators, which includes a soft bias towards membership of their chosen delegation set and the potential for liquid staking protocols to vote using the underlying stake themselves. This means that the risk is not identical to the Ethereum conversations but is still relevant when talking about LST dominance on a chain.

1370Ă—347 9.34 KB

The current state of Liquid Staking on Osmosis is a minority of stake being controlled by these protocols. At the time of writing, there is a clear leader, with Stride having 25 million OSMO under management, 10 million of which are provided by the Osmosis community pool from Proposal 641.

Chart1600Ă—990 57 KB

There is a clear competitive advantage to being the incumbent liquid staking protocol as additional usage results in higher volumes through liquidity pools, resulting in higher reward APRs for liquidity, attracting more liquidity, and so making the liquid asset both more widely and deeply used as collateral or for purely liquid staking purposes.

Suppose Osmosis wishes to have a diversified liquid staking layer. In that case, actions must be taken to limit liquid staking protocol growth through increased fees for use or subsidizing alternative Liquid Staking protocols to encourage liquidity to redistribute to ones with less developed peg backing.

Further Reading:
GSR: The Great Staking Debate, Lido Dominance and Ethereum
Lido Forums: Is Lido good for Ethereum?
Celestia Forums: Defending against LST Monopolies
Lido Forums: Why Doesn’t Lido Self Limit?
Cosmos Forums: Introducing the Dynamic Liquid Staking Tax (Blockworks Research)
DyDx Forums: dYdX Community Staking Proposal

Mechanism

This proposal makes 10M OSMO withdrawable from the stOSMO/OSMO Stableswap pool, split equally between OSMO and stOSMO.

To minimize the time without this liquidity in place, this will first be incrementally withdrawn from the stOSMO/OSMO Stableswap pool and re-deployed into the 0.05% stOSMO/OSMO supercharged liquidity pool to maintain the same level of effective capital.

stOSMO/OSMO will be placed at a range of 1.95 - 1.27 until all stOSMO is allocated to the pool. This position is approximately one-third OSMO to two-thirds stOSMO.

While this reduces the liquidity available for stOSMO depeg resilience by 50%, this concentrated liquidity position from 1.95 to the current redemption rate is approximately 5x more efficient than a stableswap pool, resulting in a net gain of 2.5x in liquidity efficiency for stOSMO resilience. This action also leaves 2.5M OSMO removed from the pairing for usage in further steps.

1.25M OSMO will be placed in a 0.05% qOSMO/OSMO supercharged liquidity pool at a range of 1.1-1.17

1.25M OSMO will be placed in a 0.05% stkOSMO/OSMO supercharged liquidity pool at a range of 0.99 - 1.05

The positions for stkOSMO and qOSMO comprise approximately one-third of the Liquid Staked asset, which will be acquired through a combination of purchase on Osmosis and Liquid Staking OSMO, according to which returns more of the LST.

The upper limits are all set to be the expected ratio of the LST to OSMO in approximately four months, meaning that this allocation to each LST is temporary and that the distribution of OSMO peg backing for all OSMO LSTs should be re-evaluated at that point based on a future governance proposal.

All positions will be returned to the Osmosis Community Pool for custody by the Osmosis DAO.

These tasks would be performed by the Osmosis Liquidity SubDAO, which is a 4/6 multisig comprised of

This multisig has previously been used in Proposal 715 to deploy liquidity to stSTARS/STARS and in Proposal 756 to deploy AXL/OSMO to Astroport. This subDAO acts as an intermediary to perform multi-stage or time-dependent transactions, such as adding liquidity to a pool with a ratio of assets that will vary before a five-day Osmosis governance proposal is completed.

Target Onchain Date: 16th April 2024 for all three proposals

4 Likes
2

I think the diversity is definitely a good thing, and in the short term this is fine, though I would like to note that the numbers involved here continue to distinctly favor stride from a governance standpoint, but of course that’s what the governance proposal did.

It might make more sense to simply eliminate all subsidies.

I hope that overall, we move toward a world where governance no longer participates in consensus and governance. Setups like these, which previously enjoyed my support, taint consensus and governance with free vote power.

Everyone who has paid for or worked for osmo is shorted by these arrangements because it dilutes their votes in both consensus and governance.

When I think about it, osmo is a governance token. I don’t want any osmo that doesn’t have a cost to enter consensus, as an osmo holder.

As a qck holder, it’s expedient, but against the philosophy that we aim for.

Recently, individuals who have been involved in governance proposals on the cosmos hub, to make acquisitions, have begun to once again discuss acquisitions. As someone who doesn’t want the hub to acquire osmo, because I think that the price of the token and quality of the community would be negatively impacted, I’m concerned by the unpaid for pumping of vote power associated with the AEZ via stride.

We have seen close governance proposals on a number of occasions and yes, 20 million osmo is enough to make a difference, especially when one considers the lobby effect of this osmo. Stride validators on osmosis (including myself) have voted themselves additional consensus and vote power.

This comes at the cost of the sole utility of osmo: governance. Everybody who worked or paid for OSMO is diluted by this measure and not only financially.

However, if we were to consider the financial stakes for a moment, 55.8% of osmo is staked, 364.97m tokens. Of those, governance has allocated 20,000,000 to stride, or about 5.95% of total vote power.

365/20 = ~5.95

Another way of saying it, is that simply repealing 100% of all governance-funded liquid stake to stride would increase what stakers earn by 5.95%.

But that isn’t all of it, I believe that there is additional governance funded liquid steak, with backbone for lab, but it’s dramatically smaller. I do believe that’s 650,000.

I think that the creation of an alternative proposal, which eliminates this practice altogether, may be the best path forward.

I do not intend to get in the way of this proposal, and due to conflict of interest, will be voting abstain.

I will begin to draft the alternative in coming days.

1 Like
3

This is a well thought out and researched proposal.

As you rightly point out - the incumbent liquid staking provider has an unnatural competitive advantage as a result of the liquidity provided by the community pool.

For an LST - liquidity is often the biggest differentiator between competing solutions - but at the same time, the overall availability of liquidity is increasingly fragmented by supporting alternative providers products.

This proposal I believe is an appropriate and equitable first step towards better outcomes with regards to decentralisation and diversification. At the same time, the proposal through the capital efficiency gains preserves the usability and security of the incumbent liquid staking solution, which preserves outcomes for users.

Whilst I support the proposal as it presents an improvement, I believe that in the long run, the market should be able to determine which liquid staking solution to support rather than being driven by community owned liquidity toward one or a few specific solutions.

1 Like
4

In order for the market to function, subsidies must be eliminated.

Furthermore, in order for people who paid for or worked for their osmo tokens to have the voting rights that they’re supposed to have, subsidies must be eliminated.

5

I hope that this kind of method will remove, or at least minimize, the liquidity benefit that is present and allow the market to decide this based on other options such as:

  • Security profile of the LST underlying custody
  • Fees charged for usage
  • Validator selection
6

@JohnnyWyles many many thanks for starting this. I sincerely love it.
There are more contentious proposals currently regarding Protocol-owned-Liquid-Staking (on Stargaze (Mintscan) and on DYDX (Mintscan)) whereas it is slowly but surely getting more criticism for the lack of LST-diversification. Seeing Osmosis taking the lead on actively working on diversification and thereby reducing the LST-risk is awesome!

That being said; I support this proposal.

  • are the mentioned ranges enough to bridge the coming 4 months? Such that it won’t take active maintenance by the liquidity DAO to manage the position?
  • after 3,5 months from now, we will reassess the outcome of this change? Any thoughts on specific metrics to take into account? Like resiliency of depegging, the share of LST-providers, decentralisation of VP on chain (which would be a very nice side-effect) and more?
  • if this change is successful, would it be foreseen to do the same procedure with the remaining funds in the stOSMO/OSMO stableswap pool?

I will be looking forwards to this proposal. I have been against utilizing the community pool for on-chain staking from the get-go and will continue to be against it. If the economic security of a chain is not good enough, then a mechanism is required to make staking more attractive for investors. Raiding the CP is not the answer, since it only dilutes staking rewards for the people who bought and staked their OSMO believing it will succeed and getting staking rewards in return. They see their APR go down by free money from the CP. The only beneficials in this on the long run are the LST-provider and the validators who receive an additional stake.

I am aware that the (validator)“community” has signalled approval for this Protocol-owned-Staking, but it can luckily also be reverted.

Thanks! This is proper validator behaviour, really cool to see that Notional is taking the right approach and responsibility.

2 Likes
7
  • are the mentioned ranges enough to bridge the coming 4 months? Such that it won’t take active maintenance by the liquidity DAO to manage the position?

They should be - the calculation was based on the APR for Stakers provided by Data Lenses. Since this varies it could be longer or shorter of course.

  • after 3,5 months from now, we will reassess the outcome of this change? Any thoughts on specific metrics to take into account? Like resiliency of depegging, the share of LST-providers, decentralisation of VP on chain (which would be a very nice side-effect) and more?

Metrics to look at would be things like % of OSMO staked, % of Stake that is LST, whether we have moved towards a more varied model (after discounting Community held LSTs)
The other ones you mentioned will also change but they might not be directly impacted by this. resilience should go up but that also depends on usage and market conditions.

  • if this change is successful, would it be foreseen to do the same procedure with the remaining funds in the stOSMO/OSMO stableswap pool?

Yes, I’d like to follow up with this, assuming that we do see the expected increase in resilience for stOSMO.
Now that 20M OSMO has been allocated for the purpose, it can remain in circulation for LST support, but we should also be careful about inflating the supply of staked OSMO, as Jacob is saying.

1 Like
8

I agree - I think this is a valuable first step and will allow the LST’s to operate on a more even playing field with a reduced external competitive advantage.

9

Just want to add some thoughts for all 3 components of this proposal!

Moving half of the stOSMO / OSMO POL Position to a Supercharged Pool

Supportive of the clear gain to capital efficiency for the stOSMO / OSMO liquidity. However, I disagree with subjecting the stOSMO position to such a short custody period in the pool. I’m not sure I understand why the position’s value needs to be re-evaluated on such a short timeframe. Let me break down some numbers demonstrating what that value is. Because of having deep stOSMO liquidity to support DeFi integrations:

  • stOSMO is now the third largest collateral asset in the Cosmos ecosystem, behind USDC and stATOM.
  • stOSMO represents over $6.5 million in locked OSMO across Mars, Umee, Shade, Kujira, Inter-protocol, and Membrane. This is more than 2x Quicksilver’s entire TVL

It’s abundantly clear that this position has brought value to Osmosis, driving scarcity for the token through a diverse range of usecases. By making aggressive changes to this position (effectively phasing out half of it in 4 months), there’s a real danger of that value evaporating, especially with yield tokenization platforms soon to launch for the first time in Cosmos, which could generate hundreds of millions of dollars in volume for the defi platform(s) they are hosted on. These protocols require deep LST liquidity on the buy side to function well, and without it, stOSMO could be left behind as a valuable collateral choice in favor of other assets like stATOM or Neutron’s wstETH.

Stride has also heavily invested in Osmosis, allocating millions in incentives over the course of the protocol’s lifecycle. While this has undoubtedly benefitted Stride as well, this has also significantly benefitted Osmosis. Some more numbers to look at:

  • The stOSMO in Osmosis’s POL position (10M OSMO worth) represents approximately 7% of Stride’s total TVL.
  • In contrast, Stride’s stAsset pools represent a whopping 32% of Osmosis’s DEX TVL, making Stride Osmosis’s largest liquidity partner.

It’s clear that Osmosis and Stride bring a lot of value to eachother. It’s odd, and a bit ironic if you think about it, to see stOSMO represented as some sort of systemic threat to Osmosis despite the huge amount of TVL that Stride has brought to Osmosis.

I agree with moving the position to make it more efficient, but think there’s a lot of risk in maintaining such a short-term position. Would like to see this be changed.

Allocating a Portion to stkOSMO and qOSMO pools

History has shown (with the ampOSMO and bOSMO POL proposals) that allocating significant POL to LSTs that people aren’t already using doesn’t subsidize adoption of those LSTs. Below, you can see the 24h trading volume history for ampOSMO and bOSMO. bOSMO has had less than $500 in volume today, and ampOSMO has had no volume at all.

image
image1920Ă—923 71.1 KB

There’s very little reason to believe the situation will be any different for pStake or Quicksilver, but I understand the need to try. i just hope that if history repeats itself we won’t continue to yeet otherwise productive funding into the void like this.

To that end, I’d propose that part of what we re-evaluate in 4 months’ time include usage scenarios like:

  • Value in secondary DeFi integrations (lending protocols, CDPs, yield-tokenization platforms, etc)
  • Trading volume (adjusted for liquidity discrepancies)

Osmosis is a DeFi protocol. It relies on volume, TVL, and secondary integrations to succeed. To ignore that where LSTs are concerned makes no sense.

3 Likes
10

This seems very reasonable - a data driven approach towards allocations makes the situation more equitable.

As you rightly say, LST’s are reliant on deep liquidity - the existing situation provides an unnatural competitive advantage sponsored by the community pool.

This proposal seems a reasonable step in the right direction towards an even playing field. It is entirely reasonable to expect that once the field is more even, the results need to measure up so as to not adversely impact Osmo holders who are represented by the community pool.

11

I think you’re confusing value and throwing off both governance and consensus. If we have a bad enough consensus problem, the value of osmo is zero. If governance fails to function, the value of osmo, a governance token, is zero.

We should end these practices immediately.

12

I’m not.

You’re choosing to view each of those two things in a vaccuum for some reason.

In the interest of keeping stake weight “organic” (as though foundation delegations aren’t already significantly impacting validator stake weights), you’re ignoring the economic impact of removing this liquidity on Osmosis.

These are two sides of the same coin, and it’s irresponsible to cast aside the economic impact such a decision could have.

13

I don’t think anyone disagrees here, but the attempt is to spread the systemic risk which is inherited by having a LST in the first place, since it relies on more than 1 protocol (not just Osmosis as a base).

I am a great fan of spreading the risk over multiple LST-providers, since that will also spread the risks. Ofcourse it also introduces risks of the other protocols, I am also aware of that.

14

What is a governance token worth if it is free?

15

I’ll also note that Imperator, an investor in stride and recipient of stride stake on many cosmos networks, is voting against quicksilver and persistence getting governance funded POL, with stake that comes to it form stride via governance.

Seems really off… but that’s what happens when free stake from the community pool votes, I suppose.

I:

  • respect their right to vote as they please
  • note the conflict of interest
  • note that the conflict of interest is caused directly by the practice of funding protocol owned staked liquidity. Other forms of POL would not cause this conflict of interest.
16

Could you expand on why you perceive it as half the position is being phased out in 4 months? As far as I can tell, this would just be moving that liquidity to purely being peg supporting, rather than supporting buys, which anyway mostly happen through the stride UI? Lending markets are depending on the ability to liquidate derivatives into Osmo or USDC (depending on what they are collateralizing)

Also wouldn’t there presumably be a prop to get re-adjust the pegs in 4 months?

17

The proposal avoids saying this. It’s not clear to me that this is the plan at all. If it is, it should be more clearly specified here.

A couple things here:

First, the liquidity would transition to being peg supporting, but after another month or two it wouldn’t even be effective for that purpose, as stOSMO continues to rise against OSMO.

Second, multiple yield tokenization protocols (including one launching directly on Osmosis) will be live in the next month or two. These require deep liquidity on the buy side to process entries. Eliminating the bulk of buy side support for stOSMO will not only harm app devs building these apps in reliance on Osmosis liquidity, but also make OSMO a less attractive asset for participating in yield tokenization. Large parties will simply choose to tokenize LSTs with more buy-side liquidity (like stATOM). Not an outcome we should be pushing for, imo.

18

Note that @JohnnyWyles already indicated that it should be ok for the coming 4 months, after which it would need repegging.

Is that a bad thing in the end?
I mean, initially all pairs were trading against OSMO. We have shifted from that already.

OSMO does not necessarily be the asset where most trading happens, the protocol should profit from it. Which it does through the taker fee and more.
And that is beneficial for Osmosis as a platform and OSMO as an asset, making it more interesting to buy, hold and pool.

19

Yes, for all three assets, this should give us time to establish a policy for peg backing diversity for any OSMO derivative and make this liquidity far more useful.

Why?
I haven’t heard a good counterargument to the reasoning that this should be routed through the minting process if the market rate is so close to the redemption rate that it cannot process the order. Processing the buy order above the redemption rate is detrimental to the user compared to following the minting process.

Suppose the market rate is below the redemption rate, which it should be by around 0.5% due to the pool spread factors + opportunity cost of unstaking. In that case, there is sufficient buy-side liquidity for typical usage.

Deeper buy-side liquidity can be provided by regular LPs with a lower risk appetite than collateralization usage. These LPs then sell the stOSMO at a premium to its worth and get yield on top of the LST itself. This is far more appealing than simply holding the stOSMO in a wallet.

I believe that a future goal for this liquidity is to use Redemption Oracle contracts, such as those used by the Stableswap pools, to set an upper tick limit for a vault that repositions regularly.
Until that is in place, these can be moved upwards every few months to minimize the inactive LST owned by the Community Pool.

1 Like
20

We totally support the reallocation to the concentrated liquidity pool but we will not support the reallocation to different LSTs. The LST provider diversification comparison to Ethereum is misleading and does not apply in the Cosmos. On the contrary, spreading the existing community allocation voted to Stride to new providers clearly poses a political problem. We suggest other LST providers should raise separate community pool spend proposal to bootstrap their own ecosystems’ liquidity.
pro-delegators-sign

1 Like