☕ Liquid Staking Tokens (LST) on other chains

A liquid staking protocol allows users to stake their tokens to earn rewards while still retaining liquidity by receiving a tradable token that represents their staked assets.

The following information is gathered using ChatGPT Deep Research.

In all, the ecosystems of Ethereum, Cosmos, and Solana recognize that staking, while vital for security and rewards, carries inherent governance implications. Each chain’s community is taking steps to balance convenience and liquidity (via exchanges and liquid staking) with decentralization. The dominant theme across discussions is that no single entity should be able to leverage user deposits to gain disproportionate control. Whether through self-imposed limits, protocol mergers, diversified validator selection, or new technical modules, the goal is to ensure user-staked assets don’t unintentionally centralize power in these blockchain networks. The ongoing debates and solutions are a testament to the community vigilance in preserving decentralization even as staking services evolve.

:full_moon: Ethereum

Token (Symbol) Est. Annual Yield % of Total Staked ETH Governance Power / Beneficiaries
stETH ~3–5% ~28.5% Lido DAO; validators chosen by DAO
rETH ~3–5% ~2–3% Rocket Pool DAO; node operators stake RPL
cbETH ~3–5% ~13.6% Coinbase (centralized issuer)
sfrxETH ~4–5% ~0.3% Frax DAO; validators managed by protocol
osETH ~3–4% ~0.9% StakeWise DAO; vault creators manage validators
BETH ~2.5–7% ~7% Binance (centralized issuer)
Kraken ETH Staking ~2.5–7% ~8% Kraken (centralized issuer)

Ethereum has off-chain governance, but a protocol with 66%+ of validators could enforce changes de facto (e.g. propose only blocks from their fork). No direct way to mint ETH through governance — needs social coordination.

Ethereum Governance Power & Risks

Staking on Ethereum does not confer formal voting rights in protocol upgrades (Ethereum governance is off-chain), but control of validators equals influence over block production and network security. Thus, “governance power” in this context refers to the ability to affect consensus (e.g. proposing blocks, transaction inclusion, censorship or outage resilience). Key risks emerge if too much stake is concentrated:

  • Validator Collusion / 51% Attacks: If a single entity or coordinated group controls a majority of validators, they could censor transactions or even rewrite parts of the chain. Ethereum’s design requires ≥66% of validators to finalize blocks – a cartel with two-thirds of stake could finalize only their chosen blocks and potentially violate liveness or safety of the chain. Even >33% control is dangerous: Ethereum researchers note that if one staking pool exceeds 1/3 of stake, it becomes a “centralization attack” on PoS finalitydecrypt.co. For instance, Ethereum Foundation researcher Danny Ryan warned in mid-2022 that “Lido passing 1/3 is a centralization attack on PoS… staking in Lido at these thresholds has a lot of [tail risk]” decrypt.co.
  • DAO or Custodian Capture: When users stake through intermediaries (Lido, exchanges, etc.), they entrust governance power to those entities. This creates a single point of failure – if the DAO’s governance is hijacked or an exchange is compromised/coerced, the huge pool of user-owned validators could be manipulated. For example, a malicious takeover of Lido’s LDO-governed DAO could theoretically reassign its 30% of validators to colluding actors or change withdrawal keys. Similarly, an exchange with a large stake could be forced by regulators to censor certain transactions (indeed, after OFAC sanctions in 2022, some large staking entities only included “compliant” transactions, leading to censorship concerns). These scenarios put the entire network at risk, not just the individual stakers.
Ethereum Community Sentiment

The Ethereum community is very aware of these centralization risks and has engaged in active debate:

  • Calls for Self-Limitation: Many community members have argued that no single staking service should dominate. Ethereum Beacon Chain community lead Superphiz famously challenged providers: “Who will be the first staking provider to publicly commit to limiting themselves to not more than 22% of validators on the chain?” decrypt.co. Even Vitalik Buterin suggested that large pools should “price themselves out” if they grow too big – “if a stake pool controls >15%, it should be expected to keep increasing its fee rate until it goes back below 15%” decrypt.co. This implies big pools like Lido or Coinbase should disincentivize further growth once they hit a safe threshold (~15% was Vitalik’s guideline).
  • Lido’s Response: In mid-2022, Lido considered a governance proposal to self-limit its market share. However, the Lido DAO overwhelmingly voted against imposing any cap (over 99% of LDO votes rejected the limit)decrypt.co. Lido’s community reasoned that capping Lido would simply push users to other (potentially less transparent) platforms, and preferred to continue growing. This decision was controversial. Decentralization advocates were disappointed – they worry Lido’s dominance (which was ~31% of ETH staked at the time decrypt.co and has remained around ~30% since) could undermine Ethereum’s neutrality. On the other hand, some argued that Lido’s internal decentralization (with many node operators) and Ethereum’s ability to withdraw stake mitigate the risk (users can leave Lido if it misbehaves stakingrewards.com).
  • Preference for Decentralized Alternatives: A segment of Ethereum stakers intentionally choose services like Rocket Pool or solo staking to support network health. These users often cite the need to “Exit Lido” or at least diversify stake. Social media and forums frequently discuss Ethereum’s “Nakamoto coefficient” (number of entities needed to reach >33% stake) – currently just a small handful of players (Lido, a couple exchanges) could collude to hit that threshold. This is viewed as antithetical to Ethereum’s decentralization ethos. Efforts like Rocket Pool, StakeWise, and new protocols are celebrated in the community for spreading out stake (Rocket Pool’s ~2% share is small but growing). Core developers and researchers keep a close eye on staking concentration and have floated ideas like making large pools charge higher fees (as Vitalik suggested) or even protocol changes if centralization gets out of hand decrypt.co.

In summary, Ethereum’s staking landscape is varied: users can stake ETH natively or through various pools, but doing so via large providers (Lido, exchanges) transfers a great deal of influence to those entities. The community acknowledges the convenience of these solutions but remains vigilant about their growth. Prominent voices continuously stress the importance of decentralizing ETH staking to avoid scenarios where one DAO or company could “capture” Ethereum’s consensus decrypt.co. This ongoing dialogue shapes how new staking products are designed (with Rocket Pool’s decentralized model and others trying innovative approaches to keep power distributed).



:full_moon: Cosmos

Token (Symbol) Est. Annual Yield % of Total Staked ATOM Governance Power / Beneficiaries
stATOM ~13.9% ~1.7% Stride DAO; Cosmos validators receive stake
stkATOM ~15.7% ~0.5% pSTAKE (Persistence); Cosmos validators

To pass malicious proposals: control >50% of voting ATOM in quorum → ~45–50 million ATOM (with low turnout, possibly less).
To veto block proposals: ~74 million ATOM (33.4% of all staked).

Cosmos Governance Power & Risks

In Cosmos Hub, governance is on-chain and stake-weighted – meaning control of staked ATOM translates directly to voting power over network proposals (upgrades, parameters, community spend, etc.). This makes the distribution of staked tokens crucial for decentralized decision-making. Key risks if user assets’ voting power concentrates include:

  • Supermajority Control (66%+): Cosmos proposals typically pass with a >50% majority (quorum requirements apply and >66.6% “Yes” can veto others in some cases). If a single entity or aligned group controls two-thirds of staked ATOM, they could pass any governance proposal unilaterally, potentially implementing malicious code or redirecting funds from the community pool. They could also halt the chain – in Cosmos’s Tendermint BFT, if ≥33% of stake goes offline or refuses to sign blocks, new blocks cannot be produced. Thus, a cartel with over 1/3 stake could stop the network or veto any proposal (via NoWithVeto votes). The Cosmos Hub today is fairly decentralized (the top 10 validators hold ~43% of stake, not one entity atomscan.com), but liquid staking could change this if one protocol amassed a huge share of ATOM.
  • Validator Collusion / DAO Capture: The introduction of liquid staking protocols adds new dimensions for collusion. For example, if Stride’s DAO (STRD holders) were compromised or motivated to act selfishly, they could delegate the vast ATOM under management to only a few friendly validators, or vote in Cosmos governance in a bloc. The Cosmos community recognized this risk – hence Stride’s plan to merge with ATOM governance to remove the separate STRD governance layer blockworks.co. Similarly, if an exchange like Binance controlled say 15-20% of ATOM stake and decided to vote in concert with a couple of large validators, they could swing votes or form a veto bloc. LSD providers and exchanges effectively act as aggregators of many users’ voting power, so if they fall under one decision-making umbrella, the risk of governance capture increases. Cosmos Hub could find itself with decisions controlled by a handful of actors (be it a big exchange or a liquid staking DAO) rather than the broad community of ATOM holders.
  • Reduced Governance Participation: Liquid staking can inadvertently dampen individual participation in governance. Many ATOM stakers who delegate natively at least have the opportunity to vote on proposals (and many do, or expect their validator to vote in line with their interests). If those ATOM instead reside in a liquid staking contract, the average user might not bother (or have the ability) to vote their fraction. Unless a mechanism like Quicksilver’s proxy voting is implemented, LSD tokens might sit idle in governance, effectively lowering turnout and making it easier for a small active group to meet quorum and pass proposals. This “governance power vacuum” is a subtle risk – power shifts by absenteeism. Some protocols may choose to abstain entirely (neutral), but that still removes those tokens from the active voting populace. The Cosmos community is discussing these concerns; for example, Quicksilver’s design explicitly tries to keep users votingstakely.io, and there are Cosmos SDK improvements (the Liquid Staking Module, LSMstakely.io) being considered to safely integrate liquid staking without losing voting capabilities.
Cosmos Community Sentiment

The Cosmos community highly values decentralization and has been proactively debating staking solutions:

  • General Acceptance with Caution: Liquid staking in Cosmos is still in early stages (only ~5–6% of ATOM’s staked supply was liquid-staked as of Q1 2024messari.io), but its growth has prompted discussions. Many community members see the utility in LSDs – enhancing ATOM liquidity and yield – but are wary of recreating an “ETH/Lido scenario” on the Hub. The Stride merger proposal in late 2023 crystallized this sentiment: Stride’s team themselves noted it was better for ATOM holders to govern the LSD protocol to avoid centralizationblockworks.co. This proposal led to lively debate on the Cosmos Hub forum and social media. Some ATOM holders welcomed the idea of the Hub “owning” its liquid staking (strengthening ATOM’s value proposition and preventing a separate token from accruing power). Others were skeptical about the merger – concerned it was too soon, or questioning how STRD holders converting to ATOM might dilute current ATOM holders. After a week of discussions, the plan was shelved for the time being due to “too much division and not enough consensus” between the Hub and Stride communitiesblockworks.co. This outcome shows that Cosmos governance is working as intended – major changes undergo extensive community scrutiny. While the merger is paused, the conversation significantly raised awareness of LSD-related risks and aligned Stride more closely with Hub interests going forward.
  • Desire to Preserve Governance and Decentralization: On Cosmos forums and in community calls, participants often emphasize that liquid staking should “do no harm” to chain governance. The introduction of the Liquid Staking Module (LSM) is one attempt to integrate LSDs at the protocol level safely, allowing staking derivatives without breaking the link to governance. Projects like Quicksilver have been lauded by community members for their approach to let users vote and choose validators even when using LSDstakely.io. The community sentiment leans towards plurality and competition – multiple LSD providers (Stride, Quicksilver, etc.) are encouraged rather than one monopolist. There is also emphasis on LSD providers distributing their stake across many validators (to not amplify the rich-get-richer effect among validators). In Cosmos, validators themselves are community members, and many top validators have weighed in: some welcome LSDs as increasing total stake and security, while others caution that if delegations concentrate into LSD platform-chosen validators, smaller independent validators could lose delegations. Initiatives like Stride’s and Quicksilver’s broad delegation policies are responses to these concerns.
  • Exchange Influence: The Cosmos community has historically been cautious about exchange-run validators having too much voting power. There have been instances where exchanges (like Binance or Poloniex in early days) constituted a large percent of voting turnout. Community discussions (on Commonwealth and Reddit) have questioned whether exchange custodial staking should be limited or if exchanges should abstain from governance votes to avoid conflict of interest. While not a crisis currently, it’s understood that if, say, Binance Staking and a liquid staking provider ended up aligning, they could together sway proposals. Thus, decentralization advocates in Cosmos continuously promote delegating to a diverse set of validators and participating in governance personally. “Not your keys, not your vote” is a mantra – meaning if you let someone else hold your ATOM (like an exchange or even a liquid staking zone), you’ve effectively let them exercise your vote.

In summary, Cosmos ecosystem participants are guardedly optimistic about staking innovations: they see liquid staking as an important piece of interchain DeFi, yet they are actively engineering solutions (like proxy voting and potential protocol mergers) to ensure governance power remains widely distributed. The prevailing sentiment is that Cosmos must avoid the pitfalls of centralization by making liquid staking an extension of the community rather than an external force. Debates on forums and Twitter reflect a strong commitment to keeping the Cosmos Hub “by the token holders, for the token holders,” even as new staking derivatives emerge.



3 Likes

:full_moon: Solana

Token (Symbol) Est. Annual Yield % of Total Staked SOL Governance Power / Beneficiaries
mSOL ~7.9% ~6.6% Marinade DAO; Solana validators
JitoSOL ~7.3% ~5.5% Jito DAO; Solana validators

Takeover:
33.4% of stake: Can halt block production (by not voting).
66.7% of stake: Can control all block production.

Solana Governance Power & Risks

In Solana, “governance power” of staked SOL manifests as influence over consensus and the validator network health, since on-chain governance for core protocol upgrades is not common. The risks of concentrated staking power are therefore tied to network security and decentralization:

  • Nakamoto Coefficient & Validator Collusion: Solana’s Nakamoto coefficient (the number of validators needed to cumulatively reach 33% of stake – the threshold to halt or control the network) is a key metric. Liquid staking shifting many users’ SOL into one platform could lower this number. For example, Marinade and other stake pools actually helped increase the Nakamoto coefficient by delegating to many smaller validators messari.io. But if a single liquid staking provider (or exchange) amassed an outsized share, they could effectively act as one “super-validator”. If one entity or cartel controls >33% of SOL stake, they could halt the network or censor transactions by withholding votes. At 66% control, they would produce 2/3 of the ledger entries and could confirm only their chosen blocks (though reaching 66% on Solana would be extremely difficult without a massive centralized custody). The risk is not just hypothetical: Solana has had instances where a handful of validators needed coordination to restart the network after an outage – in a highly centralized scenario, those few could decide unilaterally how to proceed. A diverse stake makes the network more resilient to any one validator misbehaving or going down.
  • DAO/Protocol Capture: Solana’s liquid staking protocols come with their own governance tokens (MNDE, JITO, formerly LDO for stSOL). A hostile takeover of a liquid staking DAO could redirect a lot of stake. For example, if an attacker accumulated a majority of MNDE tokens (not easy given distribution and the DAO treasury lockup, but imagine a scenario) they could potentially change Marinade’s code or delegation preferences to favor certain validators (including malicious ones they run). That could concentrate enough stake on those malicious validators to attempt attacks or at least extract excessive rewards. Similarly, if Jito’s governance or operations were compromised, an attacker could exploit the large stake trusting Jito’s system. This protocol-layer risk is why Marinade’s approach of limiting external funding and keeping governance open is seen as safer by the community, and why Lido’s exit was somewhat a relief – it removed an external DAO’s influence. In Solana, the core protocol upgrades are still overseen by core developers and the Solana Foundation, but stake distribution is the community’s responsibility, and any entity controlling a large fraction can undermine the de-facto decentralization, even absent formal governance votes.
  • Censorship and Compliance Pressures: A concentrated staking provider might be subject to legal or economic pressures. For example, if a large U.S.-based exchange or protocol held a big chunk of SOL stake, they could be compelled to censor certain transactions or programs (e.g., not process blocks that interact with sanctioned addresses). In late 2022, about 20% of Solana stake was on validators that indicated some compliance with OFAC guidelines (this was discussed after similar Ethereum issues). Solana’s high throughput and different clients make censorship currently less of an issue than in Ethereum’s proposer-builder model, but the risk remains: a few large validators could choose not to include certain transactions, and if they produce a large share of blocks, those transactions get delayed or ignored. A diverse validator set with no single entity above, say, 10-15% mitigates this – it would require a conspiracy of many to censor effectively. Thus, community members often stress not letting any one staking pool dominate for censorship resistance as well.
  • Economic Centralization and Validator Set Changes: If a liquid staking provider became dominant, they could potentially set requirements or norms for validators that wish to attract that stake. We see this with Marinade and Jito: Marinade requires validators to meet certain performance and decentralization criteria (even a bonding requirement via its Protected Staking mechanism) messari.iomessari.io, whereas Jito might favor validators that run its MEV client. If one of these were overwhelmingly powerful, it could force most validators to adhere to its standards or be left without delegation, effectively giving that protocol indirect control over the validator ecosystem’s makeup. This isn’t an attack per se, but could reduce diversity in client software or setups. For example, if Jito became the only viable way to get delegation, all validators might adopt Jito’s client to not miss out, possibly leading to a monoculture. Monocultures are risky (a bug could affect everyone simultaneously). The community generally prefers a healthy mix – Marinade’s approach coexists with others, and no single standard dictates all validators’ behavior.
Solana Community Sentiment

The Solana community’s perspective on staking solutions is shaped by the network’s drive for both performance and decentralization:

  • Marinade’s Community-First Approach: Marinade Finance has been embraced as a positive force by many in the Solana ecosystem. Its core mission is openly about decentralizing stake. Community members often cite Marinade’s delegation of stake to over 100 or even 200+ validators as a boon: “Marinade aids in decentralizing the Solana validator set without harming staker APY” messari.io. Marinade’s governance process (which moved to the Realms platform in 2023 for transparencymessari.io) allows anyone with MNDE to have a say, and because MNDE was widely distributed and not VC-dominated, the community feels a sense of ownership. On social media (Twitter, Discord), Marinade team members regularly discuss improving the Nakamoto coefficient and have even adjusted strategies in response to community input (e.g., implementing validator score changes via governance to give more weight to under-staked validatorsmessari.io). The sentiment is that Marinade’s liquid staking strengthens Solana, rather than threatening it. That said, some do keep an eye on the Marinade DAO’s large MNDE reserve – ensuring it’s used responsibly over time – and on the risk of MNDE governance attacks, though mitigations (like time-locks or gradual token release) are in place.
  • Competition and Lido’s Exit: When Lido (stSOL) entered Solana in 2021, there was initial concern among Solana enthusiasts. Marinade vs. Lido became a point of discussion. Marinade contributors argued that having an Ethereum-governed protocol run a big portion of Solana stake could be risky; they advocated for Solana’s community to rally around Marinade (a home-grown solution) rather than let Lido take over Solana staking. In fact, Marinade’s team issued public statements welcoming competition but cautioning against centralization. Over time, Lido’s Solana stake remained relatively small, and ultimately Lido decided to withdraw. The community reacted with understanding – Lido’s Solana developers (P2P Validator team) cited lack of sustained funding and usage as reasonsbinance.combinance.com. Solana users largely saw this as “Lido focusing on its core (Ethereum) and Solana focusing on its native solutions.” With Lido gone, Marinade and other Solana-native LSDs gained even more community support. This episode reinforced the idea that decentralized, chain-native solutions are preferred. It wasn’t lost on the community that Lido’s fate on Solana was decided by Lido DAO voters (many of whom are not primarily Solana users)binance.combinance.com – meaning an external community chose to turn off a Solana service. This kind of misalignment is exactly what Solana users worried about. The general sentiment was relief that Solana’s staking future is now in the hands of those directly invested in Solana.
  • Growth of Jito – Mixed Feelings: Jito’s rise as a high-yield liquid staking option has been met with both enthusiasm and caution. On one hand, many users and validators acknowledge Jito’s technological contributions (its MEV infrastructure helps maximize staking rewards and even improves Solana block propagation). Some stake providers started offering “Jito-powered” pools because of the MEV boost. On the other hand, decentralization proponents note that Jito is a VC-backed entity (Jito Labs received funding from investors) and its JITO token distribution is likely more concentrated among insiders. There is an ongoing discussion about how much stake should one MEV-oriented pool have – if JitoSOL becomes too dominant, Solana might become overly dependent on Jito Labs’ relays and validators. Community members on Solana forums and Twitter have called for ensuring Marinade remains a strong competitor to Jito, to avoid a scenario where chasing a few extra basis points of yield leads to centralization. The Solana Foundation and validator community are also exploring protocol-level MEV sharing, which could reduce the attractiveness of any single external MEV solution. In essence, the community appreciates the performance gains but is mindful: “We don’t want all SOL stake going through one funnel.”
  • Validator Community’s Perspective: Solana’s validators are vocal in governance discussions (often via the Validator DAO or community groups). They generally support liquid staking as it brings more stake into Solana (increasing security) and gives validators more delegation opportunities. However, they strongly advocate that stake pools implement decentralization-friendly policies. Marinade’s approach has been developed in close consultation with validators (e.g., the introduction of validator bonding and insurance to allow smaller validators to get Marinade delegation safelymessari.io had broad support). Validators have praised Marinade for not just concentrating on the top 10 nodes, for instance. With Jito, validators benefit if they’re included in Jito’s set (they get more stake), but those outside might feel left out. This has led to conversations about ensuring open access – Marinade is open to any qualifying validator, whereas Jito’s model might be by invitation or performance metrics. Community sentiment thus encourages transparent criteria and open participation in any staking program that holds a lot of SOL.

In summary, Solana’s community tends to view liquid staking solutions as a net positive if they align with decentralization. Marinade is seen as a community-aligned, decentralizing force (often touted as a model for other ecosystems). Lido’s experience on Solana underscored the importance of local governance – the community prefers not to outsource critical network power to an external DAO. Jito’s innovation is appreciated but comes with a watchful attitude to ensure it doesn’t unintentionally recentralize the network. Across forums, one can find Solana users acknowledging the fine balance: “We want Solana to be the high-performance chain, but not at the cost of centralization. Stake pools should help, not hurt, that balance.” The ongoing governance discussions, both informal on social media and formal via on-chain Marinade proposals, reflect a community actively steering its staking ecosystem to maintain a healthy, decentralized network.



Key Risks if Governance Power Concentrates
Risk Scenario Description and Examples
Validator Cartel / 66% Attack A group controlling ~66% of stake can finalize only their blocks or halt the network.
- Ethereum: ≥66% can finalize malicious blocks; ≥33% can halt finality.
- Cosmos: ≥66% can pass any governance proposal; ≥33% can halt the chain.
- Solana: ≥33% can stop block production or censor transactions.
Censorship & Compliance Centralized entities may be pressured to censor transactions.
- Ethereum: Large validators filtered OFAC-sanctioned transactions.
- Solana: Validators following compliance mandates could selectively exclude txs.
DAO Governance Capture Malicious actors could acquire governance tokens to redirect large staked positions.
- Lido: LDO whale could reassign 30% of validators.
- Stride: STRD control could redirect all stATOM votes or delegations.
- Marinade: MNDE attack could favor Sybil validators.
User Governance Apathy Users staking via intermediaries may not vote, lowering governance participation.
- Cosmos: stATOM holders don’t vote, shifting power to validators or abstentions.
- Ethereum: Inertia allows big pools to grow unchecked.
Network Fork or Upgrade Capture Large stakers can influence contentious forks or upgrades.
- Ethereum: Dominant pool could sway validator consensus or exchange listings.
- Cosmos: LSD protocol could threaten stake exit to resist upgrade.
- Solana: Stake majority could stall or reject core fixes.

1 Like

It seems dfinity opted for the self-imposed limits capping the canister controlled neurons to be capped at 10%.

Canister neuron limits don’t limit LST or staking. They limit the easier way of doing it.
What is meant by “self-imposed limits” = LST protocol limits, not IC limits