Solana PoS Staking’s Revolutionary Protocol: Sanctum

Original article by SHLOK KHEMANI

We explain the different approaches taken by both. Finally, we tear down Sanctum, a novel protocol that rethinks liquidity staking on Solana.

Transaction networks need a high level of security to gain trust. If someone could change a SWIFT wire instruction or a Visa transaction, people would lose trust in those systems. The same is true for blockchains. Security determines how well users accept them. For example, the Bitcoin blockchain has the highest hash rate behind it.

So, we have a basic understanding that malicious actors cannot manipulate transactions once they are recorded in the network. But the cost of transacting on Bitcoin is very high.

In recent years, low-cost networks like Solana and Ethereum have transitioned to a Proof-of-Stake (PoS) consensus mechanism. Unlike Bitcoin, which relies on computing power, these networks use staked capital as a measure of economic security.

Before we dive into how this works, here’s a quick overview of some of the terms you might see in this article:

Validator: A user who secures a PoS chain.

Stake: Validators earn the right to create blocks, process transactions, and secure the network by locking up an amount of the network’s native currency as collateral. This collateral is called their “stake.” Some networks, like Ethereum, have a minimum stake, while others, like Solana, do not.

Leader: The network selects a validator, called a leader, to create the next block. The probability of being selected as a leader is proportional to the size of their stake and other network-specific factors. Once a leader creates a block, other validators in the network verify the validity of its transactions.

If the network accepts the block, the leader receives a block reward issued by the network and transaction fees paid by users.

Slashing: If other validators deem a block invalid, the leader may lose a portion of their tokens as a penalty in a process called slashing. Validators typically have a large financial exposure to the network they help secure. As a result, they have little incentive to pass on flawed data to the network. If they do, they lose tokens through slashing.

When a PoS blockchain works as intended, the rewards earned by honest validators accumulate to form a stable return on the staked tokens, usually expressed as an annualized yield (APY). On Ethereum, this return is usually between 2-4%.

These returns from staking have three functions. First, they secure the network. Second, they incentivize long-term participation within the ecosystem. Third, they help ensure that long-term participants are not diluted by inflation.

If you think of the network as a city, staking is like building a house in that city. It keeps you there for the long term, and it appreciates over time.

The Current Dilemma

Staking has its advantages, but they aren’t cheap. Just like building a house in the real world, people may not have the time, energy, capital, and skills to set up a validator node. Everyone wants yield, but expecting to run a validator for yield may not be feasible. That’s where delegated staking comes in.

The concept is simple: users delegate their stake to a validator, which then returns the earned rewards to the user after deducting a percentage of the yield as a fee.

While delegated staking solves one problem for users, it creates another.

When users hold the chain’s native currency, it is liquid. They can sell it at any time, or deploy it in DeFi protocols to earn additional yields, such as lending and liquidity pools. However, once a token is natively staked, it becomes illiquid. Stakers must wait for the bonding period, a cool-down period, to pass before they can withdraw their stake.

On some PoS chains, this can be up to 21 days. They also give up the opportunity to earn additional yield while the token is staked. I guess you can't both hold your stake and eat (yield) in this game.

Liquidity staking is a solution that allows users to stake their tokens through a protocol that mints Liquid Staking Tokens (LST) that represent the staked assets. These LSTs can be freely traded on exchanges and used in DeFi applications to provide liquidity to users. When users want to withdraw, they can provide their LST to the staking protocol in exchange for native currency, and the protocol will then destroy the LST.

If too many users rush to exit their liquidity staked assets by trading on exchanges, it could cause a decoupling. This happened to Lido’s token last year, when the price of its liquidity staked token fell below the price it could be redeemed for — a kind of bank run.

Staking plays a central role in securing PoS blockchains, and has become one of the most important areas in crypto due to its fundamental utility in unlocking illiquid capital.

The total value locked (TVL) in liquidity staking protocols across chains accounts for more than 50% of all DeFi protocol TVL. In that ecosystem, Lido accounts for about $28 billion in staked assets. But what does Lido do?

Becoming a validator on Ethereum requires a stake of at least 32 ETH (about $100,000 as of May 7, 2024), technical knowledge, and the risk of slashing that comes with it.

This makes solo staking (running your own validator) an unattractive option for most users.

Ethereum does not support native staking delegation. This means you cannot stake your ETH directly with a validator, but instead require an off-protocol service to facilitate delegation. Those with capital but lack the knowledge or intent can delegate node operation to staking-as-a-service products like P2P or Stakefish, which charge a monthly service fee.

                     Lido’s dominance has only grown over time.

Those without 32 ETH of capital rely on platforms like Lido. Users can deposit ETH into Lido’s staking pool in exchange for liquid staking tokens, stETH. The deposit pool is evenly distributed among 39 trusted and vetted node operators. Lido charges a 10% fee on staking rewards, which is evenly distributed between node operators and the Lido DAO treasury.

Here are some numbers to help you understand the scale of Lido:

• 27% of all ETH is staked. Of the staked ETH, nearly 30% is deposited with Lido.

• Lido’s TVL is around $28.7 billion, more than 7x that of the second-largest staking protocol on-chain ($3.71 billion for RocketPool).

• Lido accounts for more than half of Ethereum’s total TVL and nearly a third of total DeFi TVL on all chains.

• 

These numbers raise two questions.

• First, how did Lido become so dominant?

• Secondly, is this dominance healthy for a decentralized network like Ethereum?

The answer to the first question lies in the interplay between liquidity and distribution.

The biggest value proposition of LST is instant liquidity. Users should be able to sell tokens at any time with the lowest possible slippage (best price). Slippage is a function of the size of the liquid pairs of LST with other assets (ETH, stablecoins) on exchanges.

The larger these pairs are, the lower the slippage, and the more widespread the adoption of LST will be.

Lido’s stETH has the highest liquidity among LST. One can buy and sell over $7 million worth of stETH on multiple exchanges with less than 2% price impact (as shown here with the ±2% depth figure). For the second largest LST, rETH, the same metric is less than $600,000.

High liquidity also helps with integration into lending protocols. Users often stake their assets as collateral for loans. It serves two functions. First, they can earn yield on the underlying asset. Second, it provides them with USD liquidity for the staked asset. These USD can then be used to trade or increase leverage by buying more of the underlying asset (ETH or SOL) to stake and increase yield.

But when a user's loan against an asset is liquidated, the protocol needs instant liquidity to prevent the collateral from going bad (becoming undercollateralized). If the liquidity of an LST is low, the likelihood that a lending protocol will accept it as collateral is reduced. stETH is currently the most supplied asset on Aave, the largest lending protocol on Ethereum.

The other half of the interaction is distribution. Users hold LST to earn additional yield or participate in the broader DeFi space. Therefore, the more protocols an LST can use, the more attractive it is to hold. Think of currencies around the world. The more accepted a regional currency is, the more valuable it is.

Lido’s LST (stETH) is like the dollar for staking assets. There is no LST in the Ethereum ecosystem that is as widely accepted as Lido’s stETH.

One ​​can use stETH on the Synthetix perpetual market on Optimism, on the Venus money market on the BNB chain, or on Aave on Arbitrum. EtherFi, a staking protocol with ~4% of all staked ETH, only accepts ETH and stETH deposits. Similarly, even new protocols like Morpheus, a peer-to-peer AI network, only accept stETH deposits.

Liquidity and distribution feed off each other. The more liquid an LST is, the more attractive it is to users. The greater the number of users holding LST, the greater the incentive for protocols to integrate it. This in turn leads to wider adoption, with more users depositing funds into Lido, generating higher liquidity.

These compounding network effects lead to a centralized winner-take-all market structure. Lido is a behemoth because it dominates this market on Ethereum, the chain with the most DeFi activity.

But does this mean that Lido’s dominance is a threat to Ethereum’s decentralized nature? Some, like the author of this article, think it is. As the Lido DAO controls about 30% of staked Ethereum, it may have an outsized influence on the network.

Given that Lido currently only has 39 node operators, there is a risk of operators colluding to conduct activities that are detrimental to the health of the network. They could theoretically conduct transaction censorship and cross-block MEV extraction. If Lido continues to grow and takes half of all staked ETH, they could start censoring entire blocks. At two-thirds of ETH staked, they will be able to finalize all blocks.

LDO holders benefit from 5% of the staking rewards that are retained by the DAO. Therefore, their incentive is to maximize the amount of stake held by Lido and the fees generated by their operators. Any decision they make will serve this goal, rather than the interests of the broader Ethereum ecosystem.

This presents a fundamental principle - agency problem. Lido is making changes to mitigate these risks.

1. First, they are working to increase operators, make it more geographically dispersed, and eventually allow any validator to join without permission.

2. Second, there is a proposal to introduce dual governance to the protocol. Both stETH and Lido holders will have a say in the direction of the project.

However, despite these changes, Lido itself is trending towards a monopoly on Ethereum staking. This brings up the long-term risks we have discussed.

The staking and LST landscape on Solana is very different from Ethereum. Solana’s staking ratio (the percentage of SOL in circulation that is staked) is over 70%, much higher than Ethereum’s 27%. However, LST only accounts for 6% of the staked supply (compared to over 40% on Ethereum).

It is valuable to explore the reasons behind this difference.

• First and foremost, staking on Solana works very differently than Ethereum. Unlike Ethereum, Solana supports delegated proof of stake. This means that users can stake any amount of SOL, with no minimum requirement, and stake directly with validators without using a third-party protocol. In fact, most wallets (including popular ones like Phantom and Backpack) allow staking directly through the wallet interface.

This allows users to easily stake their SOL natively. In contrast, due to the lack of delegated staking on Ethereum, using a staking pool like Lido is the only viable option for most stakers.

• Second, slashing on Solana is not yet active. This means that validator selection is not a critical issue and people can stake with any validator that provides a reward without taking on risk. However, on Ethereum, slashing is active, making it important for validators to choose a staking pool solution like Lido to perform.

This also means that the returns from a staking pool that distributes stake to multiple validators on Solana are not significantly different from the returns from staking directly with one of the top validators.

• Third, the DeFi ecosystem on Solana is not as mature as that on Ethereum. This means that even if users stake their SOL for LST, there are not many protocols to take advantage of. Why take on additional risks, such as smart contracts being hacked, when there is not much opportunity to gain yield? Instead, just take the easy route and stake directly.

• 

The first generation of Solana LSTs — Marinade’s mSOL, Lido’s stSOL, or SolBlaze’s bSOL — mimicked the strategies of liquidity staking protocols on Ethereum. The problem is that the problems that Lido and its peers solve on Ethereum simply do not exist on Solana.

Going back to our discussion of Lido’s dominance on Ethereum, one reason is that stETH is integrated into all the major DeFi protocols and projects in the ecosystem. This does not happen automatically, but requires years of foundational work and building trust and goodwill within the ecosystem. Participants within the Web3 industry will refer to these as business development (BD) efforts.

These networks can’t be easily replicated on new chains just because a protocol has been successful on a competing chain, especially given the tribal nature of crypto.

It’s often assumed that technical standards are adopted purely based on their efficiency. But the underlying adoption is often interpersonal. This is clearly demonstrated by Marinade’s dominance of Solana staking, surpassing Lido.

In the early months of Solana staking, there were two main players: Lido, a crypto unicorn backed by millions in venture capital, and Marinade, a self-funded project born out of a Solana hackathon. However, Lido’s stSOL never surpassed Marinade’s mSOL in TVL.

This is partly because Marinade’s sole focus (and origin) is Solana. Lido, by contrast, is expanding from a network it built itself.

Lately, with the resurgence of Solana, LST is making a comeback, spearheaded by Jito, a protocol we’ve written about before.

Jito is as close to being a Solana-native protocol as possible. Their 2023 airdrop woke Solana from its post-FTX slumber, creating a wealth effect and a resurgence of on-chain activity. Backed by VC and community goodwill, Jito is following Lido’s playbook and looking to dominate LST on Solana.

1. Jito began using its governance token, JTO, to incentivize liquidity for the JitoSOL liquidity pair on the exchange. They have the highest APY, TVL, and volume of all LSTs on the Kamino liquidity pool.

2. Second, Jito is working with other top protocols on Solana such as Solend, Drift, Jupiter, and marginfi to deeply integrate JitoSOL into the ecosystem.

3. Third, it is expanding to Arbitrum through a partnership with Wormhole, thereby achieving multi-chain and increasing the utility and appeal of JitoSOL.

   Jito has the most liquidity in LST on Kamino

4. 

With Solana back in the game, JitoSOL's activity and liquidity have surged, and Jito has perfectly timed the launch and growth of JitoSOL. Not only has it become the dominant LST on Solana, but it’s also the protocol with the highest TVL on-chain.

The Sanctum team are the OGs of Solana’s liquidity staking ecosystem. They first helped Solana create the first stake pool contract, which is now used by almost all LSTs (except mSOL). They also ran a traditional stake pool called scnSOL before creating unstake.it (now Sanctum Reserve).

Sanctum is fundamentally rethinking liquidity staking, with a mission to prevent Solana from going down the path of Lido having the dominant staking protocol, and bring a vision of an ecosystem with unlimited LSTs.

At the core of their product lies a unique insight. Sanctum’s co-founder calls it an open “secret” — LST is fungible. Let me explain what that means.

When you stake SOL to a validator, a staking account is created containing SOL, which is then delegated to the validator. This way, the validator does not have direct access to your SOL. This also means that staking is not instant. Staking accounts can only be activated at the beginning of epochs (and deactivated at the end of epochs).

Each epoch on Solana lasts about 2 days. Similarly, when you deposit SOL into a staking pool like Jito, a staking account is created and the stake is delegated to multiple validators determined by the protocol. In return, you get a liquid staking token. Another way to look at this is that LST is a tokenized version of a staking account.

This means that whether a staking account is created by staking SOL directly to a validator or by depositing SOL into a staking pool, what is behind it is the same - locked SOL. This Solana-unique mechanism is fundamental to Sanctum’s innovation in the liquidity staking space.

Reserves and Routing

Typically, when a user wants to redeem LST, they have two options.

1. They can interact with the issuing protocol, deactivate their staking account, and wait for the cool-down period (2-4 days) to expire to claim their SOL, or

2. They can trade LST-SOL pairs on a DEX to gain instant liquidity.

Since users may hold LST in the first place for the benefit of instant liquidity, they will prefer the second option. This means that LST without liquidity will be inefficient and unattractive to users. This favors big players and makes it difficult for newcomers to create an attractive LST. Liquidity creates liquidity.

Sanctum Reserve changes this equation by offering an entirely new method for redeeming LST. Recall that LST is nothing more than a wrapper around a staking account, which contains locked SOL. This means that LST can always be redeemed for its value in SOL, just not immediately.

Sanctum Reserve is a pool of over 200,000 SOL worth over $30 million. When a user wants to redeem LST, they swap their staking account with the Sanctum Reserve in exchange for instantly liquid SOL. Subsequently, Sanctum deactivates the staking account and receives the paid SOL at the end of the cool-off period.

As a result, SanctumReserve temporarily faces a shortfall in SOL for the duration of the cool-off period, which is eventually recouped. Sanctum charges a dynamic fee based on the percentage of SOL remaining in the reserve pool. This ensures efficient use of SOL during times of high liquidity demand.

The Sanctum Reserve is a significantly more capital-efficient way to liquidate LST than traditional liquidity pools. In traditional pools, the SOL in an LST-SOL pair is deposited by users who would otherwise earn a yield by staking it. Additionally, each LST requires its own liquidity pool, which fragments liquidity across the ecosystem. The Sanctum Reserve frees up SOL in swap pairs for staking by providing a universal pool to liquidate any LST, while unifying liquidity across LSTs — all with minimal slippage. In simple terms, all liquid staking tokens on Solana benefit from Sanctum’s Reserve. But how do they get staking protocols to integrate them? That’s where the Router comes into play.

Sanctum’s second product is the Sanctum Router, developed in partnership with Jupiter. As you might guess from its name, it provides a mechanism to easily and efficiently swap between any two LSTs on Solana. When a user wants to exchange an LST, say JitoSOL for hSOL issued by a Helius validator, here’s what happens behind the scenes:

1. Withdraw JitoSOL from account to a new account

2. Destroy JitoSOL

3. Deposit new account to Helius validator’s account

4. Mint hSOL

5. Transfer minted hSOL to user wallet

All of this happens in a single transaction, and benefits from the insight that different LSTs are backed by interchangeable accounts. Sanctum Router, combined with Jupiter’s routing system, ensures that any LST, regardless of its liquidity, can be exchanged for any other LST.

Router and Reserve have exchanged over 2.2 million SOL to date.

The existence of these two products changes the landscape for small LSTs. They no longer need to rely on deep liquidity pools to attract users to buy LST. Instead, they guarantee holders instant redemption and liquidity, or frictionless swaps with low slippage between any two LSTs. This also makes LST more useful in the DeFi ecosystem. For example, lending protocols can rely on Sanctum Reserve to liquidate loans against any LST.

Significantly lowering the barrier to establishing LST has already sparked a wave of innovation in the Solana liquidity staking space.

Since LST is just a wrapper around a staking account, every validator can have their own LST. But what’s the point? When staking natively, the APY for most validators is more or less the same, which means there is no way for validators to differentiate themselves. When I dove deeper into the world of Solana validators earlier this year, multiple validators told me their biggest challenge was attracting more staking volume.

LST, backed by Sanctum’s Router and Reserve, offers them a way to do just that. Issuing their own token allows stakers to participate in the broader DeFi space and come up with additional ways to reward holders of staked assets.

Laine, one of the top validators on Solana, rewards laineSOL holders with additional block rewards (beyond the APY component), giving holders more than double the native staking yield. Similarly, validator Juicy Stake recently airdropped SOL to all wallets holding at least 1 jucySOL.

Throughout this article, I’ve mentioned that liquidity staking is a tough start for small players. picoSOL, an LST issued by an independent validator in Japan, grew from 0 stake to $8.5 million in less than 30 days by being an active community member and sharing above-average rewards with holders. Recently, picoSOL has been integrated into marginfi, one of the top lending protocols on Solana.

By removing the burden of creating a liquidity pool, LST allows small, newly established, struggling, or ambitious validators to compete with the big players. This makes the Solana validator set more decentralized and competitive. Ultimately, it also provides users with more validator choices without giving up the optionality of liquidity and high APY.

Infrastructure projects, such as Solana RPC providers Helius and Jupiter, have also issued their own LST, but for slightly different reasons.

Solana recently transitioned to a stake-weighted implementation that “allows leaders (block producers) to identify and prioritize transactions proxied by staked validators as an additional Sybil resistance mechanism.” This means that a validator with 0.5% of the stake will be entitled to transmit 0.5% of the packets to the leader.

As an RPC provider, Helius’ main goal is to read and write transactions on-chain as quickly as possible. Given these network changes, the fastest way to do this is for them to run their own validator. Helius validators do not charge commissions and pass all rewards on to their stakers. For them, running a validator is an operational expense, not their core business. With the hSOL LST and the right partnerships, they can more easily attract stake (Helius can also experiment with programs like RPC credit discounts to hSOL holders.)

Jupiter runs a validator for very similar reasons and released JupSOL. The more stake Jupiter’s validators have, the easier it is for them to send successful transactions to the Solana network, allowing user orders to be fulfilled faster. Like Helius, Jupiter passes all fees to stakers.

In fact, to attract more stake, they delegated an additional 100,000 SOL to increase JupSOL’s yield, making it one of the highest-yielding LSTs on Solana. Despite being launched less than a month ago, JupSOL has already attracted over $150M in TVL.

We’re also seeing some experiments with Solana projects issuing their own LSTs.

For example, Cubik, a public money protocol on Solana (similar to Gitcoin), recently released the iceSOL LST with the help of Sanctum. All staking rewards on iceSOL are used entirely to fund public goods on Solana. So for any Solana believer holding native SOL, they can convert it to iceSOL without suffering any monetary loss while supporting public goods on the network.

Pathfinders, an NFT project on Solana, has its own LST, called pathSOL. Not only do pathSOL holders get whitelisted for NFT minting, but the LST will be locked in the NFT forever. If a user wishes to get a refund on their minting price, they can redeem their SOL back by destroying the NFT. Meanwhile, the Pathfinders team earns a yield on all the locked SOL.

Finally, Bonk, one of the top memecoins on Solana, recently launched their own validator and LST, called bonkSOL. What are the benefits of holding? In addition to earning staking returns, holders can also earn $BONK tokens as rewards.

It is conceivable that this trend will continue. For example, Tensor, with many SOL sitting idle in bids, could launch tensorSOL and accept that token as bids as a way for users to earn more yield (or add a layer of gamification by giving away accumulated yields as lottery tickets).

One ​​of the most interesting emerging trends in the Solana LST landscape is the possibility for individuals to issue their own LST.

One ​​of the early proofs is fpSOL, issued by Sanctum founder FP Lee. Those holding at least 1 fpSOL gain access to a private chat group with FP Lee (similar to a Friend.tech key), while staking rewards go to charity.

It's not hard to imagine this becoming more common, with influential individuals issuing LST as a safer option for their followers than NFTs or memecoins. They could get distribution through social media (as in the PicoSOL example, it didn't take long to attract staking), offer exclusive benefits to holders, and make money by keeping some or all of the proceeds.

Sanctum Infinity

Sanctum’s third product is Sanctum Infinity. It is a multi-LST liquidity pool that supports swaps between all LST in the pool. The team claims that Infinity has the most capital-efficient automated market maker (AMM) design possible. Let’s see how it works.

Whenever you want to buy 1 SOL worth of LST, you will always get less than 1 unit of LST. This is because LST accumulates staking rewards during its lifetime, and these rewards are reflected in its price relative to SOL. As of May 8, JitoSOL is valued at $162, while SOL is trading at $146. The JitoSOL/SOL ratio is 1.109, which means that since JitoSOL was launched, it has provided a return of about 11% to SOL. This ratio will continue to increase over time.

Each LST has a staking pool account with two parameters: poolTokenSupply (total deposited SOL) and totalLamports (deposited SOL + accumulated rewards). Lamports are to Solana what sats are to Bitcoin — the smallest unit of measurement. Dividing these two parameters gives us the staking ratio.

How Solana Stores Staking Pool Information

Sanctum Infinity uses this in-protocol information as an absolutely reliable on-chain oracle to provide perfect pricing data for every LST in the pool. Traditional AMMs rely on the ratio of the asset pairs in their pool for pricing. This can be inefficient if liquidity is low or there is a temporary imbalance caused by a large trade. The staking pool account information allows the Infinity AMM to perfectly price every LST, regardless of its liquidity.

Infinity Pool is currently a basket of LSTs that are permitted to be exchanged at the discretion of Sanctum. Users can deposit permitted LSTs into the pool in exchange for INF tokens. INF accumulates staking rewards for all deposited LSTs as well as transaction fees for swaps made inside the AMM. Therefore, INF is an LST in itself, but with additional sources of yield.

Infinity maintains the target allocation by dynamically adjusting the swap fees from one LST to another until the target ratio is reached. The fee for each LST is divided into two parts: the input fee, which is the fee paid when swapping out LSTs; and the output fee, which is the fee paid when swapping in LSTs. The total fee is the sum of the two.

If I want to swap JitoSOL (0.02% input fee, 0.03% output fee) for JupSOL (0.04% input fee, 0.05% output fee), I will have to pay JitoSOL’s input fee plus JupSOL’s output fee, which totals 0.07%. By dynamically adjusting LST’s input and output fees, Sanctum maintains the target allocation of the AMM pool.

Since launching last year, Sanctum’s product TVL has grown to over $500 million, making it the fifth largest protocol on Solana.

The term “democratization” is often used in tech circles to describe how a process that was previously difficult to enter has become accessible to people who historically could not access it. Sanctum has largely democratized liquidity staking by democratizing it. The natural extension of this argument is often that Solana’s staking ecosystem is more innovative than Ethereum’s. I think there is more nuance here.

When Lido was growing, DeFi was still a nascent sector, and Ethereum itself, at the time, was transitioning from PoW (Proof of Work) to PoS (Proof of Stake). There were too many variables and too few precedents. In contrast, Solana’s staking ecosystem was built years after Lido’s existence. As we’ve seen, Solana’s developers did try to copy Ethereum’s playbook. So, it’s safe to say they took inspiration from it.

But copying doesn’t provide any competitive advantage. The story we’ve seen between Lido, Jito, and Sanctum in this article is a story of how an existing player (from Ethereum) competed with and was surpassed by a smaller, more nimble, and more localized protocol player. Can Sanctum’s advantage on Solana last? We don’t know yet. As with most innovation cycles, new players will emerge to compete with Sanctum’s positioning in staking.

But one thing is clear: between Sanctum’s reserves ($30M worth of SOL) and their router (which is integrated into Jupiter), Sanctum is growing into something more than “just another staking provider.” That has value.