Cardano's Liquid Staking from a Security Perspective

Cardano has native liquid staking. Ouroboros PoS does not require locking coins for a certain period of time in order to allow coins to be confiscated (slashed) for staker behavior that does not conform to the needs of the protocol. In this article, we will explain how this is possible. We will also think about the advantages of slashing and whether it makes Ethereum more secure.

Liquid Staking

Cardano has native liquid staking. This means that ADA coins are liquid during staking, so they can be spent. Cardano adds another important feature to liquid staking, namely that it does not define any minimum amount of coins for staking.

The Cardano protocol does not lock ADA coins, does not slash (slash) coins for dishonest behavior, and does not require any minimum number of coins to start staking. The goal of these properties is to align the interests of users with those of the protocol.

The decentralization and security of the Cardano protocol are based on the distribution of ADA coins among users and their market value. Cardano has very simple staking, one can say almost without obstacles, as this increases its decentralization. What do users want? In the context of decentralization, they want to have control over their assets (do without third parties) and the ability to spend assets at any time (liquidity).

The reason why Cardano does not need slashing (a mechanism used by some blockchain networks to punish malicious actors) is due to the scientific work behind its Ouroboros Proof-of-Stake consensus. The team was inspired by the Nash equilibrium theory and incorporated it into the reward mechanism.

The protocol incentivizes rational behavior of participants but it cannot fully prevent irrational one. However, an irrational participant would need to hold more than 50% of ADA coins to have dominance in the network consensus.

Some PoS projects, such as Ethereum, lock coins in order to punish irrational behavior, i.e. behavior that goes against the needs of the protocol. Ethereum validators need to lock up 32 ETH on the blockchain to become a validator. This staked ETH represents their stake in the network and is used as collateral to ensure they act honestly.

The locked ETH can be at risk of being slashed, which is a penalty enforced at the protocol level associated with a network or validator failure. Slashing is designed to discourage bad actors on the blockchain. ETH can be slashed if the validator proposes a conflicting block, attests two blocks with the same target in the same epoch, etc.

From my point of view, the Ethereum team put the needs of the network too far ahead of the needs of the users. Moreover, over time it has become clear that decentralization suffers.

Users who do not have 32 ETH were forced to use third-party services. In doing so, they essentially surrendered their ETH for other tokens with roughly the same market value. But so do those who have enough ETH to run a validator but don't have the interest or technical skills to embark on the adventure of running a validator.

For any decentralized network, it is generally true that as decentralization decreases, so does security. With the increasing number of single-points of failures, the risk associated with abuse of power or an error that will affect a large amount of staked coins also increases. Accumulating a large amount of stake coins in several places represents a security risk.

Cardano has a native (protocol level) mechanism for delegating any amount of ADA to pools. This divides stakeholders into those who want to run pools (block-producing nodes) and those who only want to delegate ADA. Importantly, delegators can always have ADA coins in their own wallets and are constantly active consensus actors.

In the Ethereum network, liquid staking is currently achieved through third parties such as Lido. Lido will allow people to exchange ETH for stETH tokens. Lido will then use ETH for staking. The Lido DAO is responsible for selecting the node operators.

stETH are liquid tokens that can be spent at any time or used in DeFi. The problem is that the previous owners of ETH are only (largely) passive actors of the consensus and do not have the network in their hands as firmly as in the case of Cardano. ADA stakers can delegate to another pool at any time, thereby actively participating in the quality of block production. Stakers receive regular rewards directly from the protocol.

Native liquid staking has the advantage of eliminating third parties and leaving control of the network to the ADA owners. However, Cardano cannot prevent irrational behavior because it cannot punish it by confiscating coins. So the question is whether it is necessary.

Nash Equilibrium

The Nash equilibrium is a key concept in game theory that provides a solution for non-cooperative games involving two or more players. In the context of Cardano’s reward scheme and consensus, the Nash equilibrium is used to analyze the behavior of stakeholders in the network.

In a Nash equilibrium, each player is assumed to know the strategies of all other players, and no player has anything to gain by changing only their own strategy. If there is a set of strategies with the property that no player can benefit by changing their strategy while the other players keep their strategies unchanged, then that set of strategies and the corresponding payoffs constitute a Nash equilibrium.

Nash equilibrium theory is applied to Cardano's reward scheme. It can be said that the system reaches a Nash equilibrium when all stakeholders have settled on their strategies. This means that all pool operators and delegates have chosen the best strategy for them to participate in the network consensus and have no incentive to deviate from it. Note that a large number of independent stakeholders participate in the equilibrium state. It helps ensure that the system is fair and incentivizes rational behavior.

In the Cardano ecosystem, there are a large number of diverse stakers with different numbers of ADA coins. Despite the high diversity, it can be assumed that the majority will choose a very similar strategy.

Let's consider two stakeholders who can choose to either stake their ADA or not. The rewards for staking are higher than not staking, assuming all actors behave honestly.

If both stakeholders stake their ADA, they both get rewards. If one stakes and the other doesn't, the one who stakes gets rewards while the other gets nothing. If neither stakes, neither gets rewards.

The Nash equilibrium here is for both stakeholders to stake their ADA. This is because staking yields higher returns than not staking, so no participant can gain by changing their strategy unilaterally.

In other words, delegating ADA to the pool (ie stake ADA) is a winning strategy for all stakers. The strategy does not only concern whether to stake or not to stake (the choice is obvious) but also the choice of pools.

Delegators must choose a pool that provides them with the highest possible reward. It is the only such pool that regularly produces blocks in each epoch. For this, the pool must have a certain stake. In addition, the pool must not be over-saturated, as this reduces the rewards.

Consider two operators who can choose to either act honestly or dishonestly. If both act honestly, they both get rewards. If one acts dishonestly and the other honestly, the honest operator gets rewards while the dishonest operator gets nothing. Cardano does not reward dishonest behavior. If both act dishonestly, neither gets rewards.

The Nash equilibrium in this case is for both operators to act honestly. Acting honestly yields rewards.

Operators are interested in having their pools saturated, so they want to attract a large number of delegates (up to the point of saturation). This essentially forces them to behave honestly, as only this behavior will attract delegators and ensure a reward not only for them but also for stakers. The interests of pool operators and stakers are aligned.

It can be assumed that ADA holders will behave rationally in that they want to get the staking reward. The best strategy for a delegator is to choose a pool that is efficient and honest (no slot is missed, etc.). Pool operators have costs associated with running pools, so it can be assumed that they will behave rationally in order to get a reward (and cover the costs). They will try to produce as many blocks as possible to get a reward and attract as many delegates as possible. The result is that a large part of the stake will represent honest stakeholders.

Cardano does not have slashing, so attackers do not lose their staked ADA if they act maliciously. The only penalty is that they don't get the potential reward. The security of the Cardano network is based on the assumption that the majority of stakeholders will behave honestly because they are incentivized to do so by the reward system. This is a common assumption in many blockchain protocols and I believe it is a reasonable one, given that stakeholders have a financial interest in the health and success of the network.

Is Cardano Secure?

It's important to note that while Cardano economically incentivizes rational behavior, it cannot fully prevent irrational behavior. An irrational participant would need to hold more than 50% of ADA coins to have dominance in the network consensus. One of the security assumptions is that it is almost impossible for an individual to acquire such a large amount of coins. Moreover, with increasing adoption and the number of stakers, this can become an increasingly complex task.

The fact is that if someone were to successfully commit a 51% attack, Cardano has no defense mechanism to confiscate the attacker of ADA coins.

Ethereum's slashing mechanism is a key part of its security model, particularly in protecting against the 51% attack (in the case of Ethereum it is only a bit above 33.3%).

For an attacker to successfully carry out an attack, they would need to control more than 33.3% of the network's staking power. However, as they continue to act maliciously, their staked ETH would be slashed, reducing their staking power. Over time, this would drain their resources and make it increasingly difficult for them to maintain control over the network. This serves as a strong disincentive for any attackers.

Security is a complex topic. We will only focus on one part and that is to compare liquid staking with the ability of slash coin. If you wanted to compare the security of two blockchains, you would have to do a comprehensive analysis of all aspects.

From my point of view, the ability to punish attackers is advantageous. However, this advantage will be manifested only in the case of an attack whose practical realization is in itself very unlikely. However, it cannot be ruled out.

While Ethereum can punish attackers and drain them economically, ETH is not liquid during staking. This forces users to hand over ETH to third parties who stake on their behalf. This essentially reduces decentralization, thus also security. Because of an attack that is unlikely, centralization occurs in the Ethereum ecosystem.

Paradoxically, the attack becomes more real along with the unwanted strong influence on the consensus by third parties precisely because of the high accumulation of ETH by several entities. In other words, illiquid staking reduces security. If the existence of a defense mechanism reduces decentralization and security, the design can be considered suboptimally designed.

Lido stakes 32% of ETH. 3 largest centralized exchanges more than 20% of ETH. If one entity controls a large portion of staked ETH, it could theoretically have a disproportionate influence on the network. This goes against the principle of decentralization, which is a key feature of blockchain networks.

There is no such high centralization of stake in the Cardano ecosystem. The largest multi-pool operator in the Cardano ecosystem has a share of 7.4%. There are only two centralized exchanges in the top 10 with a total share of 6.5%.

Liquidity staking supports decentralization and thereby increases security. The high distribution of ADA coins in the hands of a large number of stakers and pool operators is an excellent protection against attacks, including a 51% attack. It is always easier to attack a center of power (be it a smart contract or a centralized exchange) than a large number of stakeholders.

Moreover, decentralization of stakes is important for reasons other than just attack. Decentralization is important from the point of view of resistance to various forms of errors, such as bugs in a smart contract, abuse of power by the exchange's CEO, etc.

Ethereum has the advantage that validators risk slashing if, for example, they propose conflicting blocks or if they are inactive. Cardano cannot punish inactivity or the event when the slot leader appends one block for two existing chains in the case of a fork.

An attack can be carried out when the attacker sets up his own pool, fills it with his own ADA coins, and deliberately does not mint blocks. Fewer blocks will be minted in Cardano, which will affect overall performance. The attacker will not be penalized for this attack and can continue it forever (but must renew the certificate regularly).

However, as far as I know, this type of attack has not yet occurred in the Cardano ecosystem. Cardano has had PoS with staking for the third year in a row, and so far stakeholders seem to be behaving rationally. An operator of a saturated pool has these two options. An operator can mint blocks and get roughly 1000 ADA coins every 5 days (depending on fees), or not mint blocks and get no rewards.

If ever Cardano were to have some form of slashing, it should only apply to the operator's ADA coins. ADA coins of delegators should remain liquid.

From my point of view, Cardano is secured, which is largely supported by liquid staking. Liquid staking enables the growth of decentralization. Small stakers, as well as whales, can hold ADA in their Cardano wallets and delegate coins to any pool. A rational choice of stakers is a pool whose operator also behaves rationally. In case of problems, delegators can always choose another pool.

Ethereum is also a secure blockchain, although it seems more centralized to me. Vitalik Buterin, the team, the Ethereum Foundation, and some leading members of the community perceive Lido (and other similar actors) as a threat and intend to address this issue. From the last paper by Vitalik, I have the feeling that he was inspired by Cardano and wants ETH holders to be able to choose a node operator. This is certainly a step in the right direction.

Conclusion

Liquid staking has one more advantage. If cryptocurrencies are to be money in the future, their properties must match. Besides the ability to be a good store of value, it is also the ability to be a medium of exchange. ADA coins can be money because they retain liquidity during staking. ADA holders have security networks in their hands, but at the same time money that they can spend (and get back). This is not the case with Ethereum. If an ETH holder stakes with a centralized exchange, they cannot spend the coins. If they stake with Lido, they get stETH tokens for ETH. Can stETH (and similar tokens) be money? Hard to say. stETH may have a similar market value to ETH, but they have no meaning from a consensus perspective (except that ETH can be reclaimed). I dare say that liquid staking is a necessity from this point of view.