March 12, 2021 · 7 min read
Slashing Penalties - The Long Term Evolution of Proof of Stake (POS)
The increasing use of Proof-of-Stake (PoS), especially in the current DeFi era has provided a way for HODLers to earn rewards for the tokens they are holding. 2020 saw the two new words becoming a rising trend of the cryptocurrency industry - ‘Liquidity Mining’ and ‘Yield Farming’. Every blockchain has a network of participants which run nodes on the blockchain and work towards reaching the consensus on the most current state of the blockchain. Proof of Work (PoW) has miners who consume energy to mine blocks, PoS blockchains have validators. So rather than hardware-warrior miners competing to solve complex mathematical problems thus verifying new transactions (with many hundreds of thousands of computers all performing the same calculations using energy and creating CO2 emissions), in PoS, validators stake their tokens to validate blocks which are to be added to the blockchain. By staking tokens in a wallet, these validators make themselves available to propose new blocks and validate the blocks proposed by other validators. Once a block receives ‘enough’ validations, the block is added to the blockchain. Most of the PoS blockchains require 2/3rds of the network majority to create the block. Validators are rewarded for both attesting and proposing the block to the blockchain as a percentage of their stake.
What happens to the attackers in PoS? - Slashing Penalties Explained
In PoW, if an attempt to mess with the network fails, you lose the cost of electricity which can be quite vague to calculate. Cryptoeconomics of PoS blockchain works two ways. The majority of the PoS blockchains have a reward mechanism for honest and truthful behavior as well as a penalty mechanism for malicious behavior. Validators in PoS blockchains have to deposit their tokens as ‘stakes’, which also acts as collateral. This collateral is locked/freezed and can’t be moved for the locked period. The penalty in the PoS blockchain charged for inactivity, dishonest validations, or any other malicious behavior is called a Slashing Penalty. The penalty may vary from being charged a fixed amount of tokens, a fixed percentage, complete slashing of the stake and banning the validator from the group for the current epoch (or permanently).
What is the purpose of slashing penalties?
Slashing discourages validator misbehaviour, which in turn promotes security, availability of the validators, and honest network participation. The two reasons for the slashing penalty are 1) to make it expensive to attack the network and 2) to make validators behave responsibly. The two major cases when the validator is charged are when the validator shows ‘downtime’ (absent to sign transactions) and double signing (signing two or more blocks at the same height). Many participants run backup nodes to keep the network running 24x7 but an incident in 2019 on the Cosmos blockchain took place where two parallel blocks were proposed and signed, and the validator was slashed from the network. Double signing is a threat to network security and slashing penalties protect the network from this threat. The slashing penalty also gives groups a further incentive to examine validators, attract, and retain the best validators and accept them as group members.
Impact and the Evolution
Over the past, slashes have happened due to incorrect algorithms, DDoS attacks, fraudulent attestations, and many similar reasons which have resulted in millions of dollars being lost by the validators for their misdoings. These incidents create a fear of slashing among the validators. Slashing has evolved with new protocols to increase this fear and keep the network more stable and blockchain. The various factors which make the slashing penalties vary across multiple blockchain are - type of misbehaviour, severity of misbehaviour, type of penalty (fixed or percentage), duration of banishment or non-engagement from validation activities, rewards and stake for the whistleblowers (if any). Then the question remains what to do with the penalized funds?
What happens to the penalized funds?
In Casper’s FFG, 4% of the funds are given to the node which reports the behaviour and the remaining 96% are burned. Ethereum 2.0 brought newer changes by keeping the minimum penalty of 1 ETH which increases with the number of people being slashed to create more fear of misbehaviour. For major mistakes, the minimum penalty of (effective balance staked )/ 32 is charged at the beginning, and then more penalties follow during each epoch. Here we see Ethereum evolving its slashing model from a fixed and flat model to a more dynamic and multilayered structure where the growth in the attack or growth in the number of attackers is resulting in the growth in the penalty charged.
A similar approach is deployed by Polkadot’s NPoS which you can see in Table 1 below. As the number of offenders (x) compared to the total number of validators (n) in the network increases, the penalty charged for the incident also increases. The majority of the PoS protocols want two third of the total validators to be active and honest during each block addition. Both Ethereum and Polkadot have designed their mechanisms to punish the validators if more than 33% of the network is inactive.
Table 1 - Polkadot NPoS slashing penalty in double signing (Grandpa/Babe Equivocation)
Celo has a fixed penalty in Celo tokens. When the slashing sum is deducted from the validator’s stake, the same stake is also deducted from the group the validator was part of. In this kind of system, the bigger validators are at advantage and can manage to perform more misconduct. To tackle this flaw, Celo removes the validator from the group leaving the decision to the group whether they want to take the charged validator back to the group. All the slashed penalties go to the community fund.
Cosmos, on the other hand, charges a fixed percentage of 0.01% and 5% respectively with a jail time which the validator has to serve. Keeping a percentage penalty maintains the equality in the system and ensures that the bigger validators realize their responsibility. We compare different PoS blockchains and their slashing penalties in Table 2 below.
Table 2 - PoS blockchains and their slashing penalties compared
We can notice the variety of penalizing models which exist in the blockchain space, which have also resulted in evolving PoS space as a whole. It is impressive to see that the developer teams are analyzing the behavior of the validators and devising mechanisms which are optimizing the above-mentioned varying parameters. Alongside these, we have PoS protocols like Algorand, Cardano, and Hashgraph which do not have any slashing penalties. Having no slashing allows the validator to validate blocks on multiple chains with the same node or on the wrong chain. Many experts in the PoS ecosystem don’t consider PoS blockchains without slashing penalties secure compared to those that do.
PoS is heavily supporting the DeFi lending and Liquidity Mining models which are making the returns from both a lucrative deal. Slashing penalties reduce the ‘total reward’ and demand validators to be more responsible. Since two-third of all staked assets need to be owned by honest actors to ensure the security of the network, and if more people abandon the network, the security is at risk. Abandoning and high APY on lending might result in few owners having large amounts of stakes/nodes which in turn undermines the security again. PoS protocols have to consider at least thrice about reducing block rewards because doing so would encourage stakers to switch to lending. Traders may short the security due to diminishing security which may push pressure on the prices. PoS are based on inflationary models. Developers need to rethink the tokenomic models on reward and security, keep the cap on interest rates, and help secure the network by rethinking the slashing penalties and rewards for their validators.