In order for a blockchain network using PoS consensus algorithm to maintain consistent throughput of transactions there are three key factors that a validator node must actively manage: uptime, block creation, and integrity. If any of these three factors are violated, the network’s overall quality as it pertains to security and throughput are placed in jeopardy. Slashing is the protocol “lever” that incentivizes nodes to have excellent uptime, throughput, and non-malicious contributions to the underlying network. PoS consensus is contingent upon token ownership and staking.
As such, slashing is the on-chain mechanism that leverages this fact by publicly destroying a portion of the total staked tokens of a validator node that is not performing up to par with the networks standards as it relates to the three key factors previously mentioned.
Slashing as described in the original Casper research paper (written by Vitalik and Virgil) follows the so-called “Casper Commandments”. The protocol programmatically scans validator votes for violations such as voting twice or voting in an abusive rapid succession. The protocol, as a result of a series of proven theorems, is able to maintain the attributes of accountable safety and plausible liveness; both of which are sub-attributes of integrity. Uptime is trivial to manage — if a validator misses more than ⅓ of the blocks of an epoch, the node is slashed. Finally, block creation and BLS key signing are checked to see if less than ⅔ of all proposed blocks are signed. If so, then the node’s stake is slashed.
It should be noted that there is a range of blockchain networks that have different slashing parameters that must be achieved in order for a node to be slashed. The more intensive the slashing parameters are for a protocol, the harder it is for non-industry grade nodes to run successfully. This is because delegators will choose to stake with validators that have healthy and reputable metrics, and more than likely stable hardware set-ups. The advantages of this higher standard is that the average node of the network has a higher quality.
Unfortunately, this can potentially slow down the level of decentralization the network can achieve. This phenomena can be attributed to the costs of hardware and CRM being greater than the potential profitability from garnering new delegators to a node. This effect is markedly noticeable in networks that have demanding base line specs for a node to be able to participate in the network. A prime example of this is the Secret Network where SGX (a key component of the baseline protocol architecture) is a mandatory requirement for a full node, drastically increasing the cost barrier of entry.
This effect is also true when the rate of slashing is increased or decreased. If 5% of a node’s total stake is slashed for certain behaviors versus a protocol with a .01% slashing penalty, there is a guarantee of modified delegator behavior — particularly those with large stakes. In the long term, governance will likely be pushed in a direction to have more demanding parameters as well as higher slashing rates in order to have the best possible network throughput and vitality. Validators with off-chain contracts, excellent CRM, professional-grade hardware/management will all be favored in the long-run by entities in an ecosystem that want a guarantee of a steady return without any slashing penalties incurred on their stake. Smaller validators may counteract networks that push for demanding slashing parameters by offering up free kickback on a periodic basis for any slashing that occurs against a node. While this strategy may be comforting for some delegators, it relies on trusting an off-chain mechanism of CRM to make-up for lack of quality control as it pertains to the three variables mentioned earlier — all of this risk can in principle be sidestepped simply by staking with a more reputable validator.
In theory, as a network’s token value increases in both price and velocity, the more important the quality of the average validator becomes. While slashing may not be an immediate problem for the average small scale delegator, it is a massive barrier of entry for institutional investors that want a stable return on staking. There is large growth potential in the near future for slashing insurance that covers either validators or institutional stakers. High quality validators will have to face a subtle trade-off: pushing for more intensive parameters through governance may draw more delegators away from smaller validators, and potentially decrease the total competition and decentralization of the network. Simultaneously, a higher slashing rate offers up a greater barrier of entry for institutional and large delegators.
Blockchain networks that strike a balance between quality of validator, decentralization, and security will win out in the long-run. How a validator brands themselves may vary from network to network based on the priority of a given blockchain’s architecture and community ethos. As usual, adaptability, excellent CRM, active governance participation, and network contributions ultimately decide how delegators choose to navigate the risk of having a portion of their stake slashed.
Foundation, Ethereum. “Validated, Staking on eth2: #1 — Incentives.” Ethereum Foundation Blog, blog.ethereum.org/2020/01/13/validated-staking-on-eth2–1-incentives/.
Tsapis, Dimitris, et al. “PoW vs PoS: Blockchain Consensus Mechanisms Explained.” Paybis Blog, 16 June 2020, paybis.com/blog/pow-vs-pos/.
V Buterin, V Griffith — arXiv preprint arXiv:1710.09437
“Slashing.” Harmony, docs.harmony.one/home/validators/definitions/others.
Boneh, Dan, et al. “BLS Signature Aggregation.” BLS Multi-Signatures With Public-Key Aggregation, Mar. 2018, crypto.stanford.edu/~dabo/pubs/papers/BLSmultisig.html.
“SGX Compliance Secret Network.” Secret Network Wiki, learn.scrt.network/sgx.html.
Author: Carter L. Woetzel