IOHK | Paper

Library > Incentives Against Power Grabs or How to Engineer the Revolution in a Pooled Proof of Stake System

Incentives Against Power Grabs or How to Engineer the Revolution in a Pooled Proof of Stake System

August/2021, DAPPS '21


Proof-of-Stake (PoS) blockchain systems, especially those that allow stakeholders to organize themselves in "stake-pools", have emerged as a compelling paradigm for the deployment of large scale distributed ledgers. A stake-pool operates a node that engages in the PoS protocol and potentially represents a large number of smaller stakeholders. While such pooled PoS operation is attractive from various angles, it also exhibits a significant shortcoming that, so far and to the best of our knowledge, has not been sufficiently understood or investigated. Pooled PoS operation, to be effective and not lead to sub-optimal dictatorial or cartel-like configurations, should enable the stakeholders to revoke and re-delegate their stake in a way that is aligned with their incentives. However, given that stake-pool operators are exactly those entities who determine what transactions are to be recorded in the ledger, they are quite likely to form a cartel and censor any transaction they want, such as those that attempt to adjust the current stake-pool lineup. In this way, a power grab takes place, where the stake-pool cartel perpetuates its control over the PoS system. We first model and observe formally the emergence of the above problem in pooled PoS systems, and then we describe an anti-censorship mechanism that takes advantage of the underlying cryptographic functions of the ledger and the nature of peer-to-peer networks to diffuse information without suppression. We provide a thorough game-theoretic analysis of this mechanism discovering various types of Nash equilibria which demonstrate that the "revolution", i.e., the strategic decision of pool members to withdraw support from a censoring cartel as well as the pool operators to step down, can be incentivized, under suitable and plausible conditions in the utility functions of the involved participants.