In our last post the main problems of the Proof of Work (PoW) consensus algorithm were discussed. The alternative consensus algorithm Proof of Stake (PoS) was touted as the solution to exorbitant energy inefficiencies and centralization tendencies. In this post we will explore PoS in more detail and discuss potential problems of the protocol. Assuming readers are familiar with PoW deficiencies, we can go on to discuss PoS characteristics. Contrary to the Bitcoin PoW, there is no planned token mining in the main Proof of Stake projects. Instead of mining tokens, users are expected to forge blocks. The frequency with which forgers get to forge a block depends on the token value staked or in other words locked up. Once their tokens are locked, forgers are called to bet on a block and thus participate in the verification of transactions.
Reliance on Hardware
To begin with, given that PoS does not rely on mining hardware for heavy computation, PoS protocols are a lot more flexible in their mechanism design. Whereas in PoW, mining hardware will not get affected by the protocol no matter what the behavior of the node, Proof of Stake can be designed in such a way that tokens can be slashed given certain conditions. In PoW, if the participating node gets banned from the network, then they can still sell the hardware to another person or legal entity. They may also redirect their computational power to another protocol such as Ethereum or Verge.
Governance flexibility in Proof of Stake
Another governance flexibility that comes with the transition from PoW to PoS is that centralization due to economics of scale can creatively be countered. As we argued in our last post, it is economically more rewarding for small miners to organize in gigantic mining pools and give up control over their tokens. This encourages centralization of the network. In Proof of Stake, rules of governance can be created such that centralization is discouraged. This is enabled by a higher degree of independence on hardware in PoS systems. For this reason Proof of Stake has often been judged as the consensus algorithm less likely to lead to the centralization of the network. Centralization of miners in the Bitcoin mining network. https://blockchain.info/de/pools retrieved 18.01.2018 Centralization of miners in the Bitcoin mining network However, depending on the characteristic of a PoS algorithm, Proof of Stake may also lead to a centralized blockchain ecosystem. A centralized blockchain ecosystem is one where control of the network rests with a single or few participants instead of being distributed across the network.
The Monopoly effect
Forgers on the PoS blockchain typically receive rewards proportional to their staked value. This means that already quite influential participants on the blockchain will become even more influential over time. Because each token has the same chance of being picked as the next validator, the odds to get chosen as the next validator rises with the amount of token one holds. Whereas PoW works to the benefit of those with the most hashing power, PoS designs the game in favor of those with the most tokens available for staking. Just like Bitcoin mining adheres to the principle of economies of scale, in Proof of Stake the profit margins of high stakers are significantly higher. Regardless of the amount of tokens staked, the wallets of participants have to be online. To be online, participants need resources including hardware, electricity, and an internet connection. In a way, there are fixed costs associated with forging blocks on a PoS consensus algorithm. Consequently, the forger who stake 1 ETH and the forger who stakes 3 ETH both have to pay the same amount of fixed cost to stake. This entails that the higher the stake of the forger the higher their profit margins. Additionally, the staking algorithm will reward those that are most active on the protocol. One might find this a desirable property of the blockchain, because it encourages participation, but a similarly unequal distribution of influence and wealth may result as in PoW. If participation is rewarded in token and more token creates disproportionate opportunity to influence the system, then unequal influence over the system may arise in Proof of Stake blockchains by default. Without a correction in the protocol a larger ETH stakeholder will grow their stake faster than a small ETH stakeholder. After some time the relative cost for some forgers to stay in the network will be too high and they will be forced off the network.
Pool of Stake solutions
So is centralization on Proof of Stake blockchains a problem that is impossible to avoid? At the beginning it was stated that one of the main benefits of PoS blockchains is the flexibility due to their reliance on code instead of hardware. So one possible way to deter centralization on a PoS blockchain is to program the blockchain in such a way that it encourages staking by small token holders. In such a model the returns to staked tokens would diminish with increasing stake. What could happen as a result are so-called Sybil attacks on the blockchain, namely the creation of many small forging identities by one person or legal entity. This way, centralized control over the network could resurface. Whether or not such Sybil attacks will be feasible depends on how efficiently identity will be verified on blockchains in the future. Preventing Sybil attacks will be vital to the goal of establishing decentralized PoS systems.
Introducing Pool of Stake
Another approach to guarantee the decentralized nature of Proof of Stake consensus algorithms is that smaller stake owners pool their stakes to lower their fixed costs on hardware, electricity and internet connection. The pooling of stakes, however, creates a vulnerability outside regulation and more importantly hacking. As a consequence, the creation of safe pools for staking tokens will become vital for enabling smaller stake holders to participate in the forging process. For this it will be vital to enable ultimate control over the tokens by the small token stakers, because handing over control of the tokens to the pool would mean further centralization. At Pool of Stake we build a staking pool which leaves ultimate control over the tokens with the small token owner. Stay tuned for our coming articles where we will reveal more of our pooling design such that small forgers may benefit from upcoming developments in the blockchain ecosystem.