What Works And What Doesn't: The Emerging Battle Between Proof-of-Work And Proof-of-Stake

If you are a crypto enthusiast, you must have heard of proof-of-work and proof-of-stake which are the two consensus mechanisms used to validate crypto transactions and indeed make new cryptocurrencies. All decentralized blockchain networks require validators to prove their existence on the network, and these two methods are the ways that a validator can prove their presence.

The first consensus mechanism to ever be used by a blockchain was proof-of-work, and it was on the Bitcoin blockchain. However, two years later, Peercoin introduced a blockchain that made use of Proof of Stake. At that point, it was a novel idea, and many people, especially Bitcoin maximalists, expected it to fail. Interestingly, it has only grown in popularity since then.

Today, proof-of-stake as a consensus mechanism is so popular that even networks like Ethereum are switching to it. And it’s not hard to see why. The mechanism does have some pretty amazing benefits that make it even more attractive than proof-of-work. But is that the whole story? Is proof-of-stake just a better mechanism than proof-of-work? Or is there something else we need to know?

Proof Of Work

It is easy to think of proof-of-work solely in terms of blockchain consensus, but it is not. It is a pretty broad concept that can be applied to any significantly complex computational system. In essence, it is a form of proof where one party proves to others, or a system, that a certain level of computational effort has been expended. Ergo, it is a way of proving that work has been done.

In crypto, proof-of-work mechanisms take the form of difficult mathematical problems that can only be solved by computational power. In many cases, the mathematical problems start pretty easy, but over time it gets harder and harder.

While it takes a lot of work for the prover to prove that work has been done, it often takes very minimal effort for the verifiers of the system to confirm the work. The idea of proof-of-work goes as far back as 1993 when it was invented by Moni Naor and Cynthia Dwork as a way to stop denial-of-service attacks and spam. The name was coined much later in 1999 by Markus Jakobsson.

Proof-of-work is a very useful way of verifying transactions because it requires verifiers to have skin in the game. By making it costly to mint a coin or validate a transaction, it ensures that only parties who care about the system can have a part to play. However, it is vital to understand that the point of proof-of-work algorithms is not to prove that certain puzzles were solved. That is merely the means. The end of these algorithms is to discourage the manipulation of data by making it infeasible due to high energy requirements.

Proof Of Stake

Unlike proof-of-work which has its history outside blockchains, proof-of-stake is more grounded in blockchain technology. Like proof-of-work, proof-of-stake is also a way of verifying transactions on the blockchain. But instead of asking for proof of computational work done, proof-of-stake systems ask for proof of tokens staked. That is, validators can only work on the chain by proving that they have some quantity of the chain’s native token staked.

This ensures that potential attackers of the chain must have a huge fraction of the tokens to make any material impact. Of course, this means that potential hackers would have to take significant losses if they were ever to validate false transactions. It also means that validators have an incentive to hold tokens, thereby creating a robust reserve for the blockchain and stabilizing prices.

Proof-of-Work vs Proof-of-Stake

Both consensus mechanisms have their benefits, and they also have their disadvantages. Today, a lot of blockchains make use of proof-of-work, but that might just be because the mechanism was the first popular and tested one. The fact that the protocol itself has existed since the 90s makes it, at least historically, more reliable than proof-of-stake.

One thing that proof-of-work has over proof-of-stake is that it is a lot more decentralized. For crypto maximalists, the more decentralized a system is, the more secure it is and the better for it. Since proof-of-work requires more computers and systems across a network to review a block, it also requires a lot more energy usage. The fact that the system runs on heavy computing means that they are very expensive to run and can be bad for the environment. The fact that these networks are also so large means they are slower and may be unable to scale for dApps that require a lot of speed.

Proof-of-stake is almost the antithesis of proof-of-work in terms of its advantages. For one, proof-of-stake does not require a lot of energy since the system does not use up computational power. This also means it is better for the environment in the long run as it uses less power. Since it is not necessarily a large network, proof-of-stake chains are also faster than proof-of-work chains.

Interestingly, asides from being a way to validate transactions on the blockchain, staking in itself have a lot of benefits outside blockchain validation. Even when you stake with third-party companies, you can still earn rewards regularly by just owning the crypto and not selling it. Amazingly, some third-party stakers like Ankr offer something called liquid staking where you can also spend your tokens through a derivative while it is staked. Aside from being a way to validate transactions on a blockchain, staking can also be a way to earn easy crypto yields. It is the equivalent of accruing interest on a savings account but with way better interest.

But all of this comes at a cost. Since it usually is not as decentralized, it may not be as tamper-proof as proof-of-work chains. Now, this does not mean that this mechanism is not safe. It just means that it is a lot harder to hack a proof-of-work system than to hack a proof-of-stake one.

When we get down to the bones of it, both systems are quite similar. For one, both of them have dire economic consequences for people who disrupt the network. With proof-of-work, disruptors would have to expend an incredible amount of energy and resources in terms of energy bills and hardware to successfully disrupt the network. With proof-of-stake, disruptors would risk losing the value of their tokens when they validate false transactions. If they act against the interest of the network and verify a bad block of transactions, their staked tokens will be slashed.

The difference between both consensus mechanisms can be described in one sentence; ease and security. Proof-of-stake is a lot easier for the validators and is a lot easier on the environment, but proof-of-work may be a fraction more secure.