← All guides

Proof of work vs proof of stake explained: two ways to keep a ledger honest

Market basics · 7 min read

Financial disclaimer, not financial advice. This article is general information and education only. It does not recommend buying, selling, or holding any crypto asset, fund, or product, and it is not personalised financial, tax, or legal advice. Crypto assets are volatile and speculative and can lose all of their value. Consider your own circumstances and, where appropriate, speak with a qualified professional in your jurisdiction before making any financial decision.

Both systems answer the same question: on a network with no central authority, who gets to add the next block of transactions, and what makes it expensive for that person to cheat? Proof of work answers with computation: miners race to solve a cryptographic puzzle, and the winner writes the block. Proof of stake answers with collateral: validators lock up the network's own tokens, and the protocol selects among them to propose blocks. Everything else in the comparison, energy use, speed, who can participate, how cheating is punished, is a consequence of that one design choice. Proof of work pays for security with electricity already spent. Proof of stake pays for it with capital still at risk.

How proof of work buys security with electricity

In a proof-of-work system, miners compete to find a hash, a cryptographic fingerprint of the block's data, that meets the network's target. The search is deliberately wasteful: it takes enormous numbers of attempts, run on specialised hardware such as ASICs and graphics processors, and the first miner to find a valid hash earns the right to add the block and collect its reward. The security comes from the cost. Rewriting past transactions would mean redoing that work faster than the rest of the network combined; controlling a proof-of-work network would require more than half of its total hashing power, which is immensely costly to assemble and to run. A miner who produces a block the network rejects has simply burned the electricity, the loss is the punishment.

How proof of stake buys security with capital at risk

Proof of stake removes the race. Validators deposit, stake, the network's native token as collateral, and the protocol selects a validator to propose each block, with selection chances weighted by the size of the stake. Other validators check and attest to the block. There is no puzzle, so there is no need for warehouses of power-hungry hardware; validators run on ordinary computing equipment. The punishment changes currency too: instead of wasted electricity, a validator that breaks the rules, proposing conflicting blocks, for example, can lose part of its staked tokens, a penalty called slashing. Honest behaviour earns rewards; dishonesty destroys the validator's own money. The deposit is the security.

Side by side: the honest comparison

QuestionProof of workProof of stake
Who writes the next block?The miner who wins the puzzle raceA validator selected by the protocol, weighted by stake
What does participation cost?Specialised hardware plus ongoing electricityCapital locked up as stake
How is cheating punished?The wasted electricity and hardware time on a rejected blockSlashing, part of the stake is destroyed
Energy useHigh by design; the spending is the securityFar lower; no computational race
Cost of a majority attackMore than 50% of total hashing powerA large share of all staked tokens
Best-known examplesBitcoin, DogecoinEthereum (since September 2022)

The energy row deserves its documented scale. When Ethereum completed its switch from proof of work to proof of stake, the Merge, on 15 September 2022, its energy consumption dropped by an estimated 99.95%, according to ethereum.org. That figure is an estimate for one network's transition, not a universal constant, but it shows the order of magnitude: removing the race removes nearly all of the electricity.

Finality: the difference a table row cannot hold

Ask when a transaction becomes irreversible and the two systems diverge in a way that matters practically. In proof of work, finality is probabilistic. Each additional block built on top of yours makes reversal harder, because an attacker would have to redo more work, but the protocol never declares a transaction permanently final. That is why services wait for a number of confirmations: each one is a further reduction in risk, not a switch from unsafe to safe.

Many proof-of-stake networks instead have an explicit finality step. On Ethereum, blocks can become final through validator votes, generally after roughly two epochs, a defined point after which reversing the block would require destroying a large amount of staked capital. Probabilistic versus declared finality is not a quality ranking; it is a different shape of guarantee, and it explains why "wait for confirmations" advice sounds different depending on which chain a deposit is arriving on.

What a majority attacker still cannot do

Majority attacks are usually described by their cost. Their limits matter just as much. A miner or validator with majority power operates inside the network's consensus rules, which every node independently enforces when it validates transactions and blocks. A block containing transactions that break those rules is not accepted by the network's nodes, whoever produced it. Majority power is therefore mainly a disruption weapon, the power to censor transactions, or to attempt to reverse one's own recent payments by reorganising recent blocks. It is not a master key: it does not rewrite the rules the nodes validate against, and it does not reach into wallets whose owners never signed a transaction. Knowing the boundary keeps the risk in proportion: serious, specific, and narrower than the headline version.

What any of this means for a holder

Less than the debate suggests. The consensus mechanism is a property of the network, not of your wallet: it does not change how you store keys, what your coins are, or what they are worth. Ethereum's own documentation makes the point about the largest consensus switch in history: the Merge "did not change anything for holders/users", no action was required, there is no "old ETH" and "new ETH," wallets worked exactly the same afterwards, and anyone claiming otherwise is, in ethereum.org's words, likely a scammer. That last point is a standing rule, not a historical footnote: any future network upgrade that supposedly requires you to send coins somewhere, swap them for a "new version," or hand over keys is the same scam pattern wearing a new date.

Misconceptions to drop

Three claims fail the checks above and should be retired on sight. First, that proof of stake coins are "backed" by the stake, the stake is validators' collateral for good behaviour, not a reserve standing behind holders' coins. Second, that one system is simply "more secure", they buy security in different currencies and fail in different ways, through mining concentration on one side and stake concentration on the other. Third, that a network changing its mechanism changes what holders own, the one time a top network actually did it, the documented holder impact was zero. The mechanism decides who writes the ledger and what cheating costs. That is a real and interesting difference. It is also the whole difference.

, Sterling Cole, Money Writer. Earning Desk

Reminder: Crypto Investing is educational content only, not financial advice. Crypto assets are volatile and can lose all value. Product terms, fees, and rules change, read the current prospectus, terms, or manufacturer documentation for any product you are considering.