Proof-of-stake vs proof-of-work for money: why Bitcoiners won't switch

Proof-of-stake uses a fraction of the energy proof-of-work does, so the obvious question is why Bitcoin doesn’t follow Ethereum and switch. The answer is that for money specifically, the energy is the security, not a bug to optimise away. Proof-of-work anchors Bitcoin’s history in real-world cost that can’t be voted down; proof-of-stake replaces that with a trust model where new users need a trusted checkpoint to know which chain is real, where stake concentrates in a few large operators, and where a native yield starts to make the asset look more like a security than money. None of that makes proof-of-stake wrong for a programmable platform. It makes it a poor fit for a neutral store of value meant to last decades.

What’s the actual difference for money?

Both mechanisms answer the same question — “which version of history is the real one?” — but differently. Proof-of-work answers with energy: the valid chain is the one with the most accumulated computational work, and rewriting it means re-doing that work at enormous ongoing cost. Proof-of-stake answers with stake: validators put up tokens and are rewarded for attesting to blocks.

The catch sits in how a new participant gets going. On Bitcoin, a node joining for the first time can verify the entire chain from scratch and determine the correct history purely from the data, because the work is baked into it. On a proof-of-stake network, a new node can’t do that — it needs a recent, trusted checkpoint from somewhere (another node, a block explorer, the client software) to know which chain is real. This is called weak subjectivity.¹ The reason it’s unavoidable: old validator keys that have since been unstaked can still sign blocks for past periods, so someone who acquires enough old keys could fabricate a plausible alternative history. Proof-of-work has no equivalent hole, because faking the past would mean re-spending the energy.² For someone who already trusts developers or an explorer, that’s a fine trade. For an asset whose entire pitch is minimising who you have to trust, it reintroduces exactly the dependency you were trying to remove.

Doesn’t proof-of-stake concentrate control among the wealthy?

It’s a real risk, and the numbers show the pressure. On Ethereum, Lido controls roughly a quarter of all staked ETH (about 24.7% in late 2025, down from a 32% peak), with Coinbase around 11.7% and Binance 8.4%.³ That matters because an entity past one-third of stake can stall finalisation. And staking rewards compound: large stakers earn yield they can restake, which can entrench their position over time. Bitcoin mining has no equivalent compounding — miners must keep spending on hardware and electricity to stay in the game, and protocol rewards alone don’t let an incumbent quietly accrete control.

There’s a deeper point underneath. Proof-of-stake makes security a function of token price and staking participation — variables governance can influence. Proof-of-work makes it a function of energy expenditure, a real-world cost that can’t be voted away. Money is uniquely prone to capture; whoever controls it tends, eventually, to abuse that control. Bitcoin’s design minimises that discretion on purpose.

Was the switch to proof-of-stake really about energy?

The timing is at least suggestive. Ethereum’s move accelerated during a wave of environmental scrutiny of crypto. In May 2021 Elon Musk announced Tesla would stop accepting Bitcoin on environmental grounds, knocking over $4,000 off the price the next day.⁴ Morgan Stanley’s sustainability research head argued every dollar of Bitcoin mined was more carbon-intensive than a dollar of gold;⁵ the Bank for International Settlements declared Bitcoin had few redeeming attributes given its energy footprint.⁶ In March 2022 a Greenpeace campaign — “Change the Code, Not the Climate,” backed by Ripple co-founder Chris Larsen — ran ads explicitly demanding Bitcoin switch to proof-of-stake.⁷ The Ethereum Foundation, for its part, framed the Merge heavily around energy, publishing “Ethereum’s energy usage will soon decrease by ~99.95%” and leading post-Merge messaging with carbon stats.⁸

Whether environmental pressure drove the decision or just supplied a convenient justification, Bitcoiners are sceptical that energy use should decide a monetary network’s consensus mechanism — because the energy is what makes attacks expensive. Remove it and you’ve changed the security model, not just the carbon footprint. Vitalik Buterin himself put the underlying point plainly back in 2015: the purpose of proof-of-work is precisely to solve the problem of which chain is real — and in proof-of-stake, that’s exactly the hard question.²

The yield problem: is staking turning ETH into a security?

This is where the monetary argument meets the regulatory one. The CFTC classified Bitcoin as a commodity in 2015; Ethereum has never received that clarity — and its post-Merge yield is a big part of why. Commodities don’t generate income. Gold in a vault produces nothing; the value is in the asset itself. Securities are the opposite — stocks pay dividends, bonds pay interest, and investors allocate by comparing yields.

Proof-of-stake gives ETH a native yield: validators earn roughly 3–4% a year, which Vitalik Buterin has himself called “interest.”⁹ As Allen Farrington and Anders Larson put it, money does not bear interest; securities do.¹⁰ By grounding value in yield, proof-of-stake networks start competing for capital as investment products rather than as money. That’s not academic: the Howey test asks whether there’s an investment in a common enterprise expecting profits from others’ efforts, and staking rewards fit the description uncomfortably well. The SEC has already acted on it — Kraken paid $30 million in 2023 to settle charges over its staking-as-a-service program, with the complaint citing the yield as evidence of a securities offering.¹¹ Ethereum’s native staking sits in a greyer zone (no central issuer, not a company), but the economic substance is similar. Bitcoin sidesteps the whole question: there’s no native yield, holding it produces nothing until you sell or lend it, and that’s exactly how a commodity behaves.

Is proof-of-stake bad?

No. Proof-of-stake may be well-suited to networks optimised for throughput, programmability, or specific applications, and the flexible governance that made the Merge possible is a genuine capability. Proof-of-stake works; the question is whether it’s the right foundation for base-layer money — an asset meant to be a neutral store of value over decades, where minimising trust and discretion is the entire point. For that job, the energy cost of proof-of-work is the feature, not waste. How that security works is covered in proof-of-work’s own mechanics.

Common questions

Is proof-of-stake more energy-efficient than proof-of-work? Yes, dramatically — Ethereum’s Merge cut its energy use by around 99.9%. The Bitcoiner argument isn’t that this is false; it’s that for money, the energy expenditure is what makes the network expensive to attack, so removing it changes the security model.

What is weak subjectivity? A property of proof-of-stake where a new node can’t determine the correct chain from data alone and needs a trusted recent checkpoint. Proof-of-work avoids this because the valid chain is simply the one with the most accumulated work.

Why does staking concentration matter? On Ethereum, a few large operators (Lido, Coinbase, Binance) control a big share of staked ETH, and an entity past one-third of stake can disrupt finalisation. Staking rewards also compound, which can entrench large holders over time.

Why is Ethereum’s staking yield a regulatory problem? Commodities don’t pay income, but staked ETH earns a yield its own founder has called “interest.” That makes it look more like a security under the Howey test — a question Bitcoin avoids because it has no native yield.

Footnotes

1. Weak subjectivity — a property of proof-of-stake where a new node needs a trusted recent checkpoint to identify the correct chain, because unstaked historical keys can still sign old blocks (enabling “long-range” attacks). ↩

2. Vitalik Buterin, “Proof of Stake: How I Learned to Love Weak Subjectivity” (November 2014). https://blog.ethereum.org/2014/11/25/proof-stake-learned-love-weak-subjectivity ↩ ↩²

3. Lido Q3 2025 Tokenholder Update — Lido ~24.7% of staked ETH (down from a 32% peak in 2023); Coinbase ~11.7%; Binance ~8.4%. https://blog.lido.fi/recap-lido-q3-2025-tokenholder-update/ ↩

4. Tesla’s announcement that it would stop accepting Bitcoin, May 12, 2021, and the subsequent price drop. CoinDesk. ↩

5. Morgan Stanley, Jessica Alsford, Global Head of Sustainability Research (2021). ↩

6. Bank for International Settlements, Annual Economic Report (June 2021). ↩

7. Greenpeace “Change the Code, Not the Climate” campaign, launched March 29, 2022, backed by Ripple co-founder Chris Larsen; effectively suspended January 2024. Greenpeace USA; CoinDesk. ↩

8. Carl Beekhuizen, “Ethereum’s energy usage will soon decrease by ~99.95%,” Ethereum Foundation Blog (May 18, 2021). https://blog.ethereum.org/2021/05/18/country-power-no-more ↩

9. Vitalik Buterin, “Why Proof of Stake” (November 2020), referring to staking returns as interest. https://vitalik.eth.limo/general/2020/11/06/pos2020.html ↩

10. Allen Farrington & Anders Larson, “Only The Strong Survive,” Axiom Venture Partners — “money does not bear interest; securities do.” https://www.axiombtc.capital/only ↩

11. SEC, “Kraken to Discontinue Unregistered Offer and Sale of Crypto Asset Staking-As-A-Service Program and Pay $30 Million to Settle SEC Charges,” February 9, 2023. https://www.sec.gov/newsroom/press-releases/2023-25 ↩