What the electricity buys

The energy argument about Bitcoin is usually conducted by two groups talking past each other. One says the consumption is enormous, which is true. The other says it secures a global settlement network, which is also true. The disagreement is about whether the thing secured is worth securing, and that is a values question, not a technical one.

What can be settled is the mechanism, and most people on both sides get it wrong.

Why the electricity is the point

Proof of work makes rewriting the chain expensive. To reverse a transaction you would have to redo the work of every block after it, faster than the rest of the network builds new ones. The cost of that attack is the cost of the electricity and hardware. Remove the energy and you remove the security. This is not an inefficiency to be optimised away; it is the product.

Whether that is the right way to buy security is a fair question, and Ethereum answered it differently in 2022 by moving to proof of stake, where the cost is capital at risk rather than electricity spent. Bitcoin will not make that change, for reasons that are as much cultural as technical.

The economics that determine consumption

Miner revenue is block reward times price. Costs are electricity, hardware depreciation and hosting. In a competitive market, miners add capacity until the marginal miner barely breaks even, which means total energy consumption tracks the price of bitcoin, not the number of transactions.

This matters for two common mistakes. Energy per transaction is a meaningless metric, because the energy would be spent whether the block held one transaction or four thousand. And the halving cuts revenue in half, which forces the least efficient machines off the network and reduces consumption until price compensates.

Our mining calculator makes the break-even visible: at typical machine efficiency, the electricity price where a miner stops being profitable is the single most important number in the industry.

Where miners actually put their machines

A miner is an electricity buyer with no location preference, no latency requirement and the ability to switch off in seconds. That is an unusual and useful profile for a grid operator.

Miners cluster where power is cheap and otherwise hard to sell: hydroelectric capacity in the wet season, flared gas at oil wells, curtailed wind at night, geothermal near volcanoes. Several grid operators pay miners to shut down during peaks, which makes them a demand-response resource, and the payments are sometimes larger than the mining revenue.

The honest counter is that miners also consume grid power in regions where the marginal generator is fossil, and that local electricity prices have risen near large operations. Both are true in different places, and general claims in either direction are usually selective.

The heat that gets thrown away

All of the electricity becomes heat. A few operations sell it: greenhouses, district heating, drying timber, warming swimming pools. This is economically marginal and technically straightforward, and it is the most underused opportunity in the industry, mostly because mining wants to be near cheap power and heat customers are near cities.

The long-run question

As the block reward halves toward zero, transaction fees must eventually pay for security. Today they are a small fraction of miner revenue. Whether a fee market alone can fund enough security is the genuine open question about Bitcoin's design, and it will not be answered for decades. Anyone who tells you it is settled, in either direction, is guessing.

What to watch

Watch fees as a share of miner revenue, which is the long-run viability signal. Watch machine efficiency in joules per terahash, which sets how much energy a given security budget buys. And watch demand-response contracts, which are the clearest evidence that grids find flexible load valuable.