The Energy Theory of Money

"It's just numbers" – and so is everything else

The complaint feels devastating: this money has no substance, nothing you can hold, just entries in a database. But pause on it, and the same charge convicts every modern currency – and misses what actually makes a thing hard to counterfeit.

The dollars in your account are also just numbers in a database; the difference people reach for is that dollars are "backed by the government." But backed by what, exactly? By the government's ability to make more of them at will – which is the opposite of backing. Touch is not the test, and it never was. The test of a monetary good is whether anyone can quietly produce more of it.

So before we go further, let me give the strongest version of the obvious rebuttal, because the loose form of this argument deserves to be killed off early. People sometimes say energy money is valuable because it cost so much to make. That is false, and an economist will tell you why in one sentence: cost does not create value. Burn a fortune in electricity baking mud pies and you have expensive mud pies, not valuable ones. Value is in the eye of the buyer – subjective, marginal, set by what the next person will give – never conferred by sweat. If the argument for this money were "it took a lot of work, therefore it is worth a lot," it would be worthless.

So that is not the argument. Here is the real one – the honest "energy theory of money," rebuilt after the naive one is cleared away – and it runs the other way. Cost does not flow into value; it stands guard over supply. The thing a monetary good needs above all is that no one can flood it – that wanting more cannot summon more. Economists call this inelastic supply, and it is the rarest property in the world of money, because almost everything that has ever served as money eventually found a cheaper way to make itself. So the thesis of this whole piece is narrow and defensible: a monetary good needs a supply that cannot be cheaply expanded, and a real, unavoidable cost of production is the most credible way ever found to guarantee it – the cost is the lock, not the value behind the door. Everything that follows is just that one sentence, worked out.

Cost does not become value.
It stands guard over supply.

Scarcity welded to the laws of physics

Bitcoin's supply is not protected by a promise or a vault. It is protected by the one move no central bank has ever managed: making "spend more to get more" mathematically impossible.

Here is the mechanism, and it is the opposite of what people assume. New coins are issued on a fixed schedule – a known number per block, halving on a known clock, ending at a hard ceiling. Miners pour real, measurable energy into the work of adding the next block, but that energy does not buy extra coins. When more energy floods in, the protocol simply raises the difficulty, so the same scheduled number comes out the other end – just guarded by a higher wall. This is the famous difficulty adjustment, and it is the whole trick. Effort cannot inflate the supply; it can only make the supply more expensive to attack. Spend a hundred times more electricity and you do not get a hundred times more coins – you get the same coins, defended a hundred times harder.

That is the property no vault ever had. Gold was sound because supply was costly to expand, but a rich enough strike or a good enough machine has always loosened it eventually. Here the loosening is foreclosed by design: try to win the race by spending more, and everyone else's spending lifts the bar to match, leaving issuance exactly where the schedule put it. A government can print a number. It cannot raise the difficulty on itself. Whatever else it is, money whose supply cannot be bought into existence – at any price, by anyone – is the furthest thing from "conjured from nothing."

A government can print a number.
It cannot print a joule.

Pour in all the energy you like. The orange bar is the work you spend; the darker bar beneath it is the difficulty wall, and it climbs to match you – so a coin still appears only when the schedule's clock comes round, never sooner, never more. Spend ten times as hard and you do not get ten times the coins; you only build a taller wall around the same coins. There is no door marked "more money for more effort." That door was never built.

Spend ten times harder.
You get a taller wall, not more coins.

Price halves; the hash rate barely flinches

On a bad day the most-watched number is the price, and it is the least informative one on the screen. Bitcoin can lose half its value in a few months while the hash rate – the total computing power guarding the ledger – barely dips, and goes on setting records cycle after cycle. The people actually securing the network do not walk out when the screen turns red. The crash is loud; the commitment underneath it is quiet, and it is the part that lasts.

Be precise about why, because there is a popular version of this that runs backwards. "Price follows hash rate," some supporters say, as if switching on machines pushed the price up. It does not – the causality runs the other way. Mining is a business: coins earned in dollars, against hardware and power paid for in dollars, so when the price rises mining grows more profitable and hash rate climbs, and when it falls the least efficient miners go dark. Hash rate is a lagging response to price, not a leading cause of it. But it is a far steadier one: miners with sunk costs and long horizons keep hashing straight through a drawdown, efficiency keeps improving, and the secular trend has climbed relentlessly even across eighty-percent crashes – the one sharp drop, the 2021 mining ban, was policy and not price, and the network clawed it back within months. So hash rate cannot lead the price, but it is the truer gauge of the network's health than the panicking quote everyone fixes on. Watching the price and ignoring the hash rate is watching the weather and missing the climate.

But discard the bad argument and a genuinely strong one is left standing underneath. The energy miners spend is the network's security budget – the running cost an attacker would also have to outspend to rewrite the ledger. Because mining revenue scales with price, that budget scales with price too: a more valuable bitcoin draws more hashing, which makes the chain proportionately more expensive to attack. So price does not follow security, but a higher price does buy more security. That is the defensible claim, and it survives the correction.

There is one honest open question to flag rather than paper over. Today most of that security budget is paid by the block subsidy – the new coins minted each block – which halves toward zero on a fixed clock. In the long run, security must increasingly be funded by transaction fees instead, and whether fee revenue alone will be large and steady enough to keep the wall high is genuinely unsettled. It is the live debate among people who take the system most seriously, and any account that waves it away should be read with suspicion.

A higher price does not follow security.
It buys it.

"But all that electricity is wasted"

If you have followed this far, the title's provocation is now sharpened to its hardest point: fine, the supply is inelastic – but burning rivers of power to defend a ledger is obscene. This is the objection the whole essay owes an answer.

Start by naming what the energy actually buys. It is not a side effect; it is the product. The power spent is the security budget – the price an attacker would also have to pay to rewrite the record, paid continuously so that rewriting it stays uneconomic. A monetary network's energy bill is the wall around its history, and a ledger this open has no other wall. Call that wasteful and you are really saying secure money should be free to attack, which is a stranger position than it first sounds.

Then compare, because "wasteful" is a word that needs a yardstick. Gold is sound money precisely because we move mountains for it – diesel fleets, cyanide leach pads, refineries running on continents of power, all so the float grows a couple of percent a year. The dollar's soundness, such as it is, rests on an apparatus too: a military that secures trade routes, a sprawl of banks and clearing houses and data centres, the whole physical scaffolding of trust. Every credible money has a costly base. The honest question is not whether bitcoin spends energy but whether it spends less for the same job – a globally settled, seizure-resistant, supply-capped ledger – and increasingly that energy seeks out the power no one else wants: stranded gas that would be flared, hydro spilling in the wet season, the cheapest marginal electron on earth. The forge can be lit anywhere a joule is going begging.