When Evergrande’s 2.44-trillion-yuan debt bomb finally went off, everyone’s favorite word was “evaporated.”
I refuse that word. Money does not disappear; it only moves from one person’s hands to another’s. So whenever money that should exist is gone, a dog must have taken it — that is my Dog-Takes-the-Money Theorem. I know it sounds like a joke, but it cuts to the essence.
“The money is gone” is not a description. It is a cover-up — cover for some party that is holding it right now. Evaporation is the dog’s camouflage.
What Doesn’t Count
One cut like this turns a punchline into a theorem.
A theorem that can never lose is superstition. So start with the counterexamples: what doesn’t count?
Buy a coffee and the money lands in the barista’s pocket — doesn’t count. That is “money that shouldn’t exist anymore, no longer existing”: you knew exactly what you were getting, money became a thing. The theorem only governs money that should exist but is gone. Insurance premiums don’t count either: you pay, you get a promise, an even trade. When does it count? When you check on a claim and the insurer refuses to pay — money that should still be there is gone, and only then does the dog show itself.
With those two boundaries drawn, the theorem makes a falsifiable prediction: wherever “money that should exist” disappears, follow the chain of transactions and you will find a party that took the gains without bearing the costs. If you can’t find one — I’m wrong.
When Goods Rot, Money Moves
The standard objection: a typhoon wrecks the pomelo harvest — surely there’s no dog here?
I worked this out while watching a typhoon in Yuhuan. When Typhoon Dolphin came through, Yangen village lost roughly 20 million yuan. The ruined pomelos were truly ruined — on the ledger of goods, real destruction. But what about the ledger of money? The pomelos the wind missed went up in price; farm-gate prices in Hualien rose from 14–15 yuan per jin to 18–20. My loss did not vanish. It moved — into the pockets of merchants whose fruit was still standing.
Money and goods are two separate ledgers. Money is a bookkeeping symbol; goods are physical things. Goods rot; entries don’t. Losing your money and losing your goods are two different events. Goods gone, you can blame the typhoon. Money gone, there is always a hand. Some dogs are concrete: an insurance company, a major shareholder cashing out at the peak. Some dogs are abstract: all policyholders, all taxpayers. Abstract does not mean nonexistent.
The Academy Adopts It
The rent-seeker is that dog.
Economics has no theorem called Dog-Takes-the-Money. But its relatives are all there: there is no free lunch — every gain hides a cost; zero-sum games — your decrease is someone else’s increase; double-entry bookkeeping — every debit has a credit, and they must balance. The closest bloodline is rent-seeking: taking wealth that others created through unproductive means. Textbooks file all of this under “market failure,” phrased politely. Dog-Takes-the-Money Theorem says it in one line — and points at the dog.
The Strongest Objections
A theorem cannot stand until its strongest opponents take the stage.
Using it takes four questions. Deposit money in a bank — who takes the spread between deposits and loans? Trade stocks — who takes the spread when you buy high and sell low? Sit down at a casino — who takes your chips? Buy insurance — who takes your premium? Every financial act should begin with these four questions. It is practically a universal formula. Has anyone won a Nobel Prize with it? Possibly. I’m just a little later than they were.
But for the theorem to stand, the strongest opposition must be heard. There are five objections; the first two bite hardest.
First: technological progress. AI creates value; new value is not a transfer; there is no dog. It doesn’t hold. That money is flowing, tragically, out of ordinary workers’ hands and into Jensen Huang’s at Nvidia. The creation is real, and so is the relocation — the two ledgers each record their own entries, and neither cancels the other.
Second: value destruction. A 200,000-yuan car, after five years, sells as scrap for 5,000 — value truly destroyed, nobody benefits. But that is the ledger of goods, not of money: the money was divided up the day the car was bought — carmaker, gas stations, taxes. Only the car was destroyed. Value destruction is the loss of a physical thing, not the disappearance of money.
The remaining three — legitimate returns, risk premia, systemic risk — can’t get a grip either: the theorem describes the fact of transfer and refuses to grade its morality. A risk premium is a transfer; in a systemic crash the money still moves, from countless retail pockets into a few.
With all five swallowed, the theorem has exactly one soft spot left: it cannot predict the future; it can only assign responsibility for the past. Accepted. An accountability tool was never in the prophecy business.
The Academic Version
Private gains, socialized losses.
Translated into academic language: when a wealth loss occurs, there exist parties who have already taken real gains from this chain of transactions yet bear none of the cost. That is the most unfair structure there is — private gains, socialized losses. Evergrande is its best case study: dividends, cash-outs, and offshore gains were banked long ago, and the 2.44-trillion-yuan hole was left to homebuyers, suppliers, and wealth-management investors. I have a dedicated page on this site that works through that ledger, item by item. The theorem’s English name — The Dog-Takes-the-Money Theorem — is the one that page already uses.
Laozi saw through this chain long ago: the Way of Heaven takes from those who have too much and gives to those who have too little; the way of man takes from those who have too little and gives to those who have too much. Dog-Takes-the-Money Theorem is the way of man in its contemporary draft — a violation of Heaven’s way, yet the dog still took it.
So the next time you hear someone say “the money is gone,” don’t sigh along.
Just ask: where’s the dog?
