The Quiet Tightening

A money system that needs your confidence ends up watching what you say and what you buy. It gets there by arithmetic.

Belief is the collateral

The money in your bank account is a promise. The bank owes you that number. The bank’s own health rests on its borrowers keeping their promises, and the government’s debt rests on lenders believing they’ll be repaid in money that still buys something. Nearly everything we call money is somebody’s IOU, because money in this system is loaned into existence.¹ When a bank writes a mortgage, it creates the deposit in the same moment.² If every loan were repaid at once, most of the money would vanish with it.

An IOU has value only while people believe it will be honoured. So in this system, belief isn’t decoration. Belief is the collateral [the thing of value standing behind a loan]. That holds for every promise in this system that carries no public cover, and for what every promise buys. Where a public scheme covers the number, the scheme carries that number up to its cap, and the scheme is itself a promise.³ When an official says “confidence must be maintained”, the sentence is literal. It’s an engineering requirement.

The world carries about 365 trillion dollars of debt against a world economy that produces roughly a third of that in a year.⁴ That debt as a whole can’t be repaid in money that holds its value. One borrower can shrink its own debt against what it produces in a year, and Canada, Belgium, and Sweden did, each by a third or more, with prices rising about 2% a year.⁵ What can’t happen is all of them at once, because the money they’d repay with is made of those same loans. There are only two exits. Default, which breaks the promise openly. Or creating more money and credit so each unit is worth less, which breaks the promise quietly. The second exit is quiet while it’s taught as health rather than named as a transfer, and looking at it directly breaks the story, not the mechanism.

A system standing on belief in a promise it can’t keep must manage belief. It manages belief first with stories. When rival stories threaten the belief, it polices stories. And when people stop believing anyway and act on their doubt, which they do through payments, it watches and then controls payments. Speech is doubt forming. Doubt acted on shows up as a payment, and most payments carry no doubt, so the ones that do have to be told apart from ordinary traffic. The system ends up watching both because they’re the same threat at two different stages.

The bridge between broken money and less freedom

This is the bridge between “the money is broken” and “society gets less free”. A monetary design problem becomes a civil liberties problem through ordinary incentives, with no bad intent needed anywhere in it. Walk across this bridge and the surveillance headlines you already see stop looking like separate news items and start looking like one system defending itself.

Step one: confidence holds everything up

This isn’t abstract. In late 2008 some letters of credit froze for days. A letter of credit is a bank’s guarantee to a foreign seller that an importer’s payment will arrive. The ones that froze were fully backed, paperwork in order. Across world trade as a whole, the record shows fewer letters of credit and tighter credit lines between banks, not a general halt, and most of the fall in trade that year came from falling demand.⁶ But where belief flickered, the goods stopped moving. Containers sat on docks because banks had stopped trusting each other’s promises. The ships existed, the goods existed, the crews existed. A tower of IOUs works only while each party believes the next will pay. The moment that belief breaks, credit that took decades to build can evaporate in days.

Step two: the promise can’t be kept, so confidence must be manufactured

Two forces are colliding. Technology keeps making things cheaper to produce, so under a money nobody can add to prices would fall and your money would buy more. But debts are fixed in pounds. A system that runs on debt can take a slow fall that comes with better tools. In that fall it’s the cost of making things that drops, and pay drifts down slowly behind it, so the money to meet a fixed payment is still there. What it can’t take is a sudden fall in the prices credit pushed up, houses first, because then the losses spread through the banks. In that fall it’s the lending that has stopped, so incomes and the house fall together and at once, and the fixed payment is hit from both sides in the same year. Houses come first because people buy them with borrowed money. The debt takes a bigger bite of every income. And where the loan was written against a house, the house is worth less too, so the bank stands to lose more on that loan. So every time a sudden fall in those prices threatens, the system is rescued with new money, and the rescues keep getting bigger.

In between, policy targets rising prices of about 2% a year, forever.⁷ Central banks publish why they aim there. They want room to cut interest rates in a downturn, they know wages resist falling, and they say the official measure of prices overstates the rise.⁸ Those are the reasons they give, and the debts aren’t among them.⁹ The debts show up in the rescues, not in the everyday 2%. Getting there takes new money that first cancels the fall better tools would have delivered, and then adds two on top. Governments and central banks create that money and credit to push prices up.¹⁰ They mean to.

Follow what that means for the saver. If the interest you’re paid is below the rate at which prices rise, you’re guaranteed to lose purchasing power [what your money actually buys] by holding the “safe” asset. That’s the mechanism working as designed, because holding returns below the rate prices rise is how the unpayable debt gets shrunk, and the bias runs that way whenever paying savers above inflation would threaten the debt.¹¹ The pages a saver reads don’t name that debt.¹² The system’s own research does, plainly and free to read.¹³ What varies is who the document is written for, and a saver who sees the transfer clearly stops volunteering for it.

The strain also grows on its own. Between 2000 and early 2018 the world added roughly 185 trillion dollars of debt to get about 46 trillion dollars of growth.¹⁴ Each new pound of debt buys less real output than the last. Meanwhile the tools are improving faster than they did, as software and AI spread, so the fall in prices they would deliver keeps getting steeper. What the record shows growing is the stock of debt and the size of each rescue, from hundreds of billions in 2008 to trillions in 2020, not the yearly pace of new money.¹⁵ A gap that reopens every time the new money is withdrawn has to be explained for as long as the system stands, and each new round needs a public reason.

Step three: the story becomes a policy instrument

In a credit system, what people expect decides which outcome arrives among those a bank’s books allow. If everyone believes the banks are sound, they keep going. If everyone stops believing on a Saturday, the banks fail by Monday, because no bank can pay everyone at once. Central banks know this, which is why they reach for “forward guidance” [official statements designed to steer what markets expect] as readily as for interest rates. In this system, talking is doing.

Where saying a true thing about a bank or about the system would cause the thing it describes, the system treats that statement as a risk to stability rather than as opinion. For whoever owes a duty over the information, the law already says so. The bank may delay telling you, and its regulator may not tell you at all.¹⁶ You can hear it in the language. “Quantitative easing” instead of “the central bank creating new money to buy bonds”.¹⁷ “Transitory” instead of “we misjudged”. A 2% inflation target taught as health, when it’s a 2% yearly transfer of purchasing power away from everyone who saves or works for wages, compounding for life.¹⁸ Even the vocabulary helps the story along, because we have everyday words for prices rising and almost none for the good kind of falling prices, so most people never even consider the alternative.

No one has to plan those euphemisms. Each choice is locally reasonable. Calm the market. Don’t spark a run. Reassure. But add up thousands of locally reasonable choices and you get institutions that speak to keep belief up rather than to describe what’s happening.

Step four: from telling stories to policing rival stories

As the strain grows, the pressure lands on speech.

When the transfer stays quiet, the anger it causes goes somewhere other than the money. New money reaches asset owners first, before wages adjust, so the people closest to its creation gain and everyone further from it falls behind them, and that widening gap produces real fury that needs a target. Renters versus landlords, old versus young, one tribe versus another. The division serves the system, because every hour spent fighting each other is an hour not spent looking at the base layer, the money underneath everything, where the transfer happens.

Rival explanations become dangerous to the system, and I mean that in a mechanical sense, not a rhetorical one. In a system where belief is the collateral, a rumour about a bank’s health can cause the run it describes. A clear explanation of what inflation does can change how millions save. So when the strain is on, the system experiences critical speech as an attack on the collateral rather than as opinion. That’s the internal logic by which speech about money, banks, and the health of the system starts being policed. It gets policed under the labels already to hand, which are “market abuse”, “financial promotion”, “scam”, and “public order”.¹⁹ Crises make people accept it, because in a crisis safety feels urgent and dissent feels risky, and the tools an emergency brings in rarely leave with it.

This is also why constitutions don’t hold the line. Money outranks law. When preserving the monetary system requires control, laws get reinterpreted to fit the need. In 1933 the United States ordered private citizens to hand in their gold and made keeping it a crime.²⁰ In 1971 it ended the dollar’s link to gold by decree.²¹ Neither was the act of a fringe state. The rules changed once the promise required it.

Step five: from policing stories to controlling transactions

The final move is forced. Stories keep people believing. They don’t keep people in. And staying in is what the arithmetic requires, because holding returns below the rate prices rise only shrinks the debt while the holders of the currency and its bonds keep holding them, and a credible exit limits how far that can be pushed against the holders who are free to leave.²²

A person who stops believing doesn’t have to say a word. They act. They move deposits out of a shaky bank, buy dollars, buy gold, buy bitcoin, move money abroad. The payment is the dissent. From the system’s side, a sell order is a rumour made real. So control follows doubt down the same path it travelled. As the strain rises, managing the story and managing the exits escalate together, and the exits get managed because stories on their own don’t keep people in. Some get closed bluntly, for everyone at once. Withdrawal limits, penalties on cash, frozen accounts, capital controls. Capital controls are rules restricting moving money out of the country or into certain assets. Closing an exit for one person and not another needs surveillance first, because you can only aim at what you can see.

A devaluation compresses the sequence into days. In December 2023 Argentina’s new government cut the peso’s official value by more than half in one televised address, issued a protocol against blocking the streets two days later, and lined the route of the first march with riot police six days after that. The controls that cap how many dollars you can buy were still in place a year later. The population took a massive pay cut by decree, and no parliament would have passed that cut as a tax.

The endpoint is the central bank digital currency, which is government money issued as programmable entries on the central bank’s own ledger, not as a claim on your bank. Once money is software run by the state, the issuer watches the payments on its own ledger and can set conditions on whatever that ledger carries, so the step from watching to controlling is a keystroke for a limit or an expiry on a balance. A rule about what you’re buying or where you’re standing needs the purpose or the place on that ledger too, and that’s the kind of rule programmable money can enforce. Expiry dates on your balance. Deeply negative rates you can’t escape into cash because cash is gone. Spending that fails because of what you’re buying or where you’re standing. An account switched off by keystroke where the state holds the account and the name together, and by a legal order through your bank where the bank still holds the name. Every feature is optional at launch. Crises afterwards make each one tempting, because now the switch exists. And AI drops the cost of both ends of this, the narrative shaping and the transaction watching, at exactly the moment the system’s need for both is rising.

Each step is forced by the one before

Money loaned into existence runs on belief. The belief is in a promise the arithmetic says can’t be kept. So doubt is the enemy. Doubt forms as speech and is acted on as payments. A system defending the promise therefore ends up watching speech and controlling payments, because each step was the locally rational defence of the promise once the step before it stopped being enough, and nobody had to choose surveillance as a goal for that to happen.

The four places people break the chain

“Strong constitutions and courts will stop this.”

They bend. If the survival of the money system requires control, laws get adjusted or reinterpreted to fit, as 1933 and 1971 showed.²³ Institutions are only as strong as the incentives pressing on them. Weimar Germany had a constitution too.

“Payment monitoring fights crime, and some misinformation really is false.”

Both partly true, and I take them seriously. But tools can target crime without watching everyone. The scope of financial surveillance widens as monetary strain rises. So where crime is the reason given for an expansion, ask whether targeted tools could have reached the crime. Watch what triggers each expansion.

“Better leaders would resist.”

Incentives beat intent. Put anyone decent in the seat and they face the same choice. They can let the debt collapse now, on their watch, or extend support and control. They were appointed to stop exactly that. The structure produces the behaviour regardless of the person.

“This reads as paranoia about tools that mostly don’t exist yet.”

The direction is already on the record in currencies under real stress. Bank freezes, capital controls, forced conversion, and policed protest. They arrive with the stress. One may be lifted when the stress eases, but they tend to persist, and the next crisis brings them back. The disagreement is only about timing and about whether your jurisdiction is special.

A money that makes no promise

The whole machine defends a promise. A money that makes no promise needs no such machine. Nobody manages narratives to protect a fixed supply, because no story, no vote, no panic, and no decree can change it. There’s nothing to take on trust, only rules anyone can verify. Two things have to hold. The network stays spread out enough that nobody can rewrite the rules, and the work that keeps it secure goes on being paid for. Unlike a promise, both are things you can check yourself. That’s why fixing the money comes before fixing speech. Remove the lie at the base and the machine built to defend the lie loses its purpose.