by An all-digital economy promises a world without friction. The same design that makes payment effortless makes denial effortless, and it hands every government a quiet way to decide whether you may eat. Pay for coffee with a glance at your phone. Send rent across the country before the barista finishes the pour. Split a dinner bill at the table with no cash, no card, and no fuss. The case for a fully digital economy arrives wrapped in that smoothness, and the smoothness is real. The danger sits one layer underneath, in a question almost no one asks at the register: who has to approve this payment for it to clear, and what becomes of me on the day they decide not to?
We are being walked toward an arrangement in which money and property exist only as entries in databases that a central authority controls. Bank balances, wages, pensions, the deed to a house, the title to a car, the small float that gets a family from one paycheck to the next, all of it rendered as numbers that someone else can edit. The pitch emphasizes speed, lower costs, and an end to the grubby inconvenience of physical currency. The cost it leaves out is the loss of any way to hold or move value that does not first pass through a gate, and the corresponding power that gate confers on whoever owns it.
The architecture of the switch
Money comes in two broad forms, and the gap between them decides who holds power over any single transaction. Physical cash works on possession. Whoever holds the note can spend it, the exchange completes between two people, and no central ledger records that it happened. A digital balance works on permission. Every payment routes through an institution that reads the account, applies whatever rules are in force, and then lets the transfer through or stops it cold. That routing step is the entire contest. When it belongs to one authority, that authority can attach conditions to your money and can revoke the privilege of spending it at will.
The word carrying the weight in the official vocabulary is “programmable.” Engineers and central bankers use it to describe money that carries instructions about where it may be spent, on which categories of goods, by which deadline, and by whom. China has staged the clearest demonstrations. During a 2020 trial in Shenzhen, the People’s Bank of China handed digital yuan to about fifty thousand residents and coded the money to expire if it went unspent within a few days, a deliberate experiment in currency that self-destructs on a clock. Researchers publishing in the Journal of Risk and Financial Management in August 2025 examined that pilot and connected it to an older concept, the Gesell currency, money engineered to bleed value unless it keeps circulating, which gives the issuer a dial on how fast a population spends.
The Chinese system has since grown into something national in scale. By November 2025 the digital yuan had processed billions of transactions, and from the start of 2026 the central bank began allowing interest on e-CNY wallets, a feature that pulls the instrument further from any role as a simple cash substitute and deeper into the machinery of state monetary control. A senior official at the State Administration of Foreign Exchange has publicly endorsed building programmable features into the currency, the same capabilities that let money be told what it may buy and when it must be used. The e-CNY runs on a centrally controlled architecture rather than an open ledger, which means the authorities hold the records and the rules in the same hand. A government with that hand can set an expiration date on a balance, restrict it to approved merchants, or empty an account it judges noncompliant.
What already happens
The capability already operates, long before any finished digital currency arrives. Today’s hybrid system, banks and payment processors layered over a shrinking base of cash, already contains discretionary levers, and those levers get pulled.
Consider Canada in February 2022. During the trucker convoy that occupied downtown Ottawa and blocked border crossings, the federal government invoked the Emergencies Act for the first time since that 1988 law was written, and brought in financial measures that let banks freeze accounts tied to the protest. Testimony to a House of Commons finance committee put the tally at roughly 206 accounts holding about 7.8 million Canadian dollars, frozen without any court order, after the national police force passed names to the banks. One member of Parliament from British Columbia reported that a constituent saw her account frozen after donating fifty dollars. Most accounts were released within days, and the emergency declaration was revoked on the twenty-third of that month, yet the precedent stood: a government had reached into private accounts and shut them on its own authority, ahead of any judicial finding. Canadian courts later concluded that the use of those emergency powers had been unlawful, a vindication that arrived long after the rent was due and the damage was done. That sequence is the lesson. The freeze is instant. The remedy crawls.
Consider as well the case of Nigel Farage in Britain in 2023. The private bank Coutts, owned by NatWest, moved to close his accounts, and a subject access request produced an internal dossier from the bank’s reputational risk committee that weighed his publicly stated political positions, his role in Brexit, his association with Donald Trump, and his views on other matters against the bank’s stated values. Both chief executives, at Coutts and at NatWest, resigned in the fallout, the latter after admitting she had discussed the case with a journalist. Honesty requires saying that the episode is contested. The Financial Conduct Authority reviewed thirty-four firms and reported that none had closed accounts primarily because of political views, while cautioning that further work was needed. An independent review by the law firm Travers Smith concluded that the decision was lawful and predominantly commercial, since the relationship was losing money, even as it found serious failings in how the bank handled confidential information. Read either way, the structural fact survives the dispute. A major bank ran a committee that documented a customer’s political opinions and treated them as grounds touching on whether to keep his accounts, and the workings of that committee stayed invisible to him until a legal request pried them loose. The machinery for cutting a person off from the financial system on grounds beyond plain commerce exists, it operates with little transparency, and the only reason the public learned its vocabulary in this instance was that the customer happened to be a famous politician with lawyers.
Both of these episodes unfolded in mature constitutional democracies with independent courts and an aggressive free press, which is the part worth sitting with. The capability lives inside ordinary institutions, available the moment a government finds a reason and the law to reach for it, with no dictatorship required. Multiply that machinery by a system in which cash has been removed and every balance lives in one central wallet, and the discretion stops being occasional and becomes total.
How cash gets taken away
The obvious objection is that none of this matters while cash remains an option, which is precisely why the migration to digital money so often begins by making cash scarce.
Nigeria offers a recent and painful case. The country launched the eNaira in October 2021, the first central-bank digital currency in Africa and the second in the world after the Bahamian Sand Dollar. Adoption stalled, with fewer than half a percent of Nigerians using it a year on. Late in 2022 the Central Bank of Nigeria redesigned the highest-denomination notes and imposed tight limits on cash withdrawals, and the cash in circulation collapsed from around 3.3 trillion naira to under one trillion by February 2023. What followed was a humanitarian emergency that exposed how little the digital alternative could carry. Markets seized, the informal economy that holds most of Nigerian daily life ground down, and riots and violent protests broke out in the weeks before a presidential election, until the Supreme Court intervened to extend the validity of the old notes. The Carnegie Endowment found that by March 2023 the eNaira amounted to less than a tenth of one percent of the cash supply, nowhere near enough to replace what had been withdrawn, and the World Bank noted that only about forty percent of Nigerian adults even held a bank account. The government, for its part, pointed to a jump in eNaira adoption during the crunch as a sign of success. By 2025 a privately issued naira stablecoin had begun to find real users, and the reason analysts gave was telling: people trusted it in part because the central bank did not control it.
India ran a larger version of the same experiment in 2016. On the evening of the eighth of November, Prime Minister Narendra Modi announced on live television that the 500 and 1,000 rupee notes, about eighty-six percent of the cash in a country of more than a billion people, would cease to be legal tender at midnight. The stated goals were to expose untaxed “black money,” to kill counterfeits, and to push the nation toward digital payments. Citizens were given weeks to deposit or exchange the voided notes, under daily withdrawal caps of a couple thousand rupees, and the informal sector that employs the overwhelming majority of Indian workers absorbed the shock. The country’s own central bank board had warned that the premise was shaky, and the warning proved sound. Analysis from Brookings found that around ninety-seven percent of the demonetized currency came back into the banking system, which meant the hoards of illicit cash the policy was meant to trap had mostly never existed in that form, since only a tiny share of illegal wealth is ever held as banknotes. There were reports of deaths tied to the upheaval, to the endless queues and the sudden paralysis of a cash economy. What endured was the surge in digital and bank-based payments. Note the asymmetry, because it recurs every time. The justification underdelivered. The migration to trackable money overdelivered. When a policy reliably produces the second result while failing at the first, the second result is worth treating as the actual aim.
Why a free society should care
Money is the substrate of a great deal more than shopping. The ability to pay underwrites dissent, movement, worship, association, and conscience. A journalist investigating a regime has to buy a train ticket and pay a source. Workers on strike need a fund that the employer’s allies in government cannot drain. The woman in a hostile jurisdiction must pay for lawful medical care in another state without leaving a record that flags her for prosecution. Members of a disfavored faith or party have to keep the lights on while officialdom decides they are a problem. Each of those acts depends on being able to transact without first satisfying some central authority’s opinion of one’s conduct.
Programmable, centrally held money dissolves that independence by moving enforcement to the point of sale. Under a system of law worth the name, punishing someone financially generally requires a charge, a hearing, and an order, the slow apparatus of due process that exists to keep power honest. A central digital wallet lets an administrator apply a rule to your balance first and leave you to contest it later, if you can still afford counsel with your funds locked. The gap between making a rule and enforcing it, the gap where courts and appeals and evidence live, closes to nothing. The twentieth century’s authoritarian governments needed cordons, camps, and a great deal of manpower to make a person vanish from economic life. A government with a master ledger needs a policy and a database. The horror of it is the quietness. No troops appear in the street. A number stops working, and a citizen discovers that the right to buy food has become a permission that can be declined.
The reassurances, and why they leak
Several honest objections deserve a direct answer, because the danger only counts if it survives them.
The first reassurance is that the United States has already shut the door. In January 2025 an executive order barred federal agencies from developing a central-bank digital currency, and that July the House of Representatives passed the Anti-CBDC Surveillance State Act by a margin of 219 to 217. The protection is thinner than it sounds. That bill has not passed the Senate, so it carries no force of law. An executive order binds agencies only until a later president signs a different one. The Federal Reserve chair’s pledge against a retail digital dollar is tied to a term that runs out around the middle of 2026. More revealing still, the day after that House vote, the GENIUS Act created a federal framework for privately issued dollar stablecoins, and those issuers are required by law to maintain the ability to freeze and block balances, a power the largest issuer has already used to lock up billions of dollars in tokens at the request of authorities. Under the American approach, the capacity to switch off a person’s money changed owners, moving into private hands that the government licenses and can direct. Banning the state’s own coin leaves the off switch intact, with someone else’s finger resting on it.
A second reassurance holds that programmable money fights crime, fraud, and the financing of terror. The empirical record argues otherwise. India’s demonetization recovered almost all of the voided cash, because the criminal hoards it targeted were largely a fiction; the tool is effective because it reliably forces a whole population onto trackable rails, and it is not effective because it seldom catches the wrongdoers named in its justification. A capability sold for one purpose and retained for another should be judged on what it actually does.
Then there is the claim that the courts will protect us. The Canadian episode answers that. Judges did rule the account freezes unlawful, and they did so long after the accounts were frozen, the bills were missed, and the point was made. Enforcement through a ledger is instantaneous. Judicial review is slow, expensive, and uncertain, and it cannot un-miss a mortgage payment.
A fourth promise is that safeguards will be built in: privacy protections, offline modes, limits on how much anyone holds. Holding limits run in the wrong direction for anyone worried about state power, since they assert the government’s authority to cap how much value a citizen may keep. The European Parliament’s draft for a digital euro would also make acceptance of the currency mandatory for many merchants, and a member of that body has framed opposition to the project as opposition to the European Union itself. A safeguard written by the party that holds the lever is a setting, and settings get changed.
What holds the line
The choices in front of us can be ranked, and they should be, plainly.
Protecting physical cash as legal tender and as working infrastructure, with real obligations on businesses to accept it, is effective because cash is the single medium an authority cannot silently switch off, read, or expire. Cash carries real costs, and pretending otherwise would be dishonest, since it does grease some tax evasion and some crime. The trade still favors keeping it, because the harm from occasional anonymous misuse stays bounded, while the harm from a financial system with a master switch has no ceiling. This protection is not effective on its own if cash is allowed to wither through neglect, ATM closures, and quiet merchant refusal until it survives only on paper.
Requiring that any public digital money be non-programmable at the spending layer, with a statutory ban on conditioning, restricting, or expiring balances, is effective because it strips out the most dangerous capability at the source. It is not effective if emergency carve-outs are written in, since the emergency is exactly when the power will be reached for.
Demanding prior judicial authorization and firm sunset clauses for any freezing or seizure of funds is effective because it restores the due process that point-of-sale enforcement erases. It is not effective where courts defer heavily to executive claims of necessity.
Keeping the payment system plural, with cash, several competing private networks, and open or permissionless options all live at once, is effective because redundancy defeats a single chokepoint. It is not effective if regulation funnels everyone onto one approved rail in the name of safety.
Refusing mandatory-acceptance rules and holding limits is effective because it keeps any digital instrument genuinely optional. It is not effective once network effects make refusal a practical impossibility for ordinary people.
My recommendation is direct. Defend all of these, and understand which one is load-bearing. The protection that actually holds is the survival of a parallel medium of exchange that no one can switch off, namely physical cash, kept alive by law and by use. Every other safeguard is a rule, and a rule is a thing the controller can rewrite when the political weather turns. Cash is a fact, and a fact cannot be edited in the dark without everyone seeing it go. Guard the rules, by all means, and do not mistake them for the wall. The wall is the continued existence of a way to pay that requires no one’s blessing.
Coda
Return to the coffee and the phone and the rent that crossed the country in a heartbeat. The convenience is the bait, and the hook is the gate it routes you through. Digital money has already arrived; the open question is whether we keep an exit. A free people needs at least one way to buy bread that passes through nobody’s approval. China built its off switch on purpose. Nigeria and India built theirs through shock and called the wreckage progress. In every case the population moved onto rails it did not own, the stated reason faded, and the new dependence remained. The off switch does not announce itself; it waits inside the architecture, wearing the label of convenience, until someone with the authority and a reason reaches over and uses it.
Leave a Reply