Stablecoins Fail The Payment Credibility Test At Scale

Have you ever tried to send what looks like a dollar and discovered it is not quite a dollar when it arrives? That awkward gap sits at the center of a fight that now reaches far beyond crypto Twitter. The head of the institution that coordinates the world’s central banks argued late this week that stablecoins still fail a basic credibility test as everyday money. Not because they are useless. Because they do not yet behave like money when the volume gets large and the recipient is picky.

I have followed this debate long enough to hear both extremes. One camp treats every official warning as a turf war. The other treats every token as a finished product. Reality, as usual, is messier. The speech at Jackson Hole did not call for a ban. It drew a line between instruments that can settle like bank money and instruments that still wobble when markets get nervous.

The Payment Question Officials Will Not Drop

Pablo Hernández de Cos, general manager of the Bank for International Settlements, put the issue in blunt terms on August 28. In their current form, he said, stablecoins are not a credible means of payment at scale. That phrase matters. He did not say they have no market. He said they do not yet pass the test that payment systems must pass if households, firms, and banks are going to rely on them every hour of the day.

His preferred path is tokenized deposits. Those are still claims on commercial banks. They can move on programmable rails. Settlement can still close in central bank money. In his view, that design keeps the foundations of the monetary system intact while still capturing the speed and programmability people want from blockchain infrastructure.

Tokenised deposits offer a more direct path to harness tokenisation while preserving the monetary system’s foundations.

– BIS general manager, Jackson Hole remarks

That is a polite way of saying the plumbing still matters more than the branding. I’ve found that markets often skip this part. A token can look liquid on a screen and still fail the three tests that make money feel interchangeable in ordinary life.

Three Features Money Needs Before It Feels Ordinary

De Cos judged stablecoins against three qualities: singleness, interoperability, and financial integrity. None of those words is decorative. Each one describes a friction people notice the moment they try to use a token as cash rather than as a trading chip.

Singleness is the simplest idea and the hardest one to fake. A dollar in one regulated bank should be worth a dollar in another regulated bank. You should not need a discount window in your head before you accept payment. Stablecoins do not always clear that bar in secondary markets. A holder of one major dollar token may have to sell it before buying another token because the recipient accepts only the second brand. During stress, either token can trade above or below one dollar. The exchange may not happen at par.

Tokenized deposits stay inside a different legal wrapper. They remain liabilities of supervised banks. A transfer can debit one customer and credit another while the banks settle across central bank accounts. That is old architecture wearing a new interface. The point is not nostalgia. The point is par value that does not depend on a market maker’s nerve.

Interoperability is the second crack. Stablecoins live across several chains and scaling networks. Moving the same token often means bridges, wrapped versions, or centralized intermediaries. Each hop adds operational risk, custody risk, or smart-contract risk. People get used to those hops until one of them fails in public.

Tokenized deposits are not innocent here. Most live projects still sit on permissioned networks that do not talk freely to rival platforms. De Cos admitted the obvious: no multi-bank, cross-border tokenized deposit system currently runs at full commercial scale. So the official favorite is institutionally cleaner, not technically finished. That honesty is useful. Hype collapses when someone says the pilot is still a pilot.

Financial integrity is the third test and the one that makes compliance teams sit up. Public chains let people hold and move assets without a regulated custodian in the middle. That can make anti-money-laundering and counterterrorist-financing rules harder to apply with the same consistency banks already face. It does not mean every self-custody transfer is dirty. It means the identification problem is different. Policymakers still have to decide how peer-to-peer transfers should be monitored without turning privacy into a slogan or a crime.

Why This Is Not A Call For A Ban

Read the speech carefully and the ban narrative falls apart. De Cos said stablecoins and tokenized deposits could coexist if regulators define roles and attach safeguards. Under that model, tokenized deposits would carry most daily and wholesale payments. Stablecoins would occupy narrower lanes, including decentralized lending and other specialized activity.

That split will annoy people who want one winner. Markets rarely deliver one winner. Cash, cards, wires, and instant payments already share the same economy. The interesting question is which instrument sits at the center of settlement and which one sits at the edge.

• Tokenized deposits keep the claim inside a bank balance sheet.
• Stablecoins usually keep the claim with a private issuer backed by reserves.
• Both can be programmable. Only one inherits the full bank safety net by default.
• Both can fragment if networks stay closed or bridges stay fragile.

In my experience, coexistence language is how officials buy time. It is also how they admit a product already has users. You do not design a specialist role for something you expect to vanish next quarter.

What The Comparative Rulebook Actually Shows

A day before the Jackson Hole remarks, a study from the Financial Stability Institute compared frameworks in the United States, the European Union, the United Kingdom, Hong Kong, and Singapore. The authors found broad agreement on a narrow issuer job: issue, redeem, manage reserves. They found wide disagreement on everything around that job.

Lending, staking, proprietary trading, and custody sit in different boxes depending on the jurisdiction. The United States and Singapore take a tighter line with specialized non-bank issuers. Under the U.S. payment-stablecoin statute known as the GENIUS Act, those extra activities generally sit outside the core permission set. A separate entity or a separate approval may still support related services. The European Union, the United Kingdom, and Hong Kong leave more room when an issuer obtains extra authorization. Banks can also operate under broader prudential umbrellas than a dedicated issuer.

The study flagged a group-level gap that I think deserves more attention than the headline about “different rules.” Restrictions usually attach to the legal entity that issues the token, not to every company in the same corporate family. An affiliate could therefore do things the issuer cannot do directly. Banks already face consolidated supervision designed to catch that game. Non-bank stablecoin groups may not face an equivalent net in every market.

The authors said supervisors may need to stretch group-level oversight to larger non-bank issuers. They also noted the usual disclaimer: the conclusions are theirs and do not automatically speak for every central bank in the network. Still, the direction of travel is clear. If the product sits next to payments, the group chart will get read as carefully as the white paper.

Reserves, Treasuries, And A Two-Way Economic Shock

Here is where the story stops being a philosophy seminar and starts touching funding markets. Payment stablecoins, at least under the U.S. statute, must hold one-for-one reserves. Eligible assets include cash, deposits, repurchase agreements, and Treasury securities with remaining maturities of 93 days or less. That design is meant to keep the token close to a dollar. It also turns issuers into large buyers of short-term government paper.

Treasury officials have argued that growth in these tokens could deepen international demand for dollars and for U.S. government debt. When the statute became law in July 2025, the Treasury secretary called stablecoins a revolution in digital finance that could generate extra Treasury demand. De Cos did not dismiss that channel. He accepted that stablecoins could lower government borrowing costs, especially when demand comes from outside the United States. Foreign users can add buyers of bills rather than merely reshuffle existing domestic buyers.

Then comes the other blade. If households move balances from bank deposits into stablecoins, banks can lose a cheap and relatively stable funding source. Issuers may park some of that cash back at banks as wholesale deposits. Wholesale money is usually more concentrated and more rate-sensitive. Banks can respond by charging more for loans or by holding more liquid assets. Smaller lenders feel that squeeze first because they lean harder on retail deposits. Credit then gets more expensive for households and small firms.

Perhaps the most interesting aspect is how ordinary this risk looks once you strip away the jargon. It is a deposit-substitution story with a digital wrapper. Money leaves a sticky account and returns as a hotter one. Anyone who lived through wholesale-funding stress already knows the plot.

Redemption Waves And Short-Term Market Pressure

Reserve design also creates contagion paths. A rush of redemptions can force an issuer to sell bills or yank large bank deposits. Those flows can hit short-term funding markets when everyone else is already hunting cash. These are scenarios, not forecasts. Modeling cited in the speech found a modest overall economic effect. The result depends on reserve mix, the stock of government debt, and whether demand is domestic or foreign.

Modest is not the same as harmless. Modest average effects can hide sharp local pain. A small lender in a deposit-heavy region does not live inside an average. I would rather see that distinction in every official paper than another paragraph about innovation in the abstract.

Two-way shock in plain language:
  More token demand can lift bill buying.
  More deposit flight can lift bank funding costs.
  Net effect depends on who is buying and what sits in reserve.

How U.S. Implementation Is Tightening The Screws

Rulemaking did not freeze after the statute passed. In April, Treasury proposed anti-money-laundering and sanctions rules that would treat permitted payment-stablecoin issuers as financial institutions under the Bank Secrecy Act. The proposal would require systems that can block, freeze, or reject transactions when the law demands it. That is the moment a token stops being only a market product and becomes a regulated payment actor with operational chores.

Some readers will call this banking-by-another-name. They are not entirely wrong. If a product wants to sit next to payroll, invoices, and retail checkout, it inherits the ugly parts of that job. Screening is one of those parts. So is the ability to stop a transfer when a sanctions list updates at an inconvenient hour.

Does that kill the original crypto pitch? It depends which pitch you mean. Permissionless transfer between strangers is a different product from a regulated payment dollar with attested reserves. Officials are now writing rules for the second product. The first product will keep existing in corners of the market. Pretending those two products are the same thing is how conversations go in circles.

Tokenized Deposits Still Have Their Own Homework

It would be lazy to crown tokenized deposits as a finished alternative. They are digital representations of commercial-bank deposits on programmable infrastructure. The claim remains a bank claim. That is the institutional advantage: capital rules, liquidity rules, resolution tools, supervision, and customer-protection frameworks already exist. Settlement through central bank money can preserve equal value across institutions.

The technical list is less flattering. Separate bank networks can become walled gardens with trapped liquidity. Smaller institutions may struggle with build costs and with network effects that favor the largest balance sheets. Always-on transfer can also accelerate a run. If money can leave at 3 a.m. on a Sunday, liquidity facilities need a Sunday personality. Legal questions remain around settlement finality, smart-contract enforcement, and the boring but vital problem of reversing a mistaken payment.

1. Keep the deposit as a bank liability rather than a separate issuer claim.
2. Settle the interbank leg in central bank money so par value holds.
3. Open the networks enough that liquidity is not trapped on one rail.
4. Build after-hours liquidity tools before 24/7 transfer becomes default.
5. Write legal rules for finality and error correction before scale arrives.

Project Agorá is the live experiment many officials point to. Seven central banks and more than forty private institutions have tested cross-border settlement using tokenized commercial-bank money and central bank reserves. Work moved from prototypes toward real-value testing in 2026. Trials are not a product launch. They show that the architecture can move value. They do not show that it can replace card networks, instant-payment schemes, or correspondent banking next year.

So the official position is institutional preference, not product maturity. Stablecoins already have wider public-chain distribution. Tokenized deposits keep a closer tie to regulated money. One has users. The other has a rulebook that supervisors already know how to read.

The Everyday User Still Meets Friction First

Strip the speech down to a kitchen-table example. You get paid in Token A. Your landlord accepts only Token B. A small exchange spread appears. Then a network fee. Then a bridge delay. Then a moment when Token B trades at 99.7 cents. Nobody writes a speech about that afternoon. People just switch back to a bank app.

That is what payment credibility means in practice. It is not a white-paper diagram. It is whether a stranger accepts your balance without checking a price chart. Cards already cleared that test. Instant bank payments cleared it in many countries. Stablecoins still trip over brand differences and market discounts when the user is not a trader.

Could that change? Yes. Uniform reserve rules, tighter redemption rights, and better interchange between tokens would shrink the gap. Some of that work is already in statute. Some of it still depends on market structure. A token that is easy to mint is not automatically easy to spend.

What Issuers, Banks, And Policymakers Do Next

Regulators now have to turn principles into operating manuals. In the United States, agencies continue to implement reserve, licensing, sanctions, and AML provisions. Other jurisdictions will keep applying their own frameworks. Differences among the five markets studied can push groups to pick a legal home with broader permissions. That is not a scandal. It is how international finance works when rules are uneven.

The comparative study suggests supervisors will look harder at entire corporate groups, especially when affiliates offer lending, staking, trading, or custody around an issuer. For central banks, the next phase is more tokenized-settlement testing plus shared technical and legal standards. Stablecoins are unlikely to leave the room. The official bet is that they occupy specialist roles under rules that support redemption, transparency, and integrity.

The fight is not whether digital dollars exist. The fight is which digital dollar still behaves like a dollar after midnight, across borders, and under stress.

Banks face a strategy choice they cannot postpone forever. Some will tokenize their own deposits and try to keep customers inside the existing franchise. Some will partner with issuers and treat tokens as a distribution channel. A few large names are already circling both options. I do not see that as hypocrisy. I see it as balance-sheet math. If cheap deposits start walking, the bank either builds a rail or rents one.

Issuers face a different squeeze. The more they look like payment companies, the more they inherit payment-company duties. Attestations get stricter. Affiliate structures get less cute. Redemption at par becomes a promise that must survive a bad week, not a good quarter. That is expensive. It is also how a token graduates from market instrument to money-like instrument.

A Practical Way To Read The Next Twelve Months

Ignore the loudest forecasts. Watch four signals instead. First, whether major tokens trade through stress without lasting discounts. Second, whether group-level supervision actually lands on non-bank issuers. Third, whether tokenized-deposit pilots connect more than two banks in more than one currency corridor. Fourth, whether bank funding costs rise in markets where retail balances migrate fastest.

Those signals are dull. They are also how payment systems get judged. A viral launch does not settle a payroll file. A polished dashboard does not freeze a sanctioned wallet. A speech at a mountain symposium does not finish a standard. But speeches do reveal the hierarchy officials want. Right now that hierarchy puts tokenized bank money at the center and privately issued tokens in a regulated side room.

Will users accept that hierarchy? Many already use tokens because they settle faster than correspondent banking or because they want dollar exposure without a local bank account. Those use cases do not disappear because a central-bank coordinator prefers another design. They do, however, meet thicker rules as soon as they touch domestic retail payments.

The Questions People Keep Asking

Why question stablecoins as everyday money? Because they can trade away from par, sit on fragmented chains, and make consistent integrity controls harder. Those limits make universal acceptance and final settlement harder to guarantee.

What is the difference between a stablecoin and a tokenized deposit? A stablecoin is generally a liability of a private issuer backed by reserve assets. A tokenized deposit remains a commercial-bank liability and settles through the regulated banking system.

Could stablecoins lower U.S. borrowing costs? They could add demand for short-term government securities, especially if foreign users drive adoption. The size of any saving remains uncertain and can be offset by higher private-sector funding costs.

Is anyone calling for a ban? No. The argument on the table is coexistence with defined roles. Specialized activity under transparent rules is the official compromise, not extinction.

Are tokenized deposits available at global scale today? No. Pilots exist. A fully interoperable multi-bank and cross-border network at global commercial scale does not.

Where This Leaves The Broader Market

Crypto markets often treat official skepticism as an attack on the entire asset class. That reflex wastes time. Bitcoin can be a scarce asset without being a checkout button. A smart-contract platform can clear loans without replacing central-bank settlement. A dollar token can be useful in trading and remittances without winning the “everyday money” contest in 2026.

The credibility test is narrower and, frankly, more adult. Can the instrument stay at one unit of the currency through stress? Can two strangers complete a payment without a conversion ritual? Can supervisors see enough to enforce integrity rules without pretending they see everything? Until those answers get cleaner, officials will keep pointing to bank tokens as the safer core.

I keep coming back to a simple bias of mine. Payment systems are judged on boring Tuesdays, not on launch days. If a tool needs an explanation every time it crosses a border or a brand, it is still an instrument. Money is the thing you stop explaining. Stablecoins have not reached that quiet status at scale. Tokenized deposits have the legal quiet, but not the network quiet. Both sides still have work. The speech just made the grading rubric public.

Watch the rulebooks, the reserve books, and the interbank experiments. The next chapter will not be written by a slogan. It will be written by whether a payment still looks like one dollar after it has moved, sat in reserve, and come back to a person who never asked to become a market analyst.