Have you ever looked at a savings balance and wondered how much of that number actually sits in cash, ready to be handed back the moment you ask? Most people never do. They treat a bank statement like a locked box. It is not. It is a promise, and promises have fine print. That is the uncomfortable thought Tether’s chief executive pushed into the open after a major global banking forum argued that tokenized bank deposits should sit at the center of digital money, not fully reserved stablecoins.
I have followed this argument for years, and I keep coming back to the same gut check. If two products both claim to be “a dollar on a chain,” why would anyone pick the one that still leans on fractional reserve lending? That question is not a slogan. It is the whole fight. Stablecoin issuers want money that is parked in liquid assets. Banks want money that can still be lent out. Regulators want something that looks familiar enough to supervise. Savers, if they ever get a clear explanation, may want something else entirely.
Why This Stablecoin Fight Suddenly Matters
The latest flare-up started when the head of the Bank for International Settlements told a high-profile economic gathering that stablecoins still fall short of what money needs at scale. Redeemability at par. Interoperability. Financial integrity. Monetary sovereignty. Those were the headlines. Tokenized deposits, he said, keep commercial-bank liabilities inside a system that already settles through central bank accounts. In that model, one bank’s digital dollar and another bank’s digital dollar can still meet at par because the old plumbing is still there.
Paolo Ardoino did not buy the framing. His reply was blunt. If stablecoins can hold reserves in highly liquid instruments such as U.S. government debt, why should a saver park cash in a product that only keeps a slice of deposits in ready money? He put it in colorful language: the emperor, in his view, is standing there without clothes. Banks have lived with that exposure for generations. Crypto-native issuers are now pointing at it in public.
Why would someone choose to put savings into a fractional reserve product while stablecoins are fully reserved?
– Tether chief executive, commenting on the tokenized deposit case
That line is doing a lot of work. It is not just marketing. It is a claim about risk, liquidity, and who gets to use your money while it sits still. I find the honesty refreshing, even if the politics around it are messy. Because once you admit that a deposit is also raw material for loans, the comparison with a ring-fenced reserve portfolio stops being academic.
What Tokenized Deposits Actually Are
A tokenized deposit is still a bank liability. The coin on the ledger is a representation of money the bank already owes you. The funds do not leave the bank’s balance sheet just because they now have a blockchain wrapper. They can still support lending. That is the feature banks like, and it is the feature critics treat as the catch.
Think of it this way. You hand a bank a hundred dollars. The bank keeps a buffer and puts the rest to work. If that same hundred later appears as a token that can move at midnight between corporate treasuries, the economic story has not changed. The token is programmable. The liability is familiar. Settlement can still run through accounts at the central bank, which is how the “singleness” of money is supposed to survive. Different banks, same par value, same redeemability story.
Stablecoins work on another logic. The issuer takes in fiat, or the equivalent, and holds a reserve book. In the version Tether’s leadership likes to advertise, that book leans on cash and short-dated government paper. The token is not, in theory, a claim on a loan book. It is a claim on a pile of liquid assets. Whether every issuer lives up to that picture is a separate, and very real, debate. The design goal is different even when the user interface looks similar.
I’ve found that this distinction gets lost in conference slides. People hear “on-chain dollar” and stop listening. The legal wrapper, the reserve mix, the redemption desk, the chain the token lives on, and the wallet that holds it all change the risk. A token that never leaves a permissioned bank network is not the same animal as a token that can jump from a self-custody wallet on one public chain to a bridge and then to another network. Both can be called digital money. Only one of them behaves like a bank deposit wearing a new jacket.
The Case Regulators Keep Repeating
The official critique of stablecoins is not one complaint. It is a bundle. First comes redeemability. If two branded dollars trade against each other in a secondary market, their prices can slip off one dollar, especially when markets get jumpy. A user holding one token who needs to pay someone who only takes another token may have to swap first. That swap is not the same as a central-bank-settled transfer between two commercial banks.
Second comes interoperability. Public chains multiply venues. Bridges, wrapped versions, and self-custody all make the same nominal asset harder to treat as one instrument. Third comes financial integrity. Moving value across open networks and unhosted wallets complicates the consistent application of anti-money-laundering and counterterrorism rules. Fourth comes sovereignty. Dollar-linked tokens used heavily outside the United States can look like digital dollarization. Local policy tools get weaker if households and firms start thinking in another country’s unit of account.
None of those points are imaginary. Anyone who watched a peg wobble already knows secondary markets can be rude. Anyone who has waited on a bridge transaction knows “the same coin” is sometimes a cousin, not a twin. And anyone who works in compliance knows an open mempool is not a core banking ledger.
Still, I keep noticing what the critique quietly assumes. It assumes the commercial banking system’s par clearing is the gold standard, and everything else is a patch. That may be true for wholesale payments inside rich economies. It is less obvious for a merchant in a market where the local bank rail is slow, expensive, or simply closed after 4 p.m. Context matters. The same product can look sloppy in one city and indispensable in another.
The Reserve Question Banks Would Rather Soft-Pedal
Ardoino aimed at the reserve structure because that is where the emotional punch lives. A fully reserved token, if the reserves are real, short, and disclosed, is a narrow product. It is closer to a money-market wrapper than to a bank. A deposit is a wide product. It funds mortgages, credit lines, and payroll loans. Society needs that credit engine. Savers, on a bad Tuesday, may not want to be the fuel.
Fractional reserve banking is not a scandal. It is the operating system of modern credit. The scandal, if there is one, is pretending a tokenized deposit is “just like cash on-chain” without saying the credit transformation is still happening in the background. When customer funds stay on the issuing bank’s books, they can still be put to work. When funds move into a separate reserve portfolio of bills and overnight cash, they generally cannot support the same loan book.
That trade-off is the heart of the policy fight in the United States right now. Banking groups have told lawmakers that rewards on stablecoin balances could pull deposits out of banks, especially community lenders. Less deposit funding, in their telling, means less capacity to extend credit. A large bank chief repeated a version of that warning while backing market-structure legislation. The concern is not theoretical to them. Deposits are cheap, sticky funding. Tokens that feel like checking accounts with extra yield look like competition.
There is a legislative wrinkle worth keeping straight. One recent statute bars payment stablecoin issuers from paying interest or yield directly to holders. Platforms and service providers can still design reward programs around the edges, depending on structure. That gray zone is why lobbyists are still arguing over a single section of a broader bill. The legal sentence is narrow. The business incentive is wide.
How Banks Are Building Their Own On-Chain Answer
This is not a thought experiment anymore. Large U.S. banks have been working on a shared deposit token network through an industry clearing utility, with a target window in the first half of 2027. The first users are expected to be multinational firms that want programmable treasury moves and faster cross-border settlement. That is a very bank-shaped beachhead: wholesale, permissioned, familiar clients.
The global messaging network used by banks has also tested a blockchain shared ledger with a group of major international lenders. The design centers on tokenized deposits for round-the-clock cross-border payments. Again, the pattern is the same. Keep the liability inside the banking perimeter. Add a ledger that does not sleep. Sell speed and programmability without giving up the deposit franchise.
A smaller pair of banks has tried a hybrid. Their dual-purpose token is meant to behave like a deposit while it stays inside a closed network and like a stablecoin if it leaves that garden. Testing on a public smart-contract chain has been underway ahead of a late-2026 target. I am skeptical of hybrids until they survive a real redemption crunch, but the instinct is easy to read. Banks want optionality. They do not want to concede the open-network use case forever.
Even the official case for tokenized deposits admits the model is incomplete. There is still no multi-bank, cross-border ecosystem that issues these tokens through a fully interoperable framework. Most live systems sit on permissioned rails. Some designs look a lot like bank-issued stablecoins with extra adjectives. Coexistence was floated as the grown-up outcome: tokenized deposits for everyday payments, stablecoins for more specialized jobs. That split sounds tidy. Markets are rarely tidy.
| Feature | Fully reserved stablecoin | Tokenized bank deposit |
| Legal nature | Claim on a reserve portfolio | Liability of a commercial bank |
| Where the cash sits | Separate liquid assets, often bills and cash | On the bank balance sheet |
| Credit creation | Limited by design | Still supports lending |
| Typical venue | Public chains and mixed custody | Permissioned bank networks first |
| Par settlement story | Issuer redemption plus secondary markets | Bank-to-bank settlement via central bank money |
Deposit Flight Is the Political Fuse
Call it deposit flight, disintermediation, or “people voting with their wallets.” The mechanism is simple. If a token feels safer, more portable, or better rewarded than a checking account, balances move. Issuers then recycle a large share of those inflows into government securities. Sovereign borrowers may enjoy firmer demand for short-term paper. Banks may face a more expensive funding mix. Households and firms could eventually see that cost in loan pricing.
That two-sided effect showed up in the same speech that promoted tokenized deposits. Stablecoin reserve demand can ease government funding. Deposit leakage can tighten bank funding. Both can be true at once. Policy then becomes a choice about which friction you prefer: a smaller bank loan channel, or a larger private market for tokenized cash that sits outside traditional credit transformation.
Ardoino flipped the moral of the story. What happens, he asked, if people decide stablecoins are the safer asset class and move savings accordingly? His punchline was that the system is already in the “find out” phase. It is a taunt. It is also a forecast. Once a critical mass of users treats on-chain dollars as the default store of value in a given corridor, the old deposit base does not automatically win on inertia.
USDT remains the largest stablecoin by circulation and has a deep footprint outside the United States. The company’s leadership has spent years pitching it as a dollar savings and payments tool in places where local banking is thin or where access to dollars is a daily headache. Investments in remittance and payments firms that serve African and Asian corridors fit that map. If those corridors already clear invoices and family transfers in a dollar token, a lecture about monetary sovereignty arrives late.
Where the Official Story Gets Uneasy
Monetary sovereignty is the argument that travels well in official rooms. If a foreign-denominated token becomes the unit people actually price goods in, local interest-rate moves land with less force. Import prices, wage talks, and capital flows start answering to another center. That is not a crypto-only problem. Dollar cash already plays that role in more than a few economies. Tokens make the same habit faster, cheaper, and harder to reverse.
I do not shrug that off. A country that loses control of its unit of account is not just losing a symbol. It is losing a tool. At the same time, scolding users who grab a more stable unit during inflation or capital controls has a hollow ring. People optimize. They always have. If the official money is leaky, a parallel dollar will show up, whether it is a paper note in a mattress or a token in a phone.
Interoperability is the more technical sore spot. A payment system that depends on exchanging brand A for brand B in a stress window is not “money” in the textbook sense. Fair. The banking answer is to keep tokens inside a club where par is enforced by membership rules. The crypto answer is to improve reserves, redemption, and market-making until deviations stay tiny. Both answers can fail. Club systems can freeze outsiders. Open systems can gap in a panic.
Perhaps the most interesting aspect is how often both camps accuse the other of dressing up old risks in new software. Banks say open stablecoins import run risk and compliance gaps. Issuers say tokenized deposits import the very maturity mismatch that made classic banking fragile. Listen long enough and you hear two risk departments talking past each other, each certain their ledger is the adult in the room.
A Closer Look At “Singleness” Of Money
Singleness is the idea that a dollar is a dollar, no matter which private name sits on the instrument, because conversion at par is reliable and cheap. Cash, insured deposits, and central bank balances are supposed to form one pile. If a token trades at 98 cents whenever markets sneeze, singleness is cracked. If two bank tokens always clear at one dollar through the same settlement agent, singleness holds.
That is a clean theory. Practice is lumpier. Even inside banking, not every claim is treated as identical. Uninsured balances, foreign-branch deposits, and money-market funds already live in a hierarchy. Stablecoins add another rung. Tokenized deposits try to stay on the insured-or-at-least-familiar rung while borrowing the speed of a token. The branding war is about which rung gets to call itself money rather than a money-like asset.
In my experience, users do not recite singleness over coffee. They ask three plainer questions. Can I get out at one dollar today? Will the app work on Sunday? Who eats the loss if the issuer or the bank stumbles? Official speeches answer the first with settlement architecture. Product teams answer the second with uptime. Lawyers answer the third with footnotes. The public only sees the number on the screen.
Rewards, Yield, And The Quiet Product War
Strip away philosophy and you get a product war. Banks sell safety, credit access, and a web of services. Crypto platforms sell 24-hour movement, global reach, and, where rules allow, some form of reward. Direct issuer yield is constrained in the U.S. payment-stablecoin lane. Indirect rewards are the battlefield. Points, fee rebates, partner yield, and structured programs all try to look like interest without wearing the costume.
- Banks fear cheap deposits migrating to token balances that feel like accounts.
- Issuers fear rules that freeze rewards while leaving bank perks untouched.
- Lawmakers fear being blamed for either a credit crunch or a messy run.
- Users fear waking up to a haircut, a freeze, or a 19th-century wire fee.
Community lenders make the sharpest political case because their funding is local and their loan books are local. If a national token vacuum pulls balances toward a handful of reserve managers buying bills in one money center, the map of credit can shift. That argument can be overplayed. It can also be underplayed. Credit is not evenly sprinkled from the sky. It follows funding.
On the other side, treating every stablecoin reward as a systemic theft of deposits ignores why some customers leave. Fees. Hours. Cross-border friction. Distrust after past bank failures. A token that settles in minutes to a family abroad is not only a yield toy. Sometimes it is the first dollar instrument that actually works for that household.
Liquidity During Stress Is The Real Exam
Pretty reserve slides do not matter if the exit door jams. A stablecoin that is “fully reserved” in 30-day paper can still struggle if everyone redeems on the same afternoon and the market for that paper gaps. A tokenized deposit that is “just a bank liability” can still struggle if the bank is the one under pressure and depositors want out faster than loans can be called.
This is where I get less romantic about both models. Liquidity is a behavior, not a label. Treasuries are liquid until they are not, for a few ugly hours. Bank par conversion works until confidence cracks and the central bank has to stand behind the system. Public-chain tokens move until gas spikes, a bridge pauses, or an issuer gates redemptions. Anyone selling a frictionless digital dollar is selling weather in a climate that still has storms.
The honest design conversation is about buffers. Cash tranches. Staggered maturities. Transparent attestation. Predictable redemption windows. Clear priority in insolvency. On the bank side, capital, deposit insurance, and access to official liquidity. On the token side, bankruptcy-remote reserves and operational redundancy across chains. Without those, “fully reserved” and “licensed bank” are both slogans.
Programmability Without The Fairy Dust
Both camps sell programmability. Conditional payments. Escrow without a human clerk. Treasury sweeps that run while the back office sleeps. That part is real and useful. A corporate treasurer who can move value with rules attached is not chasing a meme. She is cutting operational drag.
The fairy dust appears when programmability is treated as a substitute for trust. Code can move a token. Code cannot invent a buyer for a reserve asset at a fair price during a fire sale. Code cannot force two brands of dollar to trade at par if the market has lost faith in one issuer. I like programmable cash. I do not like the habit of using the word “smart” as if it abolished credit risk.
Permissioned bank networks will likely win the first wave of large corporate flows because the counterparties already know each other. Public-chain dollars will keep winning corridors where the counterparties do not share a clearing membership. That split can last a long time. It does not require anyone to be a villain.
What Savers Should Actually Compare
If you strip the speeches down to a household checklist, the comparison gets practical. Not ideological. Not tribal. Just a set of questions that should be asked before anyone moves rent money onto a new rail.
- Who is legally on the hook when you redeem, and under which insolvency rules?
- What sits in the reserve or the bank book, and how fast can it be sold?
- Is there deposit insurance, or only a claim on assets?
- Can you exit at par on a weekend, or only through a secondary market?
- Which networks and wallets can hold the token without extra wrapping?
- What happens to rewards if the rulebook tightens next quarter?
Those six points sound dry. They are the whole product. Marketing will talk about speed and inclusion. Fine. Speed is worthless if the last mile is a discount to par. Inclusion is incomplete if the token cannot be spent where the user actually lives. I would rather a slightly slower instrument that redeems cleanly than a dazzling one that needs a specialist to unwind.
There is also a behavioral trap. People treat any instrument labeled “stable” as cash. They then lever it, park it in a protocol, or use it as collateral. Stability of the peg is not the same as safety of the wrapper around the peg. A reserved token inside a risky venue is still a risky stack. Banks know this. Crypto desks know this. Casual users often learn it the expensive way.
The Coexistence Line Sounds Nice Until Volumes Arrive
Officials like coexistence because it postpones a choice. Tokenized deposits for daily pay. Stablecoins for niche jobs. Maybe that is how a calm decade would look. I am not convinced volumes will respect the seating chart. If one rail is cheaper for remittances, that rail gets remittances. If one rail is easier for trading venues, that rail gets trading balances. If one rail is the only option after local banks throttle a customer, that rail gets the customer.
Scale changes politics. A niche token used by traders is a market-integrity file. A mass-market token used for payroll is a monetary-policy file. We are already watching that migration in some regions. Once merchants quote prices in a token and workers accept wages in it, the “specialized function” story is outdated. At that point, regulators do not get to assign roles like a theater director. They get to supervise a fact on the ground.
Banks know this, which is why they are building tokens of their own rather than writing stern letters and calling it a day. Issuers know this, which is why they keep repeating the reserve argument in language ordinary savers can repeat. The next two years of rulemaking, bank launches, and redemption testing will tell us which story survives contact with actual flows.
A Few Myths Worth Dropping
Myth one: a blockchain automatically makes money safer. It does not. It makes transfer and audit trails different. The asset behind the token still has to be good.
Myth two: a banking charter automatically makes a token conservative. Charters come with supervision, which is valuable. They also come with loan books, which is the point of banks and the source of run dynamics.
Myth three: interoperability is a software upgrade away. Legal identity, sanctions screening, and par conversion are institutional problems. Bridges can move bits. They cannot invent a shared bankruptcy court.
Myth four: savers will stay put out of habit. Some will. Enough will not, especially younger users who already treat the phone as the branch. Habit is a moat until a better default appears.
What happens to the financial system if people start realizing that stablecoins are safer and move their savings into the better asset class?
That question is loaded, obviously. “Safer” depends on the issuer, the reserve, the chain, and the custody setup. But the provocation works because it names a comparison banks rarely invite. Not “crypto versus the dollar.” The dollar is on both sides. The comparison is reserved digital cash versus lendable digital cash.
Where I Land After The Noise
I do not think tokenized deposits are a trick, and I do not think every stablecoin is a public utility in waiting. The useful frame is narrower. If the job is wholesale settlement among regulated firms that already hold bank accounts, a deposit token on a shared permissioned rail is a natural evolution. If the job is portable dollar access for users who are poorly served by local rails, a reserved token on open networks is going to keep winning users whether or not a symposium likes the optics.
The reserve argument is the one that will stick with the public, because it is easy to tell at a dinner table. One product tries to keep a matching pile of liquid assets. The other product uses your balance as part of a loan machine. Both can be legitimate. They should not be sold as twins. When officials praise singleness, they should also say out loud that singleness in the banking system is backstopped by institutions most token holders do not get to vote on.
Will savers really migrate at scale? Some already have, in specific corridors. A full-scale shift inside deep banking markets would take a mix of better products, clearer law, and a shock that makes the old default look foolish. No one should root for the shock. People should still read the balance-sheet story before they treat a new ticker as cash.
The next test is not another speech. It is whether multi-bank deposit tokens can move across institutions without becoming isolated islands, and whether large stablecoins can keep redemption smooth when volatility returns. If both pass, coexistence stops being a slogan and starts being a market structure. If one fails, the other will inherit the flows, and the losing camp will say they warned us.
Until then, the uncomfortable question stays on the table. If you could hold a dollar that is meant to sit in liquid paper, or a dollar that is meant to help fund someone else’s loan, which one do you want for the slice of money you cannot afford to negotiate over? That is not a gotcha. It is the product decision hiding under all the diplomacy. And it is why a single reply from a stablecoin chief landed harder than another polished case for keeping digital money inside the familiar bank perimeter.
