Singleness of money: why central banks prefer deposit tokens

Singleness of money is the property that a dollar is a dollar in every form it takes: cash, a deposit at one bank, a deposit at another, all exchanging at par, all the time. It is maintained rather than given, and the mechanism that maintains it is settlement in central bank money at the interbank layer, which continually reconciles the private forms of a currency back to face value. The full definition lives in our glossary entry on the singleness of money; this page is about the argument now running through central banks, banks, and stablecoin issuers over whether new forms of digital money can keep that property, and what the answer means for the rails institutions choose.

The central bank case: bearer instruments drift from par

The canonical statement of the concern is BIS Bulletin 73, by Rodney Garratt and Hyun Song Shin, published weeks after the March 2023 weekend in which USDC traded well below its dollar peg while Silicon Valley Bank held part of its reserves. Their argument is structural. A stablecoin circulates as a bearer instrument: its price is whatever a secondary market says at that moment, and even a small seed of doubt about reserves or redemption can move that price away from par. Tokenized deposits, by contrast, do not circulate with their own price. A payment in tokenized deposits is a claim on the payer’s bank that becomes a claim on the payee’s bank, with the interbank leg settled in central bank money, exactly as deposits settle today.

The BIS sharpened the position in its 2025 Annual Economic Report, concluding that stablecoins fall short of serving as the backbone of a monetary system on three tests: singleness, elasticity, and integrity. On singleness, the report pointed to issuer creditworthiness and secondary-market exchange rates that can deviate from par. That analysis is one reason central banks have steered institutional tokenization toward deposit tokens and toward projects, like the tokenized-deposit pilots and consortium networks we map in the deposit token landscape, that keep settlement anchored in the two-tier banking system.

The industry case: regulation forces par

The rebuttal does not dispute the history; it disputes the inference. Regulated payment stablecoins, the argument runs, are claims on segregated portfolios of high-quality liquid assets with a legal right of redemption at par, and that machinery, rather than secondary-market prices, is what defines their value. The GENIUS Act requires 1:1 reserves in high-quality liquid assets for US payment stablecoins; MiCA requires e-money token issuers to redeem at par on demand. Commentators including Ledger Insights and the Open Banker newsletter have argued the BIS analysis understates this: a fully reserved, non-credit instrument should hold par more easily than a fractional-reserve bank deposit, deposits themselves have failed the par test in bank runs, and uniform regulation of issuers pushes regulated tokens toward exactly the standardization singleness requires. In day-to-day payments, regulated stablecoins are in practice accepted at face value.

The honest middle ground is narrower than either camp’s rhetoric. Depegs have happened, including to the most conservatively run coins, which is why institutions underwrite the risks rather than dismiss them. And par-redemption machinery has visibly strengthened, which is why those same institutions keep expanding their use of regulated coins where the rails they replace are slow or absent.

What it means for choosing rails

Three practical conclusions follow. First, the debate is jurisdictional: a treasury moving euros under MiCA, dollars under GENIUS, and an exotic-corridor payout face three different versions of the par question. Second, it is directional rather than terminal: central bank preference is shaping what banks build, so deposit tokens and consortium rails will keep arriving, while stablecoin reach keeps growing from the other side. Third, it is a portfolio question. An institution rarely gets to pick the instrument its counterparty holds, so the operating problem is moving value at par across all of them, with settlement finality it can prove.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenized deposits, with compliance enforced on every transaction and settlement in seconds instead of days.

Frame does not take a side in the singleness debate; it is built for the world the debate produces, where fiat rails, regulated stablecoins, and tokenized deposits all carry serious volume and none replaces the others. Frame is rail-neutral: a payment enters through one integration, and Frame routes it across whichever rail fits the corridor, the counterparty, and the policy that governs it. That policy is enforced in settlement by Frame’s Rules Engine, so a transfer that cannot satisfy its conditions, including which instruments are acceptable for which flows, does not settle. For banks, payment providers, exchanges, platforms, and enterprises, singleness stops being an argument to win and becomes a policy to enforce.

See how a rail-neutral settlement layer works: the Frame Blueprint.