What the Broken Counting Rules Really Mean
The subject of this final judgment is the apparent break in the M1/M2 counting rules when money-like claims move onto blockchains and across borders. The evidence points to a more precise conclusion: the categories remain useful, but tokenized deposits, tokenized money-market funds, and payment stablecoins expose assumptions that were easier to leave implicit when most relevant claims arrived through domestic bank and fund reporting. The right response is not to count every token as new money or to dismiss every token as mere technology. It is to connect function, liability, conversion, backing, reporting, and geography before drawing a monetary or policy conclusion.
Five questions keep the categories intelligible
The first question is what function dominates. A claim used mainly for immediate payments is closer to a medium of exchange and an M1-type balance. A claim held mainly as a short-term investment, with a return and a redemption step, is closer to a store of value and non-M1 M2. This distinction is not a judgment about whether one use is more valuable. It is a way to make the aggregate describe the economic role that matters for spending and liquidity.
The second question is who owes the holder what, and what backs that obligation. A tokenized deposit remains a bank deposit on a bank’s balance sheet. A tokenized money-market fund is a share in a fund portfolio. A payment stablecoin is a privately issued, dollar-linked token supported by reserve assets and a redemption promise. The three can all be called digital dollars while exposing the holder to different institutions, protections, and failure surfaces. Legal relationship does not replace functional classification, but without it the statistic can mistake a bank liability, an investment share, and a private issuer claim for one object.
The third question is what happens between transfer and spending. A demand-style tokenized deposit can be immediately converted into cash without selling an investment or waiting for a fund redemption. A tokenized money-market share can transfer instantly while its holder still waits, typically one to two business days, to receive dollars. A payment stablecoin can move peer to peer continuously on a public blockchain, while its practical route back to dollars depends on redemption terms, intermediaries, and market access. The speed of the middle step cannot answer for the whole conversion path.
The fourth question is whether the balance can be reported without counting the backing twice. Existing bank reporting already includes tokenized deposits with traditional deposits, and the Federal Reserve note identifies no double-counting risk from their bank-asset backing. Routine money-market-fund reporting can include the fund category without identifying the tokenized subset; bank deposits in fund portfolios create a small potential overlap with M1. Payment stablecoin reserves may include bank deposits or money-fund assets already represented in M1 or M2, so the treatment depends on composition rather than on a generic claim of one-to-one backing.
The fifth question is where the activity occurs. The U.S. monetary aggregates are not simply a census of all dollar-denominated claims in the world. Deposit-related components are defined around holdings at banks in the United States, while physical currency can circulate internationally. A payment stablecoin issued by a U.S.-regulated entity can circulate globally, and reserve reporting may show issuance without identifying the location or economic purpose of every holder. A global on-chain balance and a U.S. money-stock component are different measurements.
Taken together, these questions answer what is and is not broken. M1 and M2 are not invalidated by a new ledger because they were never meant to classify ledgers. They classify safe, stable claims by liquidity and use. What has become harder is the assumption that the relevant claim, its holder, its reserve, and its location all appear in the same domestic reporting system. Tokenization can separate those elements. A bank deposit may stay inside the old category while its new rail changes how it circulates. A fund share may move faster without becoming cash. A stablecoin may perform an M1-like payment function while its backing and users span several jurisdictions.
This is also why a new aggregate line called “tokenized money” would not solve the problem. It would group claims that differ in economic function and could obscure the reason each balance is counted. It could also invite double-counting by adding a token to an aggregate that already includes the deposit or fund assets supporting it. Better measurement may require additional indicators of tokenized form, use, or adoption, but an additional indicator is not the same as a new dollar of money.
The distinction matters for interpretation as much as for arithmetic.