Three Tokenized Cash Funds, Three Very Different Maturity Books

A tokenized cash fund is supposed to be the boring part of the portfolio: a dollar that sits at a dollar, backed by short-term government debt, redeemable on demand. Three of the most credible ones — JPMorgan's, WisdomTree's and Franklin Templeton's onchain government money funds — all fit that description on the surface. All three hold a $1 share price; all three invest in U.S. government paper. But the portfolios behind them, disclosed in each fund's monthly filing, are built quite differently — and the difference is precisely the kind of thing that stops being academic the moment one of these funds is posted as collateral.

Two numbers describe how much interest-rate and maturity risk a cash fund carries. Weighted-average maturity, or WAM, measures how quickly the fund's holdings re-price as rates move — a low WAM means the fund resets fast and barely moves when rates do. Weighted-average life, or WAL, measures when those holdings actually come due, counting to final legal maturity. For a fund that owns only short Treasury bills, the two are nearly identical. When they diverge, it is a tell: the fund holds paper whose final maturity runs well past its next reset or call date — a longer-dated instrument a money fund is allowed to carry at a short WAM. The gap between WAM and WAL is where a fund's reach for yield shows up, and it is invisible in the one maturity number most summaries quote.

JPMorgan runs it tight; Franklin runs it long

JPMorgan's onchain money fund sits at one end. Its WAM and WAL are both 13 days — no gap at all. Its filing shows a book of only Treasury bills and Treasury repo, everything maturing within about two weeks, no long tail. It is the closest thing to pure, short Treasury cash among the three: a fund built to move as little as possible and turn over fast. Franklin's onchain government fund sits at the other. Its WAM is 37 days, but its weighted-average life is 113 — a 76-day gap. The filing shows why: alongside its bills and repo, the fund holds federal-agency notes — Federal Farm Credit and Federal Home Loan Bank paper — with final maturities reaching into 2027 and 2028, carried at a short maturity but a long life. That is a deliberate reach for extra yield, and it is the reach that a 37-day headline does not show. WisdomTree's fund lands between the two at 19 and 39 days, its gap traced to a Treasury floating-rate note whose maturity outruns its reset. Three funds, three different ways to earn a little more on “cash” — legible only in the second maturity number.

Why the gap matters now

For most of their short history, tokenized cash funds were held the way you hold cash — to park a balance and earn a yield. The maturity of the underlying book was a detail for the fund's own risk committee, not something a holder had to price. That is changing fast. BlackRock's tokenized money fund is increasingly used as collateral for borrowing and leveraged trading, and issuers are filing new tokenized reserve vehicles built specifically to back stablecoins. In both roles the token stops being an end holding and becomes plumbing — something a lender advances against, or a stablecoin promises to redeem into. And plumbing is judged by what it does on its worst day.

On that worst day, the two maturity numbers stop being interchangeable. A book that matures in thirteen days can be turned into cash almost immediately at close to par; there is nothing to sell early. A book with a 113-day weighted-average life holds paper that, if the fund is forced to raise cash before it matures, has to be sold into the market at whatever price stress allows — and longer-dated agency notes move more, and trade thinner, than two-week Treasury bills exactly when everyone wants cash at once. That is what a lender is really setting when it decides how large a haircut to charge against tokenized collateral, and what a stablecoin issuer is really taking on when it parks reserves in one fund rather than another. The tokens look identical on a screen and move around a chain identically; the filings say they are not identical in the one moment that counts.

None of this makes the longer book wrong. Franklin's reach for agency yield is a legitimate design — more return on cash, well within what a government money fund may hold — and its investors are paid for it. The point is that the choice is now consequential to people who never see the filing: the borrower posting the token, the protocol accepting it, the holder of a stablecoin it backs. As tokenized cash scales from a few billion into the collateral layer of onchain finance, the market will have to price these funds by their books, not their tickers — the way repo desks already price traditional money-fund collateral. There is even a hint of where that leads: across the summer all three funds shortened their books as rates re-priced — JPMorgan's WAM fell from 26 days to 13, WisdomTree's from 43 to 19, Franklin's from 53 to 37 — converging toward the short end from very different starts. The open question is whether the funds that win the most collateral use get pulled all the way toward the tightest, most JPMorgan-like books, because the shorter the maturity, the less there is to underwrite — or whether yield keeps a longer book like Franklin's in the mix, and the market simply learns to charge it a wider haircut. Either way, the number that decides it is the one the label leaves out.