I picked up a few red flags in the post concerning governance, stablecoins and more. Let’s dissect it.
The two events are real. The “same-day both-sides-of-the-market” reading isn’t.
The $12.5B buyback was a scheduled cash-management op in the front end, announced the day before, inside a program Treasury has run since 2024.
Those purchases are usually funded by issuing other Treasuries, mostly bills. That is a liability swap, not a shrinking of the $35T stock, and it is far too small to set the cost of debt. The Fed funds rate, term premium, and the deficit still do that.
GENIUS also does not require “one stablecoin = one Treasury bond.” Permitted reserves include cash, Fed balances, insured demand deposits, paper with 93 days or less left, overnight Treasury repos, and government MMFs. That is T-bill demand at the front of the curve, not a standing bid for the whole book.
The $1.4T JPMorgan figure is a high-end dollar-demand scenario, not a committed coupon-buying program. And JPMorgan is not even in the 21 (next👇 post). Existing USDT/USDC reserves already sit in bills. A bank coin can reshuffle those holdings, or pull deposits off bank balance sheets that already buy Treasuries. Gross market cap is not net new demand.
The product is still unnamed, unchained, and targeted for H1 2027. Several of the same banks are also building tokenized deposits that keep the money on their own books.
The interesting ledger is the net one: how much is new foreign conversion into short bills, how much is recycled Tether/Circle reserves, and how much is cannibalized deposits. Without that, “mandatory $1.4T of bond buyers plus a $12.5B buyback” is coincidence dressed as architecture.
In conclusion, the events are real but the same-day masterstroke isn’t. Watch the net flows, not the montage.
