The Fed Went Quiet. The Treasury Did Not.
Chair Warsh used Jackson Hole to stop narrating the path of policy. The same week, Treasury was already managing the long end by buying it.
Follow the debt from issuance and market pricing to refinancing, interest expense, and buybacks.
A Treasury yield is the price of borrowing at the margin. The government’s interest bill is the accumulated result of many borrowing decisions made at different dates. Confusing those two things makes a rate decision look like an immediate reset of the entire debt stock.
This dossier separates the mechanisms. The yield curve tracks market prices. The bill share describes financing composition. Average interest rates and fiscal-year-to-date interest outlays show how financing decisions reach the budget. Buybacks are a separate operation whose stated purpose and execution need to be assessed on their own terms.
Read the analysis in the order below, then inspect the underlying series. The period comparison makes its observation dates explicit; the interest scenario holds the debt stock and refinancing share under your control. It is deliberately a sensitivity calculation, not a forecast of the deficit.
Two distinctions guide the coverage. A model’s term-premium estimate is not an observed yield, and models can disagree or revise their history. Likewise, the amount dealers offer in a buyback is not an estimate of the amount Treasury should accept: price, the announced cap, and eligible securities constrain the result. Those limits matter as much as the headline numbers.
Chair Warsh used Jackson Hole to stop narrating the path of policy. The same week, Treasury was already managing the long end by buying it.
Treasury has kept its bill share above the advisory committee's 15 to 20 percent range for three years. Bills now carry a higher average rate than notes or bonds, and the note stack is still repricing upward.
Since May 2024, dealers have offered $1.7 trillion of securities into Treasury's liquidity-support buybacks. Treasury has accepted $223 billion. Doubling the caps changed the ceiling, not the behaviour.
Bond yields can move while the policy rate stands still. Separating the observed move from its possible causes changes the balance-sheet lesson.
The operation dataset distinguishes securities offered, securities accepted, and the maximum amount Treasury announced. Our tracker filters for liquidity-support operations and sums par amounts by month. It does not combine them with cash-management buybacks.
Accepted divided by offered measures the share of submitted par taken up. Accepted divided by the announced maximum measures use of the operation’s capacity. Those denominators answer different questions; one ratio cannot stand in for the other.
An unfilled maximum alone does not prove that a program failed. Nor does a large amount offered prove that dealers would accept Treasury’s preferred price. Par amounts are not cash expenditure, and an offer ratio is not a direct measure of bid-ask spreads or market depth.
Assess execution and market liquidity separately. The operation analysis explains the distinction; the tracker keeps the quantities observable. A change in an announced limit is a change in capacity, not evidence of a completed purchase.
Compare two observation periods, with the units and actual dates kept in view.
See how refinancing and a rate change affect an annual interest run rate.
The first preserved data version fixes the observations used by our two Data Notes. Trackers remain current. This separation matters for model estimates that can be revised: a later revision must not silently rewrite the evidence behind a published argument.