Interest payments have outranked defense across most of the G7 since 2024

Divide one by forty and the United States is paying roughly 2.5% on its stock ^[1][2][8], which is less than half of the 5.33% the long bond asked in August ^[3], and that gap is the mechanism at work here: a budget line does not move when the yield moves, it moves when the paper matures. Or rather, the more careful version of that, since 2.5% is an average across bills, notes and bonds while 5.33% is the most expensive single point on the curve, which means the 2.1x you get by setting the two side by side ^[9] flatters the case. The multiple is too big and the direction it points is the one the Treasury actually faces.

The aggregate does the compounding. A $4.2 trillion increase to reach the projected end-2026 stock implies a $71.6 trillion starting point and a 5.9% rise in the stock in a single year ^[18], and the debt-to-GDP ratio in the same projection backs out developed-market output somewhere near $72.9 trillion ^[19]. Carry that whole stock at the 30-year's yield and the annual bill is about $4.0 trillion, roughly 5.5% of that output ^[10], which nobody should read as a forecast (most of that paper is neither American nor thirty-year) and which does bound where a decade of rollovers is heading.

Probably wrong on timing, but the crossover is the durable fact in this compilation, more than any single print: interest has run ahead of defense in most G7 members since 2024 ^[14], and a claim that grows through a maturity schedule is harder to legislate down than one that needs an annual appropriation, so a rearmament programme is now argued against the fastest-growing line on the same revenue instead of against discretionary slack. Italy's path toward about 9% of revenue by 2028 ^[13] is the clean version, roughly one euro in every eleven ^[20] committed before any minister opens the argument, and Italy is the eurozone's third-largest economy with a record of testing the currency union ^[7]. Japan is on the same curve, with long yields approaching 30-year highs ^[6].

The counter-thesis sits inside the same number. cryptobriefing.com's read is that a 5.33% long bond is a genuine alternative to equities for the first time in over a decade and that risk-averse institutions will take it ^[17], which is also to say that a yield that high is a bid, and a bid means auctions clear, which frames this as a repricing of fiscal headroom, not a funding shortfall. It reads differently if long yields retreat and the average never meets the market, if nominal growth runs above the effective rate and the ratios flatten without anyone cutting, or if the national estimates get revised down as briskly as they were revised up ^[11][12]. All of it is one publisher's compilation of national figures, which is worth holding at arm's length.