At $40 trillion of debt, "anything but bonds" stops being a contrarian call

Federal debt stood at roughly $39.9 trillion in mid-August and was expected to cross $40 trillion as early as that week ^[1]. Bank of America chief investment strategist Michael Hartnett expects it to reach $50 trillion by 2029 ^[3], which turns his "Anything but Bonds" framework from a tactical opinion into the arithmetic every allocator now has to answer for.

The framework is about supply, not solvency. Hartnett's argument is that persistent deficits force continuous refinancing and new issuance, that investors want compensation for absorbing it, and that this makes long-duration Treasurys unattractive relative to other assets ^[2]. The feedback loop is mechanical: more debt means more interest, more interest widens the deficit, and the Treasury issues more securities to cover it ^[12].

The flow numbers show the scale. The government borrowed $1.8 trillion in the first 10 months of fiscal 2026, including $432 billion in July alone ^[5]. That is an average of about $180 billion a month, with July running roughly 2.4 times that pace ^[2], and an annualised run rate near $2.16 trillion ^[3]. The interest bill has climbed to roughly $1.4 trillion over the past year, according to Hartnett's latest outlook ^[6] - about 78 cents of interest for every dollar of new borrowing over the same stretch ^[6].

The repricing problem is visible in the gap between what the debt costs today and what new debt costs. A $1.4 trillion interest bill against $39.9 trillion outstanding implies an average effective rate of roughly 3.5% ^[4]. The 10-year Treasury yield reached 4.6% and the 30-year hit 5.2% ^[4], so long-dated refinancing is landing about 170 basis points above the average cost of the existing stock ^[5]. Getting from $40 trillion to $50 trillion means roughly $10 trillion of additional debt ^[1], a 25% increase in the stock to be absorbed [d1b], much of it repriced upward.

For holders, higher yields cut both ways: new bonds pay more income, existing bonds lose value, and the longer the maturity the more sensitive the price ^[10]. That is the whole of the duration argument, and it does not require a fiscal crisis to hurt - only a market that keeps demanding more compensation.

Hartnett names an exit condition: the trade is unlikely to end until five-year Treasury yields fall below roughly 3.25% ^[7]. Until then he points to gold, equities, biotech and real estate ^[8]. His summary of the regime is blunt: "The U.S. stock market hit an all-time high on the same day that the U.S. Treasuries issued at their highest yield in 25 years," he said in the report. "That's reality" ^[9].

The spillover is the part operators cannot allocate around. Treasury rates set the baseline cost of borrowing across the economy, so higher yields feed into mortgages, corporate loans and consumer credit, and can slow investment, housing and spending ^[11].

What to watch: the five-year yield against Hartnett's 3.25% threshold, which the source does not price, so the distance to that trigger is unstated ^[7]; monthly borrowing prints against the $180 billion average, given July's $432 billion ^{[2] [5]}; and whether the implied 3.5% effective rate keeps drifting toward the 5.2% long-bond print as maturing paper rolls ^{[4] [4]}.