The $40 trillion national debt and the bond market’s revolt: top Wall Street strategists explain how we got into this mess

Total U.S. debt crossed $40 trillion for the first time this month. In the week since, two of Wall Street’s most closely read economists independently arrived at the same diagnosis: nobody in Washington is going to fix this, so the bond market will do it instead—whether the Treasury Department likes it or not. They also got a major helping hand from an AI-assisted column in the Wall Street Journal by hedge fund legend Stanley Druckenmiller.

On Tuesday, David Kelly, chief global strategist at J.P. Morgan Asset Management, used the milestone to walk clients through exactly how the country got here. A day later, Apollo chief economist Torsten Slok argued in his Daily Spark that the fiscal trajectory, a Federal Reserve weighing a rate hike, and a surge of AI hyperscaler bond issuance are all pointing toward the same outcome: rates that stay higher for longer. Slok closed his note by endorsing a line from billionaire investor Druckenmiller, who had made the same argument in the Journal a day earlier, with considerably more edge: the long-term Treasury yield is “the only fiscal disciplinarian the U.S. has left.”

That op-ed, of course, was an unexpected, direct attack on the Treasury Department’s decision, announced Aug. 19, to double long-dated bond buybacks from $2 billion to at least $4 billion per operation—right after the 30-year Treasury yield hit a 19-year high. “The market’s verdict was swift and correct,” Druckenmiller wrote in AI-inflected overtones. “This wasn’t liquidity management, it was price management.” His prescription, delivered in the same essay: “If the 30-year must trade at 5.5% to clear, that isn’t a crisis. It is an invoice.”

The essay carries extra weight because Druckenmiller was Treasury Secretary Scott Bessent’s mentor at Soros Fund Management three decades ago—the two, alongside George Soros, built the trade that broke the Bank of England’s defense of the pound in 1992. Now Druckenmiller is using the same playbook—reading the gap between what a government claims it can sustain and what markets will actually allow—against his own protégé. Jon Hilsenrath, the former longtime Fed and Treasury reporter for the Journal, told Fortune that Druckenmiller’s decision to publish in the Journal, rather than deliver the message privately, was telling, agreeing that it was a bit of a “Shakespearean drama.”

The math behind the milestone

Kelly’s note draws a distinction that matters more than the $40 trillion headline number itself. That figure is total federal public debt outstanding, which includes roughly $7.8 trillion the government owes its own trust funds. The measure economists actually watch—debt held by the public—will end this fiscal year at $32.3 trillion, or 100.5% of GDP, J.P. Morgan projects. That ratio was just 34.7% as recently as fiscal 2000, when the federal government posted a $236 billion surplus.

Kelly noted that former Fed Chairman Alan Greenspan fretted in 2001 about what would happen if the U.S. actually paid off all of its federal debt. “He needn’t have worried.” Kelly traced the reversal to four buckets of fiscal decisions compounding since then, measured against the last time the budget was healthy (fiscal 1996–2000):

• Tax cuts in 2001, 2017, and 2025 pulled federal revenue down from an average of 19.1% of GDP to 16.7%, a cumulative $11.1 trillion hit.
• Wars in Iraq, Afghanistan, and Iran pushed defense spending from 3.5% to 4.4% of GDP, adding $3.9 trillion.
• Social Security, Medicare, and Medicaid spending climbed from 7.8% to 10.1% of GDP as the population aged, adding $12.5 trillion.
• Everything else, boosted by crisis-response spending during the 2008 financial crash and the pandemic, added another $5.2 trillion.

In short, in the last 25 years, America voted itself a series of tax cuts, waged expensive wars, got older, and then spent its way out of a couple of crises. Add it up—$32.7 trillion, before interest costs—and it “more than accounts for” the debt surge of the 21st century, Kelly wrote. He was explicit that the real drivers aren’t the culture-war talking points dominating political debate: entitlements, defense, and tax policy, full stop.

Why the old rules stopped working

Slok’s note picked up where Kelly’s history lesson leaves off and pointed forward. Since 2006, gross federal debt has grown by $32 trillion while nominal GDP grew by just $19 trillion—debt nearly quintupling while the economy grew less than 2.5x over the same period. The forecasts offer no relief: the Congressional Budget Office projects debt held by the public climbing from 100% toward 175% of GDP under current policy, while the Office of Management and Budget sees deficits near 5% of GDP in coming years, on top of a current run rate closer to 6%. Deficits that size are normal in a recession, Slok notes. These are forecasts for a fully employed economy.

Both economists agree that deficits of this magnitude no longer automatically trigger the inflation spiral that economic orthodoxy once predicted—which is precisely why Washington has felt no urgency to act. Kelly pointed to the bond market’s own pricing as evidence. Since January, 10-year Treasury yields have risen 0.51 percentage points, while 10-year TIPS yields—which strip out inflation expectations—rose almost as much, 0.44 points. That leaves only 0.07 points of the move attributable to rising inflation fears. The rest, Kelly argues, reflects a “growing fear about the volume of government debt to be issued,” not inflation itself.

Slok’s explanation converges on the same fear from a different data set: AI hyperscalers’ surging bond issuance now competes directly with the Treasury for buyers, adding pressure as the Fed debates a hike rather than a cut.

A bigger fight is brewing

The Druckenmiller op-ed complicates a simpler story that both Kelly and Slok leave out. Hilsenrath noted that Druckenmiller’s argument isn’t that Treasury should never buy back debt—the buyback program, introduced in 2024 as a liquidity tool, can legitimately improve market functioning by purchasing older, thinly traded bonds. Druckenmiller’s real complaint is timing: Treasury enlarged the program right after the 30-year yield spiked to a two-decade high, outside its normal quarterly rhythm, which markets read as flinching at an uncomfortable price rather than managing routine liquidity[web:10].

There’s also a structural irony neither Kelly nor Slok addressed directly. The Treasury market Bessent now manages isn’t the one Druckenmiller’s generation tested in 1992. Foreign central banks used to absorb much of new Treasury issuance; that mechanism has weakened sharply since the financial crisis. In their place, hedge funds—often operating through offshore centers—have become the marginal buyer, holding $2.4 trillion in long Treasury exposure as of last September, more than mutual funds or U.S. banks, per a New York Fed analysis cited by Columbia financial historian Adam Tooze. The industry Druckenmiller helped build by betting against governments now largely finances the government whose credibility he’s publicly questioning.

Hilsenrath’s read on the stakes: a 5% Treasury yield “is not a clear and present danger to the economy… but it is a problem, which is why you have to pay attention to these market signals now.” Whether Washington listens is exactly what Kelly and Slok are both betting against.

For this story, Fortune journalists used generative AI as a research tool. An editor verified the accuracy of the information before publishing.

This story was originally featured on Fortune.com