The United States national debt crossed 100 percent of GDP this week, and the coverage has been exactly what you would expect. Breathless headlines. Grim milestone warnings. Comparisons to the immediate aftermath of World War II. The Committee for a Responsible Federal Budget called it “a total bipartisan abdication of making hard choices,” which is fair, but not particularly illuminating about what the number actually means.
Here is the thing: the $39 trillion figure that has been circulating is not wrong exactly, but it is not the number you should be looking at. And the number you should be looking at is more complicated than either the alarmed coverage or the dismissive response suggests. Understanding why requires understanding what the national debt actually is, who holds it, and what the different categories of holders mean for the economic analysis.
Two Numbers, Two Different Problems
The federal debt comes in two flavors, and conflating them is where most of the confusion starts.
The first is debt held by the public. This is Treasury securities held by external parties: individuals, pension funds, foreign governments, banks, insurance companies, and the Federal Reserve. As of March 31, [debt held by the public stood at 31.22 trillion, pushing the ratio to 100.2 percent. This is the economically meaningful number. It represents real claims on future federal revenue by parties who are not the federal government itself.
The second is intragovernmental debt. This is Treasury securities held by federal government trust funds and accounts. The gross national debt figure of 7.7 trillion, is sometimes dismissed as the government owing itself, which is partially correct but substantially misleading, because not all intragovernmental debt is the same thing.
The Fed Is Not the Government (Mostly)
The Federal Reserve holds Treasury securities through the System Open Market Account, known as SOMA. This is the portfolio the Fed has accumulated over decades of open market operations and, more recently, through quantitative easing programs that expanded the balance sheet dramatically after the 2008 financial crisis and again during the COVID pandemic.
Whether SOMA holdings count as intragovernmental debt is genuinely contested and the standard presentations paper over this entirely. The Fed is not a government agency in the traditional sense. It was established by Congress but operates with significant institutional independence. Its governors are presidential appointees confirmed by the Senate, but its monetary policy decisions are made without executive or congressional approval. When Treasury pays interest on securities held by the Fed, the Fed remits most of those earnings back to Treasury as a payment to the general fund.
So is this the government owing itself? Sort of. There is real institutional separation between Treasury and the Fed, and that separation matters for monetary policy independence. But the interest flow largely circles back, which means the net cost to the government of the Fed’s Treasury holdings is substantially lower than the face value of the securities would suggest. It is not nothing, but it is not the same as owing JPMorgan or the People’s Bank of China.
The complication is that quantitative tightening has put the Fed in the unusual position of running operating losses, which has suspended remittances to Treasury. The Fed has been recording a deferred asset, essentially an IOU to itself, representing the accumulated losses it will need to recover before remittances resume. According to the Congressional Research Service, the Fed’s net income was negative from September 2022 through November 2025, the first time since 1934 that remittances fell close to zero. The cumulative operating loss through 2024 was approximately $191.9 billion. So even the “the Fed just gives the interest back” argument has gotten more complicated in the current tightening cycle.
Social Security Is a Different Kind of Problem
The Old Age, Survivors, and Disability Insurance trust funds --- OASDI, the formal name for Social Security, hold a different category of Treasury securities entirely. These are special-issue nonmarketable securities that are legal obligations of the Treasury in a way that carries real consequences.
Here is why this matters. When Social Security collects more in payroll taxes than it pays in benefits, the surplus goes into the trust funds in the form of these special Treasury securities. When it pays out more than it collects, which is the situation the program has been in since 2021, it redeems those securities, and Treasury has to come up with the cash from somewhere. That means borrowing from the public, cutting spending elsewhere, or raising revenue. There is no magic accounting maneuver available when the trust fund calls in its chips.
According to the 2025 Social Security Trustees Report, the combined OASI and DI trust fund reserves declined by 2.72 trillion. Under current projections, the funds are expected to be exhausted in 2034, at which point benefits would be payable only to the extent that current payroll tax revenues cover them, roughly 81 percent of scheduled benefits at that time, declining further thereafter.
This is not the same as the government owing itself. It is the government having made specific commitments to specific beneficiaries, with a financing mechanism that is now running short. The institutional form is intragovernmental, but the economic substance is an obligation to the public in the most direct sense possible. Treating it as meaningless accounting misses what these securities actually represent: pre-funded claims that are now being drawn down on a predictable schedule.
The Number That Actually Matters
None of this means the fiscal situation is fine. It is not fine. The debt held by the public at 100 percent of GDP is a genuine problem, not because 100 percent is a magic threshold --- it is not, but because the trajectory is what matters and the trajectory is bad.
The CBO projects that debt held by the public will reach 120 percent of GDP by 2036 and 175 percent within thirty years under current law. The One Big Beautiful Bill Act, which extended the 2017 tax cuts and added new ones, is projected to add roughly 1 trillion in fiscal year 2026, nearly triple what they were in 2020.
The relevant economic question is not whether debt exceeds GDP but whether the interest rate on the debt exceeds the growth rate of the economy. When that condition holds on a sustained basis, you get a debt spiral: interest costs crowd out other spending, the economy slows, slower growth means lower revenue, lower revenue means more borrowing, more borrowing means higher interest costs. The CBO’s projections suggest this condition could emerge by the early 2030s under plausible interest rate assumptions.
Japan is frequently cited as evidence that high debt-to-GDP ratios are survivable. This is true but limited as an analogy. Japan’s debt-to-GDP ratio currently stands at approximately 236 percent, and it has run at these elevated levels for decades. But Japan has done so with low interest rates, a high domestic savings rate, and a current account surplus that keeps the debt predominantly in domestic hands. The United States does not have all of those features and cannot simply assume they will persist indefinitely.
What the Coverage Gets Wrong
The breathless coverage of the 100 percent milestone makes two related errors. The first is treating the gross debt figure as the economically relevant number when the debt held by the public is what actually drives the macroeconomic dynamics. The second is treating all intragovernmental debt as equivalent when the Fed’s SOMA holdings and the Social Security trust funds represent fundamentally different kinds of obligations with different implications for policy.
The dismissive response, “it is just the government owing itself, don’t worry about it,” makes the opposite error, treating intragovernmental debt as economically meaningless when the Social Security trust fund in particular represents real commitments with real redemption pressures arriving on a predictable schedule.
The honest answer is that the fiscal situation is serious, the gross debt figure overstates the immediate problem, the public debt figure understates the long-term problem by excluding implicit liabilities, and the institutional complexity of the Fed’s balance sheet and the trust fund accounting makes the standard presentation misleading in both directions.
100 percent of GDP is not a cliff. It is a useful marker that tells us we have been making choices, on taxes, on spending, on what we are willing to pay for and what we are willing to borrow for, that are not sustainable on their current trajectory. The number is not wrong. But the story being told about it usually is.
For a deeper dive into the mechanics of federal debt accounting, the Congressional Budget Office’s February 2026 Budget and Economic Outlook is the authoritative starting point. The Committee for a Responsible Federal Budget maintains running analysis of the fiscal picture.