The Trillion-Dollar Interest Bill

For most of American history, the federal interest bill was a manageable nuisance — a cost of doing business that rarely exceeded 15% of federal revenue. During the 2010s, when the Federal Reserve held interest rates near zero and bought trillions of dollars in Treasury bonds through quantitative easing, the interest expense actually fell even as the debt doubled. It was the great paradox of the post-crisis era: the government could borrow more and pay less. In 2003, when the debt was $6.8 trillion, interest payments totaled $292 billion. In 2015, when the debt was $18.2 trillion — nearly three times larger — interest payments were only $402 billion. The zero-rate environment had turned the debt into something close to a free lunch.

The lunch ended in March 2022, when the Federal Reserve began the most aggressive rate-hiking cycle in four decades. Over 16 months, the federal funds rate rose from 0.08% to 5.33%. The impact on Treasury borrowing costs was immediate and brutal. Every maturing bond had to be refinanced at the new, higher rates. Every new dollar of deficit spending was borrowed at 4–5% instead of 0–1%. The quarterly annualized interest cost jumped from $637 billion in Q1 2022 to $823 billion by Q4 2022 — a $186 billion increase in nine months. By Q4 2023, it had reached $1.04 trillion. By Q4 2025, it stood at $1.228 trillion.

In 2023, interest payments surpassed defense spending for the first time in American history. The numbers were unambiguous: interest hit $938 billion while defense expenditures totaled $790 billion. The gap widened in 2024 to $1.12 trillion vs. $856 billion, and by 2025 the interest bill exceeded defense by more than $300 billion. This crossing was not just symbolic. It means that America now pays more to service its past borrowing than it spends on its entire military apparatus — the 11 aircraft carrier strike groups, 1.3 million active-duty personnel, 750 overseas military bases, and the nuclear triad. The Pentagon’s budget is capped by political choice; the interest bill is set by mathematical necessity.

The comparison between interest and defense spending is not arbitrary. Defense has traditionally been the largest discretionary spending category — the thing the government chooses to spend money on, as opposed to mandatory programs like Social Security that run on autopilot. When interest surpasses defense, it signals that the cost of past borrowing has overtaken the cost of current national priorities. It is the fiscal equivalent of a family spending more on credit card payments than on rent.

The two lines tracked closely for years. In 2000, defense spending was $302 billion and interest was $354 billion — defense was actually below interest. But the post-9/11 military buildup pushed defense to $554 billion by 2007 while falling rates shrank interest to $352 billion. For fifteen years (2003–2022), defense comfortably exceeded interest. Then the rate shock reversed the relationship. The quarterly data shows the exact moment: in Q3 2023, annualized interest ($975B) overtook annualized defense ($793B). The lines have not recrossed since.

The velocity of the interest increase is what alarmed fiscal analysts. From 2019 to 2025, interest payments more than doubled: $577 billion to $1.18 trillion. No other major spending category grew even half that fast. Social Security grew 36%, Medicare grew 42%, defense grew 21%. Interest grew 105%. And unlike Social Security or defense, interest spending is entirely outside congressional control. It is determined by two factors — the size of the debt and the prevailing interest rate — neither of which Congress can directly set. The interest bill is the budget item that budgets itself.

The most dangerous feature of the interest bill is that it is self-reinforcing. The government borrows to pay interest, which increases the debt, which increases the interest, which requires more borrowing. This is not a metaphor — it is the literal accounting. In FY2025, the federal government will collect approximately $5.2 trillion in revenue and spend approximately $1.2 trillion on interest alone. That leaves $4.0 trillion for everything else — but everything else costs $5.8 trillion. The resulting $1.8 trillion deficit adds to the debt, which will generate an additional $70–$90 billion in annual interest at current rates, which widens the deficit further.

Economists call this dynamic the “r minus g” problem. If the interest rate on the debt (r) exceeds the growth rate of the economy (g), the debt-to-GDP ratio will rise mechanically even without new deficits. During the 2010s, r was near zero while g averaged 2–3%, so the ratio naturally stabilized. Today, the weighted average interest rate on outstanding Treasury debt is approximately 3.3% and rising (as low-rate bonds mature and are replaced by higher-rate ones), while nominal GDP growth is roughly 5%. For now, g exceeds r — but the margin is narrowing, and the CBO projects a crossover in the early 2030s as rates remain elevated and growth moderates.

The weighted average maturity of outstanding Treasury debt is approximately 6.2 years, which means roughly one-sixth of the debt rolls over each year. As of Q4 2025, bonds issued during the zero-rate era (2020–2021) at 0.5–1.5% are being replaced by bonds at 4.0–4.5%. Each refinancing ratchets up the effective interest rate. Even if the Fed cuts rates significantly, the full benefit won’t flow through to Treasury borrowing costs for years, because much of the debt is in longer-term bonds. The interest bill has momentum — it will keep rising even in a falling-rate environment, simply because the stock of low-rate debt is shrinking while the total debt keeps growing.

Twenty-two cents of every dollar the government collects now goes to interest payments. In 2015, the figure was twelve cents. The interest-to-revenue ratio has nearly doubled in a decade, even as revenue itself grew substantially. This is the clearest measure of what the debt costs in practical terms: for every tax dollar you send to Washington, twenty-two cents goes to bondholders before a single soldier is paid, a single Social Security check is mailed, or a single road is paved.

At the current trajectory, the CBO projects the interest-to-revenue ratio will reach 25% by 2030 and 30% by 2035. At 30%, the United States would be spending more on interest than on the combined total of Medicare and Medicaid. At that point, the compounding trap becomes nearly inescapable: the interest bill is so large that meaningful deficit reduction requires either enormous tax increases, enormous spending cuts, or financial repression (inflating the debt away). None of these options is painless, and all of them are politically explosive. The yield curve — which determines the price of all this borrowing — is the subject of Episode 6.