The Government's Own Forecasters Said 4.1%. Investors Are Charging 5.11%. Here's What the Gap Actually Costs.

Priya does her own taxes, tracks her own budget to the dollar, and had never once thought about what the federal government pays to borrow money as something that touched her own finances directly, until a line in a September news article stopped her: net interest on the national debt, $970 billion in the fiscal year that just closed, more than the $916.6 billion the government spent on national defense that same year. She did the division herself before she believed it. Divided across the roughly 135 million households in the country, that is about $7,185 a household, every year, before anyone touches a single government service. (Priya is a composite, built to show a typical realization; the numbers around her are real.)

That number alone is not new information; commentators have cited the $1 trillion interest milestone for months. What is less widely understood is why the bill keeps growing faster than the government's own forecasters expected, and that mechanism has a specific name and a specific, calculable cost of its own.

The gap between the forecast and the bill

In February 2026, the Congressional Budget Office's Budget and Economic Outlook assumed the 10-year Treasury yield would average 4.1% for the year, rising gradually to 4.4% by the early 2030s. That assumption feeds directly into the government's own projections for how much it will cost to finance the debt.

The actual 10-year Treasury closed at 5.11% on September 23, 2026, a full percentage point above what the CBO built its forecast on. SwitchWize's own analysis of the 10-year Treasury's climb to 5% covers the yield move itself in detail. This piece covers something narrower: why the government's own forecasters missed by this much, and what that specific miss costs.

Why yields include a term premium at all

The 10-year Treasury yield is not simply a bet on where short-term interest rates will average over the next decade. It also includes a term premium, the extra return investors demand to hold a long-term bond instead of continually rolling over short-term ones, as compensation for the risk that inflation or rates move against them over a longer horizon. A rising term premium means investors are demanding more compensation for that risk, independent of what they expect the Federal Reserve to do with short-term rates.

10-year Treasury term premium, Kim-Wright estimate, monthly since 1990. Source: FRED (THREEFYTP10).

One well-known estimate, published by the St. Louis Fed's public data service, puts the 10-year term premium at 0.96 percentage points in September 2026, up from 0.50 a year earlier, in September 2025. That is nearly a doubling in twelve months, and it sits above the 36-year average of 0.82 points. The term premium spent much of 2020 and 2021 below zero, when investors were effectively paying for the safety of long-term Treasuries rather than demanding compensation for holding them. Today's reading is a sharp reversal from that period, not a return to some historical extreme.

Heavier federal borrowing to fund the deficit, elevated inflation, and reduced foreign demand for Treasuries (foreign holders now own about 25% of the debt, down from roughly a third a decade ago) are the most commonly cited drivers of a rising term premium. None of these is a forecasting error on the CBO's part. They are conditions that shifted after the February forecast was published, which is a structural feature of any point-in-time budget projection, not a mistake specific to this one.

What the gap actually costs

The government's debt splits into two piles: money it owes to outside lenders (investors, foreign governments, pension funds), about $32.4 trillion, and money it owes to itself, mainly Social Security and other trust funds, which doesn't carry this same market interest-rate risk. The 1.01-point gap between the CBO's 4.1% assumption and the actual 5.11% yield is a real, calculable cost on that first pile, not an abstraction. Applied to the $32.4 trillion, per the U.S. Treasury's own daily debt figures, a 1.01-point higher borrowing cost adds approximately $327 billion a year in interest. Divided across roughly 135 million U.S. households, that is about $2,424 per household, every year, on top of what the government would have paid at the CBO's original 4.1% assumption.

SwitchWize calculation. Net interest from the U.S. Treasury's Monthly Treasury Statement for fiscal year 2025. Household count is an estimate; see Method below.

This calculation is deliberately simple: it is today's debt at a higher rate, not a projection of what happens next. A separate, harder question, what the gap costs over time as debt is repaid and re-borrowed at higher rates through the coming decade, is exactly what the Committee for a Responsible Federal Budget has modeled using the CBO's own long-run framework, estimating the elevated-yield path could add roughly $1.7 trillion to the debt over ten years and push interest costs toward $2.7 trillion a year by the decade's end. That is a different, longer-horizon number than this piece's simple today-only figure, and it depends on assumptions about future borrowing that this piece's math does not need to make.

Even the experts disagree on the exact size

The term premium is not a price anyone can look up directly, the way you can look up a Treasury yield. It is a calculated estimate: researchers take the actual 10-year yield and split it into two pieces, what investors expect short-term rates to average over that time, and the extra compensation on top of that. Different research teams split it slightly differently, so they land on different numbers. The 0.96-point figure used throughout this piece comes from one well-known estimate. A second, independent estimate from the Federal Reserve Bank of New York has recently put the same 10-year term premium closer to 0.6 percentage points, roughly a third lower.

Both agree on the direction: the term premium has risen meaningfully over the past year. They disagree on the exact size, which is a genuine limitation of this entire piece's math, not a flaw specific to one estimate over the other. The dollar figures above should be read as a reasonable calculation built on one credible estimate, not a precise, uncontested fact the way a Treasury yield itself is.

What this means, beyond the number itself

None of this is a reason to panic about a specific household's finances; the $2,424 figure is a share of a federal cost, not a bill anyone receives directly. What it does explain is something more useful: why the 10-year Treasury yield keeps running ahead of official forecasts, and why that is a fiscal and market phenomenon with a name, not a mystery. The same term premium showing up in the government's borrowing costs is the same term premium priced into every mortgage, auto loan, and corporate bond that takes its cue from the 10-year Treasury.

1. A forecast gap this size is a real cost, not noise. $327 billion a year is a genuine number with a clear mechanism behind it, worth understanding before dismissing headline debt figures as abstract.
2. The term premium, not just the debt total, is the thing to watch. The $40 trillion headline number moves slowly and predictably. The term premium can move meaningfully within a single year, as it just did, and it is the more sensitive gauge of how the bond market's own risk appetite for lending to the government is changing.
3. Model disagreement is a feature of this kind of analysis, not a reason to discard it. Two credible models landing 0.36 points apart on the same question is normal for an estimate this abstract, and it argues for directional confidence (the term premium is rising) over false precision about the exact number.

Priya's own conclusion was smaller than the headline figures but more useful to her directly: she does not need to do anything different with her savings or her mortgage because of a $2,424 number she will never see a bill for. What she changed was how she reads the next headline about Treasury yields moving. When a rate moves and a news story calls it "the market," she now knows enough to ask whether it is the Fed, expected inflation, or the term premium doing the moving, because they are three different things with three different causes.

Quick answers

What is causing the gap between the CBO's Treasury yield forecast and the actual yield? The term premium, the extra return investors demand to hold long-term Treasuries, has nearly doubled over the past year, from 0.50 to 0.96 percentage points by one widely used estimate, driven by heavier federal borrowing, elevated inflation, and reduced foreign demand.

How much does that gap cost? About $327 billion a year applied to the $32.4 trillion the government owes to outside investors, roughly $2,424 per U.S. household, on top of the $970 billion, $7,185-per-household bill the government already paid in fiscal year 2025.

Do all term-premium estimates agree? No. Two well-known estimates currently differ by about 0.36 percentage points on the same question, though both show the same rising direction over the past year.