This article explains what the US Treasury yield curve is, why it sometimes “inverts,” and why even the Federal Reserve’s own economists warn against reading it as a simple recession alarm. It describes United States Treasury markets and Federal Reserve policy; other countries’ government bond markets behave similarly in some ways but have their own conditions.
What is the yield curve, exactly?
The yield curve plots the interest rate, or yield, that the US government pays to borrow for different lengths of time, from a few months up to 30 years. Normally, lenders want more compensation for tying money up longer, so longer-dated yields sit above shorter ones and the curve slopes upward.
Commentators usually boil the curve down to a single number: the spread between two maturities. The best known is the 10-year Treasury yield minus the 2-year Treasury yield. The Federal Reserve Bank of St. Louis publishes this daily as a series called T10Y2Y, calculated from yields published by the US Treasury Department. Another common version subtracts the 3-month yield from the 10-year yield.
When the spread is positive, the curve slopes upward. When it falls below zero, the curve is “inverted”: short-term borrowing costs more than long-term borrowing.
What the curve looks like today
As of October 6, 2026, according to the St. Louis Fed’s FRED database, the 10-year minus 2-year spread was positive, meaning the curve was not inverted on that measure. The 10-year minus 3-month spread was also positive on that date. Daily figures move, so the precise level matters less than the shape: an upward slope.
Why does the curve invert?
Short-term yields are heavily influenced by where the Federal Reserve sets its policy rate and where traders expect it to go. Longer-term yields reflect an average of expected future short rates, plus extra compensation for holding longer bonds, often called the term premium.
That produces a few typical shapes:
| Curve shape | What the spread looks like | Common reading |
|---|---|---|
| Upward-sloping | Long yields above short yields | Normal conditions or expected steady policy |
| Flat | Long and short yields about equal | Markets unsure whether rates will rise or fall |
| Inverted | Short yields above long yields | Markets expect policy rates to fall in future |
An inversion typically appears when the Fed has pushed short-term rates up to slow inflation and investors expect that, later on, rates will have to come back down. Those future lower rates pull longer-term yields below today’s short-term ones. The link to how bond prices respond to interest rates is direct: yields move inversely to the prices of existing bonds.
A worked example with illustrative numbers
Suppose, purely for illustration, that the 2-year Treasury yields 4.0% and the 10-year yields 3.7%. The 10-year minus 2-year spread is 3.7 − 4.0 = −0.3 percentage points, an inversion.
Now suppose the 2-year yield falls to 3.5% while the 10-year stays at 3.7%. The spread becomes 3.7 − 3.5 = +0.2 points. The curve has “un-inverted,” but nothing about the economy needed to improve for that to happen. The 2-year yield simply fell, perhaps because traders expected rate cuts. This is why analysts often note that the curve can return to a positive slope shortly before, or during, a downturn rather than only after it ends.
Has an inversion preceded recessions?
The National Bureau of Economic Research, the body that dates US business cycles, lists recessions with peaks in December 1969, November 1973, January 1980, July 1981, July 1990, March 2001, December 2007 and February 2020. The 2020 contraction ran from February to April, a span of two months, according to the NBER’s table.
Federal Reserve research has found that yield-curve slope has had real predictive power. A 2018 FEDS Note from Federal Reserve Board economists used the 10-year minus 3-month spread in statistical models to estimate recession probability over the following one to four quarters. The same note was cautious. Its authors wrote that “we should be cautious about relying too heavily on its predictions,” pointing out that their simple model’s estimate fell close to zero in early 2008, even though a recession was beginning around then.
Why do Fed economists urge caution?
Two main reasons emerge from the Fed’s own notes.
First, the headline 10-year minus 2-year measure is not necessarily the best signal. Federal Reserve economists have argued that a different measure, the near-term forward spread, is statistically more powerful. This compares the market’s implied three-month rate six quarters ahead with the current three-month rate, and so reads as a market forecast of the Fed’s path over roughly the next year and a half.
Second, the curve does not cause recessions. In a March 2022 FEDS Note, the authors argued that it is not valid to interpret inverted term spreads as independent measures of impending recession. They describe attributing predictive power to the spread itself as a case of reverse causality: the spread reflects what market participants already expect about policy and growth. In that note, they reported a 2-10 spread of about 0.25 percentage points against a near-term forward spread above 2 points, showing how differently the two measures can read at the same moment.
What does the inversion tell you about timing?
Timing is the weak point of any single-indicator story. The Fed notes above model recession risk over windows of one to four quarters rather than pinpointing a date, and the NBER’s table shows recessions of very different lengths, from two months in 2020 to 18 months from December 2007 to June 2009. A signal that cannot say when, or how deep, offers little for planning. Because markets are forward-looking, as covered in why markets can rise even when the economy feels weak, the curve can also invert well before economic data shows any strain.
There is also a distinction between confirmed fact and interpretation here:
- Confirmed: the spread is measurable daily, and its history is public.
- Documented in Fed research: slope has had statistical links to later recessions, and so has the near-term forward spread, more strongly.
- Analysis and opinion: whether any given inversion signals a coming recession, and when, is a judgment, not a certainty.
Why the same signal can mean different things in different eras
Bond yields are not set only by expectations of Fed policy. The 2018 FEDS Note observed that a decline in the term premium explained much of the flattening of the curve at that time, and warned that a model estimated over a long sample may overstate the recession probability as a result. In plain terms: if investors accept lower compensation for holding long bonds, perhaps because of heavy demand for safe assets, the curve can flatten or invert without any change in the outlook for growth.
The same note described the term spread narrowing from about 3.6 percentage points in the second quarter of 2009 to 1.2 points in the first quarter of 2018, a slide that happened while the economy was still expanding. Professional forecasters surveyed at that point put the probability of a recession beginning in 2018 at 19 percent, well below the 30 percent from the simple model. Neither number turned out to be a verdict; they simply show how much the answer depends on the method.
How should a long-term investor read the headlines?
When headlines say the curve has inverted or un-inverted, a few questions help put them in context:
- Which spread is being quoted, 10-year minus 2-year, 10-year minus 3-month, or something else?
- Is the move coming from short yields, which track Fed policy expectations, or from long yields, which can be affected by the term premium?
- Is the claim about what has happened, or a prediction of what will?
None of this makes the curve useless. It is a compact summary of what bond investors collectively expect from interest rates. But as with short-term market moves generally, a single indicator rarely justifies a confident conclusion on its own. For wider background on why the central bank’s rate path matters across the economy, see why interest rates matter more than most people think.
The yield curve as a lens, not a verdict
The yield curve condenses a lot of expectation about Federal Reserve policy, inflation and growth into one line. As of early October 2026, the 10-year minus 2-year spread in the United States was positive. Its record as a signal is real but imperfect, and Federal Reserve economists themselves stress that it reflects expectations rather than causing outcomes. Treat it as one input to understanding markets, not a forecast.
This article is general information, not personalised financial advice; rules and rates change.
Sources: Federal Reserve Bank of St. Louis; National Bureau of Economic Research; Federal Reserve Board
