TL;DR. The yield curve plots U.S. Treasury interest rates against how long you’re lending the government money. Most of the time short rates are below long rates. When that relationship flips — short above long — it’s called an inversion, and historically every U.S. recession since 1955 has been preceded by one, typically 6–24 months in advance.
What the curve actually shows
Every weekday, the U.S. Treasury auctions or trades debt at a range of maturities: 1-month and 3-month bills, 2-year notes, 5-year, 10-year, up to 30-year bonds. The yield on each is the return an investor earns if the bond is held to maturity. When you plot those yields on the y-axis against maturity on the x-axis, you get the yield curve. The Federal Reserve’s H.15 statistical release publishes these daily, and the Fed of St. Louis hosts the same series on FRED.
Snapshot: where the curve sits right now
Here is a recent snapshot of the U.S. Treasury curve:
| Maturity | Yield (April 22, 2026) | Spread vs 10Y |
|---|---|---|
| 3-month Treasury bill | 3.60% | –0.70 pp (below 10Y) |
| 2-year Treasury note | 3.79% | –0.51 pp (below 10Y) |
| 10-year Treasury note | 4.30% | 0 (reference) |
| 30-year Treasury bond | 4.90% | +0.60 pp (above 10Y) |
The shape in one chart
Why shape matters: three regimes
- Normal (upward-sloping). Long rates above short rates. Investors demand a premium — a term premium — for locking up money longer. This is the default state of a healthy economy.
- Flat. Short rates and long rates are roughly equal. Often seen in mid-to-late cycle when the Fed has tightened enough to slow growth but hasn’t broken it.
- Inverted. Short rates above long rates. Bond investors are saying: we expect the Fed to cut in the future because growth is going to slow. The curve is literally pricing in a coming downturn.
Why inversion matters
Two spreads are the ones market watchers quote most:
- The 10Y – 2Y spread (or “2s/10s”): a classic intermediate-horizon measure of curve shape.
- The 10Y – 3M spread: the Federal Reserve’s own preferred recession-probability input. Work by Estrella and Mishkin at the New York Fed (1996) showed this spread predicts recessions on a 12-month horizon better than the 2s/10s.
Every U.S. recession since 1955 has been preceded by an inversion of one or both spreads, according to research by the Federal Reserve Bank of San Francisco. The lead time is not fixed: historically it has run roughly 6 to 24 months, with a median closer to 12. The curve re-steepens before the recession actually begins, which is part of why timing the turn is so difficult.
The New York Fed publishes a monthly recession-probability estimate derived from the 10Y–3M spread that is updated on its research page.
What actually drives the shape
- Short end is anchored by the Fed’s policy rate. The 3-month and 2-year yields move closely with market expectations for the federal funds rate.
- Long end reflects expectations for future short rates plus a term premium for duration risk and uncertainty. Inflation expectations, growth expectations, foreign demand for Treasuries, and the Treasury’s own issuance mix all push and pull on the long end.
- Quantitative tightening (QT) and easing (QE) directly move the long end. When the Fed buys long Treasuries, their yields fall; when it lets them run off the balance sheet, yields drift higher, all else equal.
Common mistakes
- Treating inversion as a timer. Inversion signals direction, not date. The S&P 500 has often risen in the year after initial inversion before rolling over later.
- Watching only one spread. 2s/10s and 10Y–3M can disagree. Both flashing red is a stronger signal than either alone.
- Ignoring the steepener. The curve usually re-steepens just before recession begins. A sudden bull steepening (long rates falling faster than short) after a period of inversion has been a more actionable late-cycle signal than the inversion itself.
- Confusing real with nominal. Real yields (TIPS) tell a different story than nominal yields — watch both.
What to read next
If this clicked, the natural next concepts are duration (how much a bond’s price moves when yields change) and real yields (yields adjusted for expected inflation via TIPS). Both get at why the long end moves when it does.
A quick history lesson: the 10Y–2Y spread and recessions
A handy way to visualize the curve’s signal power is to plot the spread between the 10-year and the 2-year yield over time. When the line drops below zero, the curve is inverted. The gray bars below mark official U.S. recessions as defined by the National Bureau of Economic Research.
Sources
- Federal Reserve — H.15 Selected Interest Rates (daily Treasury yield release).
- Federal Reserve Bank of St. Louis — FRED Treasury yield series (DGS3MO, DGS2, DGS10, DGS30).
- Federal Reserve Bank of New York — Yield Curve as a Leading Indicator (FAQ and monthly probability update).
- Federal Reserve Bank of San Francisco — Economic Forecasts with the Yield Curve, Economic Letter 2018-07.
- Estrella & Mishkin (1996), “The Yield Curve as a Predictor of U.S. Recessions,” FRBNY Current Issues in Economics and Finance.
Disclosure: This article was produced with AI assistance and reviewed before publication. It is for informational purposes only and is not investment advice.