The yield curve: shapes, slopes and what each end tells you

The yield curve is a line drawn through the yields of US Treasury securities, from bills that mature in a month to bonds that mature in 30 years. Each point answers one question: what annual return does the market demand to lend to the US government for this long? Put the answers side by side and you get a shape, and that shape is one of the most watched pictures in finance.

It matters because almost every other interest rate is priced off it. Bank deposit rates and money-market yields track the short end. Mortgage rates, corporate bond yields and the discount rates analysts use to value stocks lean on the long end. When the curve moves, borrowing costs across the economy move with it, and the slope between the two ends carries information about what investors expect the Federal Reserve, inflation and growth to do next.

This guide explains how the curve is built, why it usually slopes upward, what each end reflects, the vocabulary traders use for its moves, and how to read it, using Treasury data from 1960 to October 2026. The recession-warning question gets its own page, Yield curve inversion: what it signals and its recession record; here the focus is the whole curve.

What the yield curve is

A yield curve compares bonds that differ in one way only: time to maturity. US Treasuries are the standard choice because they share the same issuer and the same credit risk, they trade in a deep market every business day, and the government issues them at regular maturities across the whole range. Differences in their yields are therefore mostly about time, not about who is borrowing.

In practice "the yield curve" usually means the constant-maturity Treasury curve. Every business day, Treasury takes indicative bid-side prices for the most recently auctioned bills, notes and bonds and fits a smooth par yield curve through them (with a monotone convex method since December 2021). Reading that curve at fixed maturities, from 1 month to 30 years, gives the yield a new security of exactly that maturity would carry. The Federal Reserve republishes these as its H.15 release, and those are the numbers on FRED and in the charts on this page. Because they are fitted at fixed points, you can compare the "10-year yield" on two dates even though the actual note that was 10 years from maturity has changed.

Yields here are quoted as annual rates in percent. A move of one hundredth of a percentage point is one basis point; a spread between two maturities is quoted in percentage points (pp) or basis points.

The four basic shapes

Analysts sort curves into a handful of shapes. Real curves are rarely textbook-clean, but the labels are useful shorthand.

• Normal (upward sloping). Longer maturities yield more than shorter ones. This is the usual state. The curve on 28 October 2026 is an example: 4.07% at 1 month, 4.67% at 2 years, 5.05% at 10 years.
• Steep. A normal curve with an unusually wide gap between the ends. On 31 December 2021 the 1-month bill yielded 0.06% and the 20-year bond 1.94%, a gap of almost 1.9 points created mostly by a front end pinned near zero.
• Flat. Yields are about the same at every maturity. Flat curves are often a transition between normal and inverted.
• Inverted (downward sloping). Short maturities yield more than long ones. On 3 July 2023 the 6-month bill yielded 5.53% and the 10-year note 3.86%. The 10-year minus 3-month spread was −1.58 points and the 10-year minus 2-year spread −1.08 points.

There are also humped curves, where the middle sits above both ends. The 2023 curve was humped at the front, peaking at 6 months. And segments can disagree: on 28 October 2026 the 20-year yield (5.41%) was slightly above the 30-year (5.37%), a small kink at the long end of an otherwise upward-sloping curve. The 20-year has carried a premium over the 30-year on 995 of the 1,604 trading days from June 2020 to October 2026 in the daily data; the label "normal" describes the overall slope, not every segment.

What sets the short end

The front of the curve, from 1 month to about 1 year, is tied closely to monetary policy. The Federal Open Market Committee sets a target range for the federal funds rate, the overnight rate banks charge each other. A 3-month bill is an alternative to rolling overnight money for three months, so its yield cannot stray far from where investors expect the overnight rate to average over those three months.

The data show the link plainly. At the end of 2021 the target range was 0–0.25% and the 1- and 3-month bills yielded 0.06%. The Fed then raised the top of the range eleven times, from 0.25% in March 2022 to 5.50% in July 2023. On 3 July 2023, with the range at 5.00–5.25%, the 3-month bill yielded 5.44%. On 28 October 2026 the range was 3.75–4.00%, after a 25 basis point increase that took effect on 17 September 2026, and the 1-month bill yielded 4.07%.

Notice that the 3-month yield in July 2023 sat above the range then in force. Bills price the expected path of policy over their life, not today's rate. A further increase took effect on 27 July 2023, and the bill market had already absorbed it. The 2-year yield extends the same logic further out: it is roughly the market's average expected policy rate over two years, plus a small premium. That is why the 2-year is the maturity traders watch most closely around FOMC meetings.

What sets the long end

Long yields have two building blocks.

1. Expected future short rates. Lending for 10 years is an alternative to lending for one year ten times in a row. If investors were indifferent between the two, the 10-year yield would equal the average of the 1-year rates they expect over the decade. This is the expectations hypothesis. Expected short rates depend on expected inflation and on where investors think the Fed's rate will settle once today's cycle is over.
2. The term premium. Investors are not indifferent. A 10-year note's price swings much more than a bill's when rates change (see Bond prices and yields: why prices fall when yields rise), and its real value is exposed to a decade of inflation surprises. The term premium is the extra yield investors demand for that risk, over and above the expected path of short rates. It cannot be observed directly; economists estimate it with models, and different models give different numbers.

You can turn the expectations part into arithmetic. If two maturities are priced consistently, the yield on the longer one implies a forward rate: the rate the market is locking in for the later period.

That ambiguity explains why the curve normally slopes upward even when nobody expects rates to rise. With flat expectations and a positive term premium, longer maturities still yield more. Over the 1960–2026 sample in the chart below, the monthly 10-year yield averaged about 1.4 points above the 3-month bill yield.

A long history of the slope

The most common single-number summary of the curve is a term spread: a long yield minus a short one. The two most quoted are 10-year minus 3-month and 10-year minus 2-year. The chart below shows the first, monthly, since 1960.

Some landmarks from the monthly data behind it:

• The widest reading was +4.42 points in September 1982, when bill yields fell quickly as the Fed eased while the 10-year stayed high.
• The most negative was −2.65 points in December 1980, at the height of the Fed's fight against double-digit inflation, when bills yielded far more than long bonds.
• The spread went below zero in 1966–67, 1969–70, 1973–74, 1978–81, 2000, 2006–07, 2019–20 and late 2022 through November 2024. What happened next each time is the subject of Yield curve inversion: what it signals and its recession record.
• By September 2026 it had widened back to +1.05 points, a little below its long-run average.

The picture to take away is that the slope is cyclical. It tends to be steepest early in an economic expansion, when the Fed has cut rates to low levels and investors expect them to rise, and flattest or inverted late in a tightening cycle, when short rates are high and investors expect them to come down.

The vocabulary of curve moves

Traders describe a change in the curve with two words: one for the slope and one for the direction of yields. Bond prices rise when yields fall, so a fall in yields is "bullish" for bond holders and a rise is "bearish".

Two episodes from the data show the difference.

The second example carries a lesson of its own: "the curve steepened" is incomplete without saying which maturities. Different segments can move in different directions on the same day.

How to read the curve

A few habits make the curve far more informative.

Separate policy from expectations. The front end mostly tells you what the market thinks the Fed will do over the next year or two. Compare the 3-month or 6-month yield with the current target range: a bill yielding well below the range implies expected cuts, and one above it implies the market sees some chance of increases.

Read the long end as two things at once. A rising 10-year yield can mean higher expected inflation, higher expected real growth, a higher term premium, or all three. The inflation part can be separated using inflation-protected securities; see TIPS and breakeven inflation: how inflation-protected bonds work and the decomposition on the The 10-year Treasury yield: why it matters and what drives it page.

Name the spread. 10-year minus 2-year and 10-year minus 3-month usually agree in sign but not always. In the monthly data the 10-year minus 2-year spread went negative in 1989 and 1998 while the 10-year minus 3-month did not, and in 2019 the reverse happened.

Watch the change, not just the level. A 1-point spread that has been widening for a year means something different from the same spread after a collapse from 3 points.

Remember who is buying. Central bank asset purchases, foreign reserve managers, pension funds and banks all buy Treasuries for reasons that have nothing to do with forecasting. Large Fed purchases from 2020 to 2022 (see Quantitative easing and tightening: the Fed’s balance sheet) are widely thought to have pushed long yields lower than expectations alone would have, which makes the slope in that period harder to read.

Why the curve matters for stocks and borrowers

The curve sets the menu of borrowing costs for the whole economy. A steep curve with a low front end is cheap short-term money and dearer long-term money. Banks, which typically borrow short through deposits and lend longer, tend to earn wider margins when the curve is steep and narrower ones when it is flat or inverted.

For stocks, the long end matters through valuation: future profits are discounted at a rate that starts from long Treasury yields. The front end matters through the cost of short-term credit and through what investors can earn on cash. When bills yield 5%, as they did through much of 2023 and 2024, cash is real competition for every other asset. The article How interest rates affect stocks: discount rates and equity duration works through those channels with market data; How the economy affects stocks: growth, inflation, jobs and rates puts the curve alongside the other macro signals.

Pitfalls and limitations

• The term premium is invisible. Every reading of "what the market expects" quietly assumes a term premium. Change that assumption and the implied rate path changes. Published estimates come from models and disagree with each other.
• Constant-maturity yields are fitted, not traded. The H.15 curve is a smooth estimate. It is excellent for comparing dates but is not a price at which you can buy a specific security.
• Measurement conventions differ. The monthly 3-month series in the long-history chart (TB3MS) is a secondary-market rate quoted on a discount basis, while the daily constant-maturity series is fitted from bond-equivalent yields. The gap is small but can flip the sign of a spread that sits near zero.
• One country's curve. The US curve is influenced by global demand for dollar assets. Moves abroad can move US long yields even when nothing changed at home.
• The slope is not a forecast you can trade on mechanically. A curve that predicted well in the past can mislead in a world with different inflation, a different Fed balance sheet or different buyers. The inversion that began in 2022 is the best recent reminder.

Yield-curve measures compared

How Finabase does it

Finabase does not draw a yield-curve chart or calculate spreads for you. What it does give you is the raw material.

• The Treasury yields list shows ten points on the curve as live quotes: 1, 3 and 6 months and 1, 2, 3, 5, 10, 20 and 30 years (symbols such as US2Y=T and US10Y=T), with the usual change, 1-week to 2-year performance and 52-week range columns. Reading the Last column from top to bottom traces the curve; subtracting the 2-year from the 10-year gives you the 2s10s spread.
• Each maturity has its own symbol page, such as the 10-year yield and the 2-year, with a chart you can set to any range, so you can compare how the two ends moved over the same period.
• Index versions of the 5-, 10- and 30-year yields (^FVX, ^TNX, ^TYX) appear in the Bonds group on /indices, and Treasury futures (2-, 10- and 30-year contracts) are in the futures list.
• For context on rate decisions, the Fed news topic collects headlines.

The curves and spreads on this page were drawn by Finabase Research from public Federal Reserve data, not from an in-app feature.

Frequently asked questions

What does an upward-sloping yield curve mean?

An upward-sloping curve means investors demand higher yields to lend for longer, which is the normal state. It reflects a positive term premium and, often, expectations that short rates will stay level or rise. Since 1960 the 10-year yield has averaged about 1.4 points more than the 3-month bill in monthly data.

Who controls the yield curve?

No one controls the whole curve. The Federal Reserve sets the overnight policy rate, which anchors the shortest maturities; longer yields are set by buyers and sellers in the Treasury market based on expected rates, inflation and risk. The Fed can influence long yields indirectly, through guidance and by buying or selling bonds.

Why can the 10-year yield rise when the Fed cuts rates?

Because the 10-year depends on expected rates over a decade plus a term premium, not on today's policy rate. If a cut raises expectations of future inflation, or if investors demand more compensation for holding long bonds, long yields can climb while short yields fall. That combination is a bull steepening at the front and a bear move at the back.

Is the yield curve the same as the 2s10s?

No. The 2s10s (10-year minus 2-year yield) is one spread, a single number summarizing the slope between two points. The yield curve is the whole set of yields across maturities. Different spreads can tell different stories on the same day.

Where can I see the yield curve every day?

The U.S. Treasury publishes daily par yield curve rates, and the Federal Reserve republishes constant-maturity yields in its H.15 release. In Finabase, the /treasury page lists live yields for ten maturities from 1 month to 30 years.

Summary

The yield curve lines up Treasury yields by maturity. Its short end is pinned by the Fed's policy rate and the market's expectation of where that rate goes next; its long end adds up expected future short rates and a term premium that can only be estimated. It usually slopes upward, by about 1.4 points between the 3-month bill and the 10-year note on average since 1960, but it bends with the business cycle: steep and near zero at the front at the end of 2021, deeply inverted by July 2023, and upward sloping again from 4.07% to 5.37% on 28 October 2026. Read it by separating what the front says about policy from what the back says about inflation and risk, always name the spread you are using, and treat its signals as evidence rather than a forecast.