Skip to content

The yield curve and inversions

By Investor Wisdom Daily Team·

Priya (a hypothetical saver) is comparing certificates of deposit at her bank. The 1-year CD pays 4.3%. The 5-year CD pays 3.8%. That feels backwards: why would the bank pay less for locking her money up longer? Rate tables like this are a small, everyday version of what headlines call an “inverted yield curve”. This lesson shows how to read the real curve from official data, why it usually slopes the other way, and what inversions have and have not told us in the past.

A yield is the yearly return a bond offers at today’s price. The yield curve lines up the yields of one borrower’s debt from shortest to longest maturity on a single day. Usually longer loans pay more, so the line slopes upward. When short-term yields rise above long-term ones, the curve is inverted. That shape reflects what markets expect interest rates to do and how much extra they demand for tying money up longer; it is a reading of today’s prices, not a prediction machine.

The yield curve is a simple chart. Maturity runs along the bottom axis, from one month to 30 years, and yield runs up the side. Each point shows what one issuer’s debt yields at that maturity on one date. In the US, “the yield curve” almost always means US Treasury securities, which share one issuer, so differences between points come from time, not credit quality.

If you are new to charts, Reading a chart covers axes and scales. This lesson assumes you know the basics from How a bond works: prices fall when yields rise, and longer bonds swing more.

You do not need a news headline to see the curve. Two free official sources publish it.

The Treasury table. The US Treasury publishes a table called “Daily Treasury Par Yield Curve Rates” (Treasury). As of November 2026, each row is a trading date and each column is a maturity: 1 Mo, 1.5 Mo, 2 Mo, 3 Mo, 4 Mo, 6 Mo, 1 Yr, 2 Yr, 3 Yr, 5 Yr, 7 Yr, 10 Yr, 20 Yr and 30 Yr. Reading one row from left to right gives you that day’s curve.

A “par yield” is the coupon rate at which a bond would be priced at exactly its face value. Treasury’s methodology page says its official curve “is a par yield curve derived using a monotone convex method,” built from “indicative, bid-side market price quotations (not actual transactions)” collected “at or near 3:30 PM each trading day” (Treasury methodology). In plain terms, the numbers are a smoothed estimate from dealer quotes, not prices someone actually traded at.

FRED. The St. Louis Fed’s FRED database carries the same story as time series. As of November 2026:

Series What it shows Units
DGS2 2-year Treasury constant maturity yield Percent, daily
DGS10 10-year Treasury constant maturity yield Percent, daily
T10Y2Y 10-year minus 2-year Percent, daily

“Constant maturity” means each yield is estimated for an exact maturity, such as precisely 2 or 10 years, rather than taken from one particular bond whose remaining life shrinks every day. A “spread” is simply the difference between two yields. T10Y2Y is the spread most headlines mean. When it drops below zero, the curve between 2 and 10 years is inverted.

Worked example 1: reading a spread yourself

Section titled “Worked example 1: reading a spread yourself”
  1. Pick a date in the Treasury table.
  2. Read the 2 Yr and 10 Yr columns.
  3. Subtract the 2-year yield from the 10-year yield.
  4. A positive result means the curve slopes up between those points; a negative one means it is inverted there.

The numbers below are hypothetical, chosen only to show each shape.

Shape 2-year yield 10-year yield 10y minus 2y
Normal (upward sloping) 4.00% 4.50% +0.50 points
Flat 4.00% 4.00% 0 points
Inverted 4.30% 3.80% -0.50 points

Spreads are often quoted in basis points, where one basis point is one hundredth of a percentage point. So +0.50 points is 50 basis points, and -0.50 points is -50 basis points. Priya’s CD table from the opening has the same shape as the inverted row: the 5-year rate is 0.5 points below the 1-year rate.

Two forces shape the curve.

The first is what investors expect short-term rates to do. If markets expect short rates to fall, locking in a long rate today can look attractive, which pulls long yields down relative to short ones.

The second is the term premium: extra yield that investors ask for tying their money up longer and carrying more price risk. Worked example 3 below shows why that risk is real. Researchers do not agree on how large the term premium is at any time. A 2018 San Francisco Fed letter noted that it “has been close to zero or even negative” (SF Fed, 2018), so long yields do not always carry a cushion.

Worked example 2: how expectations can invert the curve

Section titled “Worked example 2: how expectations can invert the curve”

Imagine a hypothetical world with no term premium. A 1-year bond pays 5% today, and everyone expects the 1-year rate to be 3% a year from now. What should a 2-year bond pay?

  • Roll two 1-year bonds: $10,000 × 1.05 × 1.03 = $10,815 after two years.
  • Lock one 2-year bond at 4% a year: $10,000 × 1.04² = $10,816.

The two routes end up almost exactly level, so a 2-year yield of about 4% is “fair” here, below the 5% 1-year yield. Nothing about default risk was involved: the curve inverted purely because short rates were expected to fall. That is one reason inversions can show up when markets expect the central bank to cut rates, often because they expect the economy to weaken.

Worked example 3: spread and price sensitivity

Section titled “Worked example 3: spread and price sensitivity”

Take two hypothetical zero-coupon bonds, each repaying $1,000 at maturity, with no coupons along the way. As in How a bond works, we use annual compounding to keep the arithmetic simple. Now suppose every yield rises by 1 percentage point.

Bond Price before Price after 1-point rise Change
2-year, 4% to 5% $924.56 $907.03 -$17.53 (-1.90%)
10-year, 4.5% to 5.5% $643.93 $585.43 -$58.50 (-9.08%)

The same 1-point move costs the 10-year bond about 4.8 times as much in percentage terms. That gap is a big part of why investors usually want more yield for longer maturities, and why the normal curve slopes upward.

This example ignores coupons, semiannual compounding, taxes, fees and bid/ask spreads. Real bonds will show different numbers, though the direction of the effect is the same.

Inversions get attention because of their history. The 2018 San Francisco Fed letter by Bauer and Mertens, which used data from January 1955 to February 2018, a period with nine recessions, and measured the spread “as the difference between ten-year and one-year Treasury yields,” stated:

  • “Every U.S. recession in the past 60 years was preceded by a negative term spread, that is, an inverted yield curve.”
  • “[A] negative term spread was always followed by an economic slowdown and, except for one time, by a recession.”
  • “The delay between the term spread turning negative and the beginning of a recession has ranged between 6 and 24 months.”

The “one time” was in the mid-1960s, when an inversion was followed by a slowdown but not an official recession.

That is an observed pattern in past data, not a forecast. It does not tell anyone that a recession will happen, when, or how markets will respond.

Also note which spread is being used. The SF Fed letter used 10-year minus 1-year. FRED’s headline series is 10-year minus 2-year. Other researchers use other pairs, such as a 3-month bill against the 10-year. These can disagree, and headlines rarely say which one they mean.

Open the CD calculator and recreate Priya’s choice with $10,000:

  • A 60-month CD at 3.8% APY earns about $2,050 over five years with yearly compounding.
  • A 12-month CD at 4.3% APY earns about $430 in its first year. If she could renew at 4.3% every year for five years, she would earn about $2,343, but nobody can promise that renewal rate.

Notice what the inverted rate table is really asking: will short-term rates still be high when the 1-year CD matures? Run a 12-month CD at 3% and at 2% to see what a lower renewal rate would earn in a later year. The rates here are hypothetical, not quotes from any bank.

  1. “An inversion means a recession is coming.” In the SF Fed study, inversions came before past recessions with a 6 to 24 month lag and one false alarm. That is a historical pattern, not a schedule.
  2. Mixing up spreads. 10-year minus 2-year, 10-year minus 1-year and 10-year minus 3-month are different measures and can give different answers on the same day.
  3. Treating Treasury par yields as trade prices. They are smoothed estimates built from dealer quotes, as Treasury’s methodology explains.
  4. Using the curve to time stocks. The curve describes Treasury yields. It says nothing direct about stock prices or any single company.
  5. Assuming “longer always pays more”. Example 2 shows how expectations alone can make a 2-year yield lower than a 1-year yield.
  • Small sample, dated study. The record rests on nine recessions up to 2018, so one more case could change the picture a lot.
  • Long, variable lag. A 6 to 24 month range spans 18 months, too wide to time anything.
  • The term premium debate. Some argue that a low or negative term premium makes inversions less meaningful. The SF Fed authors argued the signal still held. Treat this as an open question.
  • Not about stocks directly. The curve describes Treasury yields. It says nothing direct about stock prices or any single company.
  • Indicative quotes. Treasury’s par yields come from dealer quotes, not trades.

For how bonds and stocks fit together in a mix, see Asset allocation basics and Risk and diversification. For how prices form in the first place, see How markets work.

  1. On a hypothetical day the 2-year yield is 4.35% and the 10-year yield is 4.10%. What is the 10-year minus 2-year spread, in points and basis points?
Answer to question 1

4.10 − 4.35 = −0.25 points, or −25 basis points. That part of the curve is inverted.

  1. Every yield rises by 1 point. Which loses more value in percentage terms, a 2-year or a 10-year zero-coupon bond?
Answer to question 2

The 10-year bond. In worked example 3 it fell about 9.08%, against about 1.90% for the 2-year, because more of its value sits far in the future.

  1. Can a curve invert even if investors see no extra risk in longer bonds?
Answer to question 3

Yes. In worked example 2, with no term premium, an expected fall in short rates from 5% to 3% made a 2-year yield of about 4% fair, below the 5% 1-year yield.

  1. Which spread did the 2018 SF Fed study use, and which does FRED’s T10Y2Y series show?
Answer to question 4

The study used 10-year minus 1-year. T10Y2Y is 10-year minus 2-year. Check which spread a headline is quoting before comparing it with the study.

  • The yield curve plots Treasury yields against maturity for one date.
  • You can read it yourself from Treasury’s daily par yield table or FRED’s DGS2, DGS10 and T10Y2Y series.
  • A negative 10-year minus 2-year spread means that part of the curve is inverted; expectations of falling short rates can cause it on their own.
  • Longer bonds lose more value when yields rise, which helps explain why longer maturities usually yield more.
  • Inversions preceded past US recessions in the SF Fed’s 2018 study (which used 10-year minus 1-year), with one mid-1960s false alarm and a 6 to 24 month lag. That is history, not a forecast.

Key terms

For educational purposes only. This material does not provide investment advice.