What is the yield curve, and why does an inversion matter?
By Akash Soni·
The curve is just government borrowing costs plotted by maturity. When short rates exceed long ones it has inverted, which has preceded every US recession in the past 60 years. On 19 October 2026 it was not inverted: the ten-year sat 0.38 points above the two-year.
The yield curve is a line plotting the yield on government debt against how long until it matures. Normally it slopes up, because lenders want more to be paid back in ten years than in three months. When short-term yields rise above long-term ones the curve has inverted, and in the United States that has preceded every recession of the past sixty years. On 19 October 2026 the curve was not inverted: the ten-year Treasury yielded 5.15 per cent against 4.77 per cent on the two-year.
How do I read a yield curve?
Along the bottom is time to maturity. Up the side is yield. Each point is the market's current required return for lending to the government for that length of time.
The US Treasury publishes the figures every business day as the daily par yield curve, which is free and is the series almost everyone quotes. On 19 October 2026 it read, in part: 4.11 per cent at one month, 4.27 at three months, 4.50 at one year, 4.77 at two years, 4.94 at five, 5.15 at ten, 5.53 at twenty and 5.48 at thirty.
Three shapes get names. Upward sloping, which is normal. Flat, where short and long are close. Inverted, where short is above long.
What is the spread, and which one should I watch?
Rather than read the whole curve, analysts take the difference between two points and call it the term spread.
Two are standard. The ten-year minus the two-year is the most quoted. The ten-year minus the three-month is the one Federal Reserve Bank of San Francisco research names as the most useful term spread for forecasting recessions.
On 19 October 2026 the ten-year minus two-year spread was plus 0.38 percentage points, and the ten-year minus three-month spread was plus 0.88 points. Both positive, so both say the same thing.
The distinction between the two measures is not academic. Reviews of the recession-prediction record find the result changes depending on which spread is used, so a claim about the curve's accuracy should name its measure.
Why would short rates ever exceed long rates?
Because the two ends are set by different things.
The short end tracks the policy rate closely. When a central bank raises rates to slow the economy, bills reprice almost immediately. How that transmission works is covered in how the Federal Reserve works.
The long end is the market's average expectation of short rates over the whole period, plus a premium for the risk of holding a long bond. So a ten-year yield below a two-year yield is the market saying it expects short rates to be lower on average over the next decade than they are now, which usually means it expects cuts, which usually means it expects the economy to weaken.
That is the whole mechanism. An inversion is not a cause of anything. It is a price that aggregates an expectation.
How good is the inversion signal, honestly?
Good on direction, poor on timing, and the research says both.
Michael Bauer and Thomas Mertens, writing for the San Francisco Fed in March 2018, found that every US recession in the past sixty years was preceded by a negative term spread, and that a negative spread was always followed by an economic slowdown and, except for one occasion, by a recession. The exception was the mid-1960s, when the curve inverted without a near-term recession following.
Two caveats belong next to that record.
The lead time is wide. Summaries of this work put it at roughly six to twenty-four months, which is too broad to act on. An investor who went to cash at the first inversion would in several episodes have spent more than a year out of a rising market before being right.
“Every U.S. recession in the past 60 years was preceded by a negative term spread, that is, an inverted yield curve.”
And the relationship is correlation. The same researchers warn that such correlations do not identify cause and effect, which complicates their interpretation. The curve is not a mechanism; it is a summary of what bond investors collectively expect.
What is the term premium, and why does it complicate the signal?
Because part of the long yield is compensation for risk rather than a forecast.
A ten-year yield can be decomposed, roughly, into the market's expectation of average short rates over the decade plus a term premium: extra return demanded for locking money up and bearing the risk that rates move against you. Both pieces are unobservable and have to be estimated.
That matters for reading an inversion. If the curve flattens because expected short rates fell, the signal is about growth. If it flattens because the term premium compressed, for example because a large buyer wanted long bonds for reasons unrelated to the economy, the same shape carries a different message.
Central bank bond buying is the obvious case. Programmes that purchase long-dated government debt push the term premium down by design, which flattens the curve without anyone expecting a recession. Pension and insurance demand for long assets does the same thing more quietly.
None of this makes the curve useless. It is the main reason researchers are careful to call the relationship a correlation, and the main reason each inversion produces a genuine argument about whether the mechanism still holds.
Who actually sets these yields?
Nobody sets them. They are auction and secondary market prices.
The Treasury decides how much debt to issue at each maturity and holds regular auctions. Buyers bid, and the yield is the result. After issuance the bonds trade continuously, and the Treasury's daily par yield curve is derived from those market prices rather than announced as policy.
That is what makes the curve worth watching: it is one of the few macroeconomic indicators that is a market price rather than a survey or a statistical estimate, so it updates continuously and cannot be revised after the fact. The jobs report gets restated; the ten-year yield on a given day does not.
What is the curve saying in October 2026?
That the market is not pricing an imminent downturn through this channel.
The curve was upward sloping on 19 October 2026 at both standard measures. Across the 200 trading days of 2026 up to that date, the ten-year minus two-year spread ranged from plus 0.20 to plus 0.74 percentage points and was never negative.
One detail is worth noting and not over-reading. The thirty-year yield, at 5.48 per cent, sat marginally below the twenty-year at 5.53. A small dip at the very long end is a recurring feature of the curve, driven by supply and demand for specific maturities rather than by a macroeconomic signal, and it is not what anyone means by an inverted curve.
For context on the other side of the ledger, the National Bureau of Economic Research has not dated a US recession since the two-month downturn from February to April 2020.
Does the curve predict anything besides recessions?
It is used for several things, with decreasing reliability.
It is the discount curve. Long-dated cash flows, from pensions to infrastructure projects, are valued against it, so a shift in the curve changes present values across the economy whether or not any recession follows.
“Such correlations in the data do not identify cause and effect, which complicates their interpretation.”
It sets bank economics. Banks borrow short and lend long, so the slope is close to a direct read on lending profitability. A flat or inverted curve compresses that spread, which is why bank shares react to curve moves rather than just to rate levels.
It anchors mortgage pricing. US thirty-year mortgage rates track the ten-year Treasury far more closely than the policy rate, which is why a Fed cut does not automatically lower mortgage rates.
What does the curve not tell you?
Four things it is routinely asked about and cannot answer.
When. The lead time between inversion and recession has run from months to years.
How deep. The curve's shape carries no information about the severity of any downturn that follows.
Whether this time is different. Every inversion produces that debate, and the honest position is that a sample of a dozen episodes cannot settle it.
Anything about equities directly. The curve is a government bond price. Its relationship to share prices runs through expectations and discount rates, not through any mechanical link, and the difference between stocks and bonds matters for exactly this reason.
The short version
The yield curve is government borrowing costs by maturity, published daily and free. An inversion means short-term yields exceed long-term ones, which has preceded every US recession in the past sixty years with one false positive and a lead time measured in quarters to years.
On 19 October 2026 it was not inverted, and it had not inverted at any point during 2026.
This article explains a market measure. It is not investment advice, and nothing here is a forecast or a recommendation to buy or sell anything.
Frequently asked questions
What is the yield curve?
A line plotting the yield on government debt against time to maturity, from one month out to thirty years. The US Treasury publishes the figures each business day as the daily par yield curve, which is free and is the series most analysts quote. Normally the curve slopes upward, because investors require more to be repaid in ten years than in three months.
What does an inverted yield curve mean?
That short-term yields are above long-term ones. Mechanically it means the market expects short rates to average lower over the long horizon than they are today, which usually implies expected rate cuts and therefore an expected weakening of the economy. It is a price that aggregates an expectation, not a cause of anything.
Is the yield curve inverted right now?
No. On 19 October 2026 the three-month Treasury bill yielded 4.27 per cent, the two-year note 4.77 per cent and the ten-year note 5.15 per cent, so the curve sloped upward. The ten-year minus two-year spread was plus 0.38 percentage points and the ten-year minus three-month spread was plus 0.88 points. The spread was never negative at any point in 2026.
Does an inverted yield curve always mean a recession?
Close to always on the historical record, but not reliably on timing. San Francisco Fed research found every US recession in the past sixty years was preceded by a negative term spread, and that a negative spread was always followed by a slowdown and, except once, by a recession. The exception was the mid-1960s. The lead time has run from roughly six to twenty-four months.
Which yield curve spread should I watch?
The ten-year minus two-year is the most quoted, but San Francisco Fed research names the ten-year minus three-month as the most useful term spread for forecasting recessions. The choice matters: reviews of the prediction record find the result changes depending on which spread is used, so any claim about the curve's accuracy should name its measure.
Why does the yield curve invert before a recession?
Because the two ends are set by different forces. Short yields track the policy rate closely, so they rise when a central bank tightens. Long yields reflect the market's average expectation of short rates over the whole period plus a risk premium. A long yield below a short one is the market pricing in future cuts, which it typically does when it expects growth to weaken.
How long after an inversion does a recession arrive?
The documented lead time is wide, roughly six to twenty-four months, which is too broad to act on. An investor who moved to cash at the first sign of inversion would in several historical episodes have spent more than a year out of a rising market before being proved right on direction. The signal carries information about direction and almost none about timing.
Does the yield curve affect mortgage rates?
Yes, and more than the policy rate does. US thirty-year mortgage rates track the ten-year Treasury yield far more closely than the federal funds rate, because a mortgage is a long-dated asset priced against long-dated government debt. This is why a central bank rate cut does not automatically lower mortgage rates, and sometimes coincides with them rising.
Why do banks care about the shape of the yield curve?
Because banks borrow short and lend long, so the slope of the curve is close to a direct read on lending profitability. A steep curve widens the spread between what a bank pays for deposits and earns on loans; a flat or inverted curve compresses it. That is why bank share prices react to changes in the curve's shape rather than only to the level of rates.
Is a thirty-year yield below a twenty-year yield an inversion?
Not in the sense anyone means. On 19 October 2026 the thirty-year yielded 5.48 per cent against 5.53 on the twenty-year, a small dip at the very long end. That pattern recurs and is driven by supply and demand for specific maturities rather than by a macroeconomic signal. An inverted curve means short yields above long ones, measured at the standard spreads.
Sources
What each one is, and whose it is.
- 1
Daily Treasury par yield curve rates, 2026 series (read for 19 October 2026), US Department of the Treasury (October 19, 2026)
DatasetIndependent of the vendor - 2
Economic forecasts with the yield curve, FRBSF Economic Letter 2018-07, Michael D. Bauer and Thomas M. Mertens, Federal Reserve Bank of San Francisco (March 5, 2018)
PaperIndependent of the vendor - 3
Information in the yield curve about future recessions, FRBSF Economic Letter 2018-20, Federal Reserve Bank of San Francisco (August 27, 2018)
PaperIndependent of the vendor - 4
US business cycle expansions and contractions, National Bureau of Economic Research
DatasetIndependent of the vendor