Data as of the close on Wednesday 18 November 2026, unless a date is given.
Welcome to Third Moment. This first note does two things: it reviews where the U.S. economy stands, and it shows how research will be done here. We do not make predictions. We publish probabilities, we date them, and we score them in public once the outcome is known. The name is the method: the third moment of a distribution is its skewness, and most of what matters in markets sits in the shape of the distribution, not in its central point.
The subject is the most important price in finance: the yield on the 10-year U.S. Treasury note.
The short version
- The 10-year yield closed at 5.07% on 18 November. It ended 2025 at 4.16%, touched a low of 3.95% on 27 February, and has closed above 5% on 28 trading days since mid-September.
- Three forces pushed it there: an oil shock after the war with Iran began on 28 February, an economy that grew 3.4% annualised in the third quarter, and a Federal Reserve that raised rates in September for the first time since 2023 and is all but certain to do it again on 9 December.
- Almost all of the rise is in real yields. Market-implied inflation over the next ten years has barely moved. That tells us investors still trust the Fed, which is good news today and the main source of risk tomorrow.
- Our base case for 31 March 2027 is a 10-year yield between 4.85% and 5.20% (40%). The distribution is skewed: we put 35% on yields above 5.20% and 25% on yields below 4.85%.
- At these levels the carry is real, but over four months it only absorbs a rise of about 25 basis points before a holder loses money.
How we got here
Source: TradingView (TVC:US10Y daily closes, TVC:US02Y weekly closes); Federal Reserve for the target range. Scroll or pinch to zoom.
The year opened quietly. The Fed had cut its target range to 3.50% to 3.75% in December 2025, inflation was running at 2.4%, and the 10-year sat just above 4%.
On 28 February the United States and Israel struck Iran, and within days Iran closed the Strait of Hormuz, the passage for roughly a fifth of the world's oil (timeline). Brent crude went from $73 to above $100 in two weeks. A ceasefire in April and a memorandum in June brought prices back toward $72 by early July, then the truce collapsed on 8 July and oil climbed again, to a weekly close of $117 in late October.
Source: TradingView (TVC:UKOIL), weekly closes through 18 Nov 2026.
Oil is back near $93 as Gulf exports recover, but the damage to inflation was done twice. Headline consumer price inflation rose from 2.4% in February to 4.2% in May, eased over the summer, and was back at 3.7% in September.
Source: U.S. Bureau of Labor Statistics, via TradingView. The 12-month change for October 2026 cannot be computed because October 2025 prices were not collected during the government shutdown.
The October report, published on 10 November, is the one that moved markets. Prices rose 0.4% on the month and core prices 0.3% for the third month running, both above expectations (BLS). Because October 2025 is missing, there is no official 12-month figure; the consensus estimate was 3.9%. What worries the Fed is less the level than the breadth: over the last three months core prices have risen at an annualised 3.7%, the fastest since April 2024, and services at 5.0% (KPMG). The shock is no longer only about energy.
An economy that will not slow down
Higher oil prices usually act as a tax on growth. Not this time, or not yet.
| Indicator | Latest | Previous | Date of data |
|---|---|---|---|
| Real GDP, annualised | 3.4% | 2.2% | Q3 2026 (advance) |
| Consumer spending, annualised | 3.8% | Q3 2026 | |
| Business investment, annualised | 12.7% | Q3 2026 | |
| Payrolls, monthly change | +165,000 | +60,000 | October 2026 |
| Unemployment rate | 4.1% | 4.2% | October 2026 |
| Average hourly earnings, 12 months | 3.1% | October 2026 | |
| ISM manufacturing index | 55.5 | 54.5 | October 2026 |
| ISM services index | 54.5 | 54.9 | October 2026 |
| Federal deficit | $2.0tn, 6.2% of GDP | 5.8% of GDP | Fiscal year 2026 |
Sources: BEA via PBS/AP; BLS via Fox Business; ISM via TradingView's economic calendar; CBO, 9 Nov 2026.
Two engines explain the resilience. The first is investment in artificial intelligence: spending on information-processing equipment grew at a 37% annualised rate in the third quarter. The second is fiscal policy: a deficit of 6.2% of GDP with unemployment at 4.1% is a large stimulus by any historical standard.
There is one clear casualty. Thirty-year mortgage rates have been above 7% since mid-September, and existing home sales fell to an annualised 3.75 million in October. Housing is where tight money bites first, and it is biting.
A Fed that is hiking again
The Fed raised its target range to 3.75% to 4.00% on 16 September, by a unanimous vote, its first increase since July 2023. It held in October, but two officials dissented in favour of a second hike. The minutes of that meeting, released on 18 November, say that nearly all officials now expect a quarter-point increase on 9 December, and that two of them raised the possibility of a half-point (AP).
The committee is not of one mind. Governor Christopher Waller says inflation is "headed in the wrong direction" and backs a half-point. New York Fed President John Williams answers that the situation does not call for "dramatic action". Chair Kevin Warsh has declined to give forward guidance at all, on the grounds that it would lock the Fed into a path.
Markets have filled the silence. A quarter-point in December is fully priced, and futures imply more than 80 basis points of tightening by September 2027, up from about 50 before the October inflation report (Bloomberg). That would take the policy rate to about 4.7%.
What is inside a 5% yield
A 10-year yield can be split into two parts: the real yield, which is what investors earn after inflation, and breakeven inflation, which is the inflation rate they expect and are paid for.
Source: Federal Reserve Bank of St. Louis (FRED series DFII10 and T10YIE), via TradingView, weekly, latest available.
Since the eve of the war, the 10-year yield has risen by about 1.1 percentage points. The real yield went from 1.72% to 2.89%. Breakeven inflation went from 2.25% to 2.19%: it is lower than before the oil shock.
This is the single most important fact in the bond market today. Investors are not saying that inflation is out of control. They are saying that the Fed will have to keep real interest rates high for a long time to control it, and that they want more compensation to hold long bonds while the Treasury borrows $2 trillion a year. The term premium, the extra yield for holding a 10-year note instead of rolling short bills, was estimated at 0.87% in early November on the New York Fed's model, after peaking in October at its highest since 2014 (The Private Banker).
It also defines the two tails. If that trust holds, real yields near 2.9% are the highest since 2008 and are attracting record inflows into bond funds. If it breaks, breakevens have a long way to rise, and nothing in today's price pays for that.
One reassurance on supply: on 4 November the Treasury said it expects to keep the size of its note and bond auctions unchanged "for the next several quarters" (details). The deficit is large, but it is not about to surprise the market with more long-dated paper.
Three paths to the end of March
We look out to 31 March 2027. That horizon covers three Fed meetings (9 December, late January and mid-March), four inflation reports and four jobs reports.
Path 1: the range holds (40%). The Fed raises by a quarter-point in December and once more by March. Core inflation stays near 0.3% a month, oil stays between $85 and $100, growth cools a little under the weight of 7% mortgages. All of this is already priced, so the 10-year moves sideways between 4.85% and 5.20%. This is the most likely path because it requires no surprise.
Path 2: yields fall (25%). Something gives. The candidates are visible: a deeper sell-off in technology and AI shares, like the one that pulled the 10-year down to 4.84% in early November; a faster normalisation of Gulf oil exports; a housing market that drags consumption down with it. The Fed hikes in December and then signals a pause. The 10-year falls below 4.85%, and in the more severe version (7%) below 4.50%.
Path 3: new highs (35%). Inflation keeps broadening. Monthly core readings of 0.4% would bring half-point hikes into play, and a third leg up in oil would do the rest. The 10-year moves above 5.20%, and in the more severe version (13%) above 5.50%, a level not seen in more than two decades. This is where the skew lies: breakeven inflation has room to rise, and a Fed that refuses to give guidance leaves the market to set the risk premium on its own.
These are our own judgements, not the output of a model. They will be scored against the Treasury's Daily Par Yield Curve rate for 31 March 2027.
The probability-weighted average of this distribution is about 5.07%, almost exactly today's yield. We are not calling the direction. We are saying that the two sides are not symmetrical: the right tail is nearly twice as heavy as the left (13% against 7%).
What a 5% yield pays for
Carry is the argument for owning bonds here, so it is worth doing the arithmetic. Take a 10-year note bought at par on 18 November with a yield of 5.07%, held to 31 March 2027.
Our calculation. Semi-annual coupon, repriced at the yield on the horizontal axis with 9.6 years to maturity, accrued interest included, no financing cost.
If the yield does not move, the holder earns 1.8% over the period. With a duration of about 7.4 years at the horizon, each rise of 10 basis points costs roughly 0.74%. The break-even is a yield of 5.32%: above it, four months of carry are gone. In the five scenarios above, the returns are approximately +7.5%, +4.9%, +2.2%, -0.2% and -2.8%, for a probability-weighted +1.9%.
That is only a little more than the 1.6% a three-month Treasury bill pays over the same period at today's 4.26%, without the risk. Our reading is therefore a patient one. A 5% yield with a real return near 2.9% is attractive for an investor who measures the result over years, and we would rather add duration gradually than wait for a peak that nobody can time. Over the next four months, though, the 10-year pays for the central case and not for the right tail. This fits how we look at risk more broadly: constructive on an economy that keeps growing, cautious about the price paid for it.
Our calls, to be scored
Every probability below is dated 19 November 2026 and will be marked right or wrong on this site.
| # | Call | Probability | Resolves |
|---|---|---|---|
| 1 | The Fed raises its target range by exactly 25 basis points on 9 December | 80% | 9 Dec 2026 |
| 2 | The 10-year yield ends the year between 4.85% and 5.35% | 65% | 31 Dec 2026 |
| 3 | The 10-year yield closes above 5.31%, its October high, at least once | 30% | by 31 Mar 2027 |
| 4 | The 10-year yield closes below 4.75% at least once | 20% | by 31 Mar 2027 |
| 5 | The 10-year yield is above 5.20% on 31 March 2027 | 35% | 31 Mar 2027 |
Yields are the 10-year rate in the U.S. Treasury's Daily Par Yield Curve Rates.
What would change our mind
- Core inflation at 0.2% or below in the November and December reports (10 December and mid-January) would move weight from Path 3 to Path 2.
- Brent above $110 again would do the opposite.
- Breakeven inflation above 2.5%. That would mean the market has started to doubt the Fed, and it would make Path 3 our base case.
- Payrolls turning negative or unemployment at 4.4% would put the low-yield tail back in play quickly.
We will come back to this note on 31 March 2027, whatever the result.
This article is for information only and is not investment advice. Market data: TradingView, pulled on 19 November 2026. Charts show closing levels and may differ slightly from official Treasury par yields.
