Markets
10-year yield's biggest one-day drop since 2024
The 10-year Treasury yield fell 17 bp to 4.96% on Oct 22, 2026, its biggest drop since Aug 2024, as the Nasdaq slid 2.6%. What it meant for stocks and bonds.
The 10-year Treasury yield fell 17 basis points in a single session on Thursday, October 22, 2026, from 5.13% to 4.96%, according to the U.S. Treasury's daily par yield curve. That is the largest one-day decline since August 2, 2024, and it pushed the benchmark yield back below 5% for the first time since September 22. On the same day the Nasdaq Composite fell 2.6% to 26,008.86 and the S&P 500 lost 1.21% to 7,663.43, while the Dow finished essentially flat. Bonds rallied and growth stocks fell together: a classic flight to safety, triggered this time by doubts about AI spending rather than by weak economic data.
4.96%
10-year Treasury yield
close, Oct 22 (Treasury par curve)
−17 bp
One-day change
largest since Aug 2, 2024
−2.6%
Nasdaq Composite
26,008.86, worst day since June
−1.21%
S&P 500
7,663.43, worst day since July
−5.8%
Nasdaq below its record
Oct 6 close 27,599.89
1.26 pp
10-year yield minus CPI
4.96% − 3.7% (derived)
What happened on October 22
The session started with Meta. Its third-quarter report on Wednesday evening showed revenue up 26% but quarterly capital spending of $34.9 billion and negative free cash flow (our Meta and Tesla earnings report has the details). The shares closed down about 7% on Thursday, according to Xinhua's market wrap carried by Bernama, and the selling spread across the companies that supply and finance the AI build-out:
| Stock or index, Oct 22, 2026 | Move | What was cited |
|---|---|---|
| Meta Platforms | about −7% | Q3 capex and spending outlook |
| Philadelphia Semiconductor Index | nearly −6% | AI capex worries |
| Lam Research | −11.8% | warned supply constraints will delay shipments |
| GE Vernova | −10.6% | AI financing worries after Meta's outlook |
| CoreWeave | about −10% | cost of capital, debt-heavy balance sheet |
| Intel, AMD, Micron, Nvidia | each more than −4% | sector sell-off |
| Tesla | about −3% | despite an adjusted EPS beat |
Six of the eleven S&P 500 sectors still closed higher. Consumer staples rose 3.11% and utilities 1.98%, while communication services fell 3.01% and technology 3%, per the same Xinhua report. That split is typical of a rotation out of growth and into defensive stocks rather than a broad liquidation. Investing.com reported that CoreWeave's junk-rated bonds were trading at yields near 15% in the secondary market, which shows how much AI-related credit, not only equity, was being repriced.
What makes the day unusual is the backdrop. Brent crude settled at $113.17 a barrel, up 4% and its highest since May, after further attacks on Saudi targets and reports of possible U.S. and Israeli strikes on Iran after the November 3 midterm elections. Higher oil usually pushes yields up through inflation expectations. Yahoo Finance's live coverage summed up the reversal: concerns about slower AI infrastructure spending outweighed inflation worries from higher oil, and money moved into Treasuries.
The labor market gave no reason for the rally either. The Labor Department reported that initial jobless claims fell by 6,000 to 194,000 in the week ending October 17, with the four-week average at 197,750 — both low by historical standards. Bonds rallied on risk appetite, not on weak data.
The move across the yield curve
The decline was concentrated in the 2- to 10-year part of the curve. Bills barely moved, because they are anchored by the Fed's policy rate, which markets did not expect to change at next week's meeting.
Measured by Treasury's daily par yields, a 17-point drop in the 10-year has happened only a handful of times in the last five years. The table lists every day since January 2022 when the 10-year fell by 17 basis points or more.
| Date | 10-year change | 2-year change | Context |
|---|---|---|---|
| Nov 10, 2022 | −30 bp | −27 bp | cooler-than-expected October 2022 CPI |
| Sep 28, 2022 | −25 bp | −23 bp | Bank of England gilt-market intervention |
| Mar 10, 2023 | −23 bp | −30 bp | Silicon Valley Bank failure |
| Nov 14, 2023 | −19 bp | −22 bp | cooler October 2023 CPI |
| Aug 2, 2024 | −19 bp | −28 bp | weak July 2024 jobs report |
| Mar 17, 2023 | −17 bp | −33 bp | banking stress |
| Oct 22, 2026 | −17 bp | −16 bp | AI-spending sell-off in tech |
Two things stand out. Every earlier day on the list was driven by macro or banking news; October 22 was driven by one sector's earnings. And on most earlier days the 2-year fell faster than the 10-year, because traders were pricing rate cuts. This time the 2-year and 10-year fell almost equally, which fits a move into safe assets more than a change in Fed expectations.
How the 10-year got to 5%: the 2026 context
The rally needs to be seen against the year's sell-off in bonds. The 10-year yield ended 2025 at 4.18%, hit its 2026 low of 3.97% on February 27, and climbed to 5.31% on October 5 — the highest close in Treasury's data since at least 2022 — as headline CPI rose from 2.4% in January to 4.2% in May and the Fed raised rates in September.
| 10-year | 2-year | Fed funds (upper) | |
|---|---|---|---|
| Sep 2024 | 3.81 | 3.66 | 5 |
| Oct 2024 | 4.28 | 4.16 | 5 |
| Nov 2024 | 4.18 | 4.13 | 4.75 |
| Dec 2024 | 4.58 | 4.25 | 4.5 |
| Jan 2025 | 4.58 | 4.22 | 4.5 |
| Feb 2025 | 4.24 | 3.99 | 4.5 |
| Mar 2025 | 4.23 | 3.89 | 4.5 |
| Apr 2025 | 4.17 | 3.6 | 4.5 |
| May 2025 | 4.41 | 3.89 | 4.5 |
| Jun 2025 | 4.24 | 3.72 | 4.5 |
| Jul 2025 | 4.37 | 3.94 | 4.5 |
| Aug 2025 | 4.23 | 3.59 | 4.5 |
| Sep 2025 | 4.16 | 3.6 | 4.25 |
| Oct 2025 | 4.11 | 3.6 | 4 |
| Nov 2025 | 4.02 | 3.47 | 4 |
| Dec 2025 | 4.18 | 3.47 | 3.75 |
| Jan 2026 | 4.26 | 3.52 | 3.75 |
| Feb 2026 | 3.97 | 3.38 | 3.75 |
| Mar 2026 | 4.3 | 3.79 | 3.75 |
| Apr 2026 | 4.4 | 3.88 | 3.75 |
| May 2026 | 4.45 | 3.98 | 3.75 |
| Jun 2026 | 4.44 | 4.14 | 3.75 |
| Jul 2026 | 4.75 | 4.28 | 3.75 |
| Aug 2026 | 4.75 | 4.34 | 3.75 |
| Sep 2026 | 5.29 | 4.88 | 4 |
| Oct 22 | 4.96 | 4.65 | 4 |
The chart shows the main story of the past two years: the Fed cut its upper bound from 5.50% to 3.75% between September 2024 and December 2025, yet the 10-year yield ended that stretch higher than it started (4.18% at the end of 2025 vs 3.81% at the end of September 2024). Long-term yields have been driven by inflation and term premium, not by the policy rate. Since September 16 the fed funds target range has been 3.75%–4.00%, and the effective rate has been 3.88% (Federal Reserve data via FRED).
Inflation is the other half. Headline CPI, from BLS data, ran at 2.4% in January and February 2026, jumped to 3.3% in March and peaked at 4.2% in May, then eased to 3.3%–3.5% over the summer before reaching 3.7% in September. Core CPI stayed in a 2.4%–2.8% band all year. Subtracting September CPI from the 10-year yield gives a rough real yield of 1.26 percentage points (4.96% − 3.7%); before Thursday it was 1.43 points (5.13% − 3.7%). This is a backward-looking proxy — the market's own measure is the 10-year TIPS yield — but it shows how much of the bond yield still exceeds current inflation.
Within October the 10-year has now given back 35 basis points from its peak:
| % | |
|---|---|
| Sep 30 | 5.29 |
| Oct 1 | 5.24 |
| Oct 2 | 5.28 |
| Oct 5 | 5.31 |
| Oct 6 | 5.27 |
| Oct 7 | 5.28 |
| Oct 8 | 5.22 |
| Oct 9 | 5.24 |
| Oct 13 | 5.21 |
| Oct 14 | 5.19 |
| Oct 15 | 5.1 |
| Oct 16 | 5.09 |
| Oct 19 | 5.15 |
| Oct 20 | 5.12 |
| Oct 21 | 5.13 |
| Oct 22 | 4.96 |
What it meant for a typical portfolio
The quickest way to translate a yield move into a bond-fund return is duration: price change ≈ −duration × change in yield. An intermediate-term bond fund with a duration of about 6 years therefore gained roughly 6 × 0.17 = 1.0% in price on Thursday, before interest income. A fund closer to the 10-year itself (duration around 8) gained about 1.4%.
The duration rule also puts the year in perspective. The 134-basis-point rise in the 10-year from February 27 to October 5 cost that same 6-year-duration fund roughly 6 × 1.34 = 8.0% in price, partly offset by higher coupons. Thursday's rally recovered about an eighth of it.
For stock investors, the day was a reminder of how concentrated the index is. The S&P 500 was still up 11.9% for the year at Thursday's close (7,663.43 vs 6,845.50 at the end of 2025), and the Nasdaq 11.9% (26,008.86 vs 23,241.99), but both now sit below their October 6 records — the S&P 500 by 2.0% and the Nasdaq by 5.8%. If your portfolio leans on a handful of mega-cap tech names, comparing it with the S&P 500 using time-weighted returns shows whether the concentration has paid off; the method is explained in how to compare your portfolio to the S&P 500.
Three scenarios from here
These are scenarios, not forecasts. Each one describes what would have to happen and what the duration arithmetic implies.
If the flight to safety continues
- AI-spending worries keep pulling money from tech into Treasuries and defensive sectors
- A further 25 bp fall in the 10-year would add about 1.5% to a 6-year-duration bond fund
- Staples and utilities would keep leading, as they did on Oct 22
If oil and inflation take over again
- Brent above $110 keeps headline CPI high; the November 10 CPI report shows how far energy has spread
- A return of the 10-year to its 5.31% October high (+35 bp) would cost that fund about 2.1%
- Markets would again price a December hike, pressuring both stocks and bonds
A third, middle path: the Fed holds rates on October 28 as markets expect, the Q3 GDP and PCE inflation data on October 29 arrive close to expectations, and yields settle between 4.9% and 5.1%. In that case, the next earnings reports from the largest AI spenders would matter more for the index than the bond market.
What to watch next
| Date (2026) | Event | Why it matters |
|---|---|---|
| Oct 27–28 | FOMC meeting (statement Oct 28) | first decision since the September hike; no new projections |
| Oct 29 | BEA: Q3 GDP (advance) and September personal income and outlays (PCE prices) | growth and the Fed's preferred inflation gauge |
| Nov 3 | U.S. midterm elections | oil and geopolitical risk cited in press reports |
| Nov 6 | BLS: October employment report | first payrolls number after the yield drop |
| Nov 10 | BLS: October CPI | whether energy is spreading into core inflation |
| Dec 8–9 | FOMC meeting with new Summary of Economic Projections | the meeting markets have priced for a possible hike |
Background on the Fed decision is in our October FOMC preview, and the September inflation numbers are in our CPI and 2027 COLA report.
Methodology and sources
- Treasury yields are daily par yield curve rates from the U.S. Treasury (CSV files for 2022–2026, downloaded October 23, 2026). One-day changes are the difference between consecutive closes, rounded to whole basis points; the "largest since" ranking compares every trading day from January 3, 2022 to October 22, 2026. Press reports quote slightly different intraday or market-close figures (Xinhua: 4.957%); we use Treasury's official curve throughout.
- Index levels for the S&P 500 and Nasdaq Composite are daily closes from FRED (St. Louis Fed). The S&P 500 close for October 22 (7,663.43) was not yet in FRED and comes from the Xinhua wrap; it matches FRED's October 21 close of 7,757.17 minus 1.21%. Records are the highest closes in FRED's series since 2016.
- Fed funds are the target range upper bound (DFEDTARU) and effective rate (DFF) from Federal Reserve data via FRED. CPI is year-over-year CPI-U from BLS data via FRED.
- Derived numbers: real yield proxy = 10-year yield − latest CPI y/y; bond-fund price change = −duration × yield change (a first-order estimate that ignores convexity and coupon income); 60/40 example uses the S&P 500's −1.21% and a 6-year duration. Stock moves come from the Xinhua/Bernama, Yahoo Finance and Investing.com reports listed below.
GetHossa's stock research shows valuation and reported numbers for the companies named here; nothing in this article is a recommendation to buy or sell any security.
This article is for information only and is not investment advice.
Frequently asked questions
Why did the 10-year Treasury yield fall on October 22, 2026?+
Investors sold technology stocks on worries about AI spending after Meta's earnings and moved money into Treasuries. The 10-year par yield fell 17 basis points to 4.96%, even though oil rose and jobless claims were low.
Was it the biggest drop in the 10-year yield in years?+
Based on Treasury's daily par yields, the 17-basis-point decline was the largest since August 2, 2024 (−19 bp). Since 2022 only six other days saw a decline of 17 bp or more.
How much does a bond fund gain when yields fall 17 basis points?+
Roughly duration × yield change. A fund with a 6-year duration gains about 1.0% in price (6 × 0.17%), and one with an 8-year duration about 1.4%, before interest income.
How did a 60/40 portfolio do on October 22?+
With 60% in an S&P 500 fund (−1.21%) and 40% in a 6-year-duration bond fund (about +1.0%), a $100,000 portfolio lost about $318, or 0.32%, versus $1,210 for an all-S&P 500 portfolio.
Is the Fed expected to change rates on October 28?+
Press reports before the meeting said traders largely expected the Fed to hold its target range at 3.75%–4.00%. A possible hike has been priced for the December 8–9 meeting instead; market odds change daily and are not forecasts.
Sources
- U.S. Treasury: Daily Treasury Par Yield Curve Rates, 2026 (and 2022–2025 files)
- U.S. Department of Labor: Unemployment Insurance Weekly Claims, October 22, 2026
- Federal Reserve: FOMC meeting calendars, 2026
- BEA: Release schedule (Q3 2026 GDP advance estimate, personal income and outlays)
- BLS: Consumer Price Index release schedule
- FRED: S&P 500 (SP500), daily close
- FRED: NASDAQ Composite Index (NASDAQCOM)
- FRED: Consumer Price Index for All Urban Consumers (CPIAUCSL), BLS data
- FRED: Federal funds target range upper limit (DFEDTARU)
- Bernama / Xinhua: US stocks end mixed as AI spending jitters spark tech sell-off (Oct 23, 2026)
- Yahoo Finance: Stock market today, Thursday, October 22 — S&P 500, Nasdaq slide as tech stocks sink
- Investing.com: Why is CoreWeave stock sliding today? (Oct 22, 2026)
This article is for information and education only. It isn't investment or tax advice. Rules and broker practices change — check the current rules or talk to a tax advisor.
See your portfolio next to the S&P 500
Import your Interactive Brokers history or add trades by hand — GetHossa shows your real return, your TWR and your lead over the S&P 500.
Create account