The 10-Year at 5.3%, Two Weeks After Quarterly Expiration — 26 Years of Yields and Stocks

Early on October 1, 2026, the US 10-year Treasury yield touched 5.33%, its highest level since 2002. It started the year at 4.19%, so it has risen by more than a full percentage point in nine months.

Two weeks earlier, Friday, September 18, was quarterly expiration: the third Friday of March, June, September and December, when index options and futures expire together (often called triple witching). The 10-year stood at 5.01% that day. On September 30, the last day of the quarter, it closed at 5.29%.

Those two weeks prompted me to check four things:

Where the 10-year is now, and why it rose
How yields and stocks have moved together historically
How institutions read 5%, and whether the banks can handle it
Whether quarterly expiration and the 10-year are actually related

1. Where the 10-year stands

Here is this year, using the daily Treasury yields published by the US Treasury.

Date	2-year	10-year	30-year
Jan 2	3.47%	4.19%	4.86%
Feb 27 (year's low)	3.38%	3.97%	4.64%
Jun 30	4.14%	4.44%	4.91%
Aug 31	4.34%	4.75%	5.25%
Sep 30	4.88%	5.29%	5.64%

In September alone, the 10-year rose 54 basis points, and so did the 2-year. The 2-year is the Treasury maturity most closely tied to expectations for Fed policy. When short and long yields rise together, it suggests this move came largely from expectations of further Fed hikes, not just longer-term worries.

The Fed did raise its policy rate by 25 basis points to 3.75–4.00% on September 16. The statement said inflation remains elevated. Bloomberg's October 1 report pointed to persistent inflation, heavy government borrowing, strong growth, oil prices and developments in the Middle East, and capital demand from AI investment.

2. Yields and stocks: the relationship has flipped more than once

You often hear that rising yields mean falling stocks. The data says that hasn't always been true.

What the research shows

1970s–1990s: on days yields rose, stocks tended to fall. Inflation was the dominant risk, and inflation shocks pushed bonds and stocks down together.
2000–2021: the opposite. Yields and stocks often rose together. Inflation was stable, and yields were driven mainly by growth expectations, and good growth news lifts both.
Since 2022: back toward the pre-2000 pattern, triggered by the highest inflation in 40 years and rapid Fed hikes.

That is the shared picture from Campbell, Pflueger and Viceira (2020), Ilmanen (2003) and AQR's research team (2023). The key point is what is driving yields, not their level. When inflation drives yields, bond prices and stocks tend to fall together. When growth drives yields, yields and stocks tend to rise together.

What I measured: 26 years

I calculated, for each year since 2000, the correlation between daily S&P 500 returns and daily changes in the 10-year Treasury yield, on a scale from −1 (opposite) to +1 (same direction).

Note: this uses yield changes, so the sign is the reverse of the usual stock–bond correlation, which uses bond returns.

Period	On days yields rose, stocks…	Correlation
2000–2021 (22 years)	mostly rose too	0 or above in 21 of 22 years (2006: −0.12)
2022–2024	moved slightly opposite	−0.18 · −0.07 · −0.07
2025	slightly same direction again	+0.13
2026 (through Sep 30)	clearly opposite	−0.44, the most negative since 2000
September 2026 only		−0.70

This year, the "yields up, stocks down" relationship is the strongest of the past 26 years. That fits what the research describes as an inflation-driven regime.

There is one twist. Since the start of the year, both have gone up. The 10-year went from 4.19% to 5.29%, and the S&P 500 went from 6,845 to 7,652 (+11.8%). Day to day they moved in opposite directions, but over nine months stocks have absorbed the higher yields. A common explanation in the market is that AI earnings are supporting valuations.

When yields did hit stocks

Episode	10-year	S&P 500
1994 "Great Bond Massacre"	about 5.2% → 8.0%	about −10% from the high
2013 taper tantrum	up more than 1 point in about two months	about −6% (still up about 30% for the year)
Q4 2018	peaked at 3.24%	−19.8% from Sep 20 to Dec 24
2022	about 1.5% → 3.9%	−19.4% for the year
October 2023	briefly above 5% (first time since 2007)	−10.3% from the July high

The recurring theme is speed. In 2018, Goldman Sachs noted that the speed of yield changes often matters more for stocks than the level. In months when yields rose by more than two standard deviations (roughly 40 basis points or more at the time), S&P 500 returns were typically negative. September's 54-basis-point rise was faster than that. Even so, the S&P 500 fell only 0.5% for the month.

3. How institutions read 5%

The forecasts: already exceeded

Few institutional forecasts I could find carried a clear date. I have only listed the ones I could confirm.

JPMorgan (July 31 note): raised its year-end 10-year forecast from 4.70% to 4.85%, citing higher inflation expectations and a rising term premium. The yield is already more than 40 basis points above that. (Confirmed only through summary articles.)
PIMCO (September 22, Bloomberg): trimmed its underweight on long Treasuries once yields topped 5%. I read that as PIMCO seeing more value above 5%.
Fed July minutes (primary source): the equity premium, adjusted for the level of long-term rates, was at a level "that has only been lower in recent history during the dot-com bubble." In other words, stocks look expensive relative to bonds.

The "5%" line

RBC Capital Markets (Lori Calvasina): a base case of 7,900 on the S&P 500 with the 10-year at 4.5%. A stress case of 7,400 if inflation runs at 3.8%, the 10-year hits 5% and the Fed hikes, and 6,300 if earnings also fall another 5%. Two of the stress case's three conditions, the 10-year at 5% and a Fed hike, have now been met. I have not checked the inflation condition. (I also could not confirm the publication date.)
According to late-September coverage, JPMorgan Private Bank sees 5% and Société Générale sees 5.5% as the level where stocks come under pressure. (Confirmed only through secondary coverage.)

💡 The one dated year-end forecast I could confirm has already been passed. The question now is less "how high can it go" than "how long can stocks live with yields in the 5s."

4. Can the banks handle it? A comparison with 2023

Silicon Valley Bank failed in March 2023 for a simple reason. Bonds it bought when rates were low lost value as rates rose, As deposits flowed out, it had to sell them at a loss, and that sale set off a run. So with yields at their highest since 2002, a natural question is: how large are unrealized losses on bank bond portfolios?

What the FDIC reports

	End of 2022 (just before SVB)	End of Q2 2026
Unrealized losses on bank securities	$620.4 billion	$326.7 billion (5.5% of cost)
├ Held to maturity (HTM)	$340.9B	$216.9B
└ Available for sale (AFS)	$279.5B	$109.8B

As of Q2 2026, unrealized losses were about half the end-2022 level. The Fed's May Financial Stability Report also noted that banks have shortened the maturity of their securities and that capital ratios are high by historical standards.

But the numbers lag

The $326.7 billion figure is as of June 30, when the 10-year was 4.44%. It is about 90 basis points higher now, so the Q3 figure is likely to be larger. The FDIC's Q3 report is due around late November.
In the same report, all of the quarter's deposit growth came from uninsured deposits (balances above the insurance limit). Those were the deposits that left SVB all at once.
Past-due and nonaccrual rates on non-owner-occupied commercial real estate at large banks stood at 3.08%, about five times the pre-pandemic average of 0.59%.
Five US banks have failed so far this year, up from two in each of 2024 and 2025. The largest was Nano Banc in California, closed September 25, with $736 million in assets. All were small banks. I found nothing on the scale of SVB.

💡 These numbers do not say a bank run is coming. I am noting two things, though: the "better than 2023" figure dates from when the 10-year was 4.44%, and uninsured deposits are growing. I'll check again when the Q3 report comes out in late November.

Quarter-end funding was calm

At quarter-end, banks often pull back on short-term lending ahead of their reporting date, and repo rates (short-term secured lending rates) can spike. A Fed study found that at the September and December 2024 quarter-ends, the spread between SOFR and the Fed's overnight reverse repo rate widened to as much as 25 basis points.

This September 30 was calm, according to New York Fed data.

SOFR, the benchmark overnight repo rate, was 3.90%, up 2 basis points from the prior day and below the top of the policy range (4.00%)
Use of the Fed's Standing Repo Facility (SRF) was $1.2 billion, a small amount

There was no quarter-end funding squeeze this time. The rise in yields reads as coming from inflation, the Fed and Treasury supply (section 1), not from money markets. The next date to watch is the Treasury's quarterly refunding announcement on November 4, which sets how much long-term debt will be issued.

5. Quarterly expiration and the 10-year: is there a link?

In the eight trading days from this September's quarterly expiration (September 18) to quarter-end, the 10-year rose 28 basis points. So do yields tend to move more after quarterly expiration?

In what I searched, I found no academic or central-bank research on this question, so I checked it myself.

Period: March 2000 – September 2026 · 106 quarterly expirations · daily 10-year yields from the US Treasury Three comparisons plus a direction check, set in advance — four tests in all (so I couldn't pick them after seeing the results):

Comparison	Quarterly expiration	Otherwise	Result
① Size of the yield change, expiration → 10 trading days later	average 13.1 bp	average 13.4 bp	No difference
② Size of the yield change on expiration day	4.1 bp	other Fridays 4.7 bp	Slightly smaller; no difference once four comparisons are accounted for
③ Size of the yield change in expiration week	9.7 bp	other weeks 9.4 bp	No difference

I also checked direction separately. Yields rose in the 10 trading days after expiration 51% of the time, which is a coin flip.

Across 106 quarterly expirations in 26 years, I found no pattern of unusually large yield moves. Quarterly expiration is an equity-index options and futures event. It runs on a different calendar from the bond market's own schedule (Treasury auctions, refunding announcements, Fed meetings), and the result is consistent with that.

Measured over the same eight trading days, this September's 28-basis-point rise tied for 4th-largest of 107 (matching September 2023). I read it as coming from the Fed's hike just before expiration (September 16) and a run of strong inflation and growth data, not from expiration itself. I did not test that cause.

💡 If you suspected that yields move more after quarterly expiration, this check says they haven't. I read the timing this September as a coincidence of two calendars.

What I'm watching now

What	When	Why
How strongly yields and stocks move in opposite directions on the same day	daily	−0.44 this year, −0.70 in September. If this weakens, it may signal that the inflation-driven phase is fading
The speed of yield increases	monthly	+54 bp in September. Historically, stocks have been hit by speed more than level
Treasury quarterly refunding	Nov 4	Supply of long-term Treasuries
FDIC Q3 banking report	late November	How much bank unrealized losses grew with yields in the 5s

What I don't know yet

I could not confirm most institutional forecasts in dated primary sources. I have listed only what I could confirm and flagged what came from secondary coverage.
I could not find an official, industry-wide estimate of how much bank unrealized losses rise per 1-point increase in yields.
I don't know how long this year's "yields up, stocks down" relationship will last. It persisted weakly in 2022–2024 and then flipped again in 2025.

This post is a record of public data and my own calculations. It is not a forecast of rates, stocks or banks, and it is not a recommendation to buy or sell any security. Investment decisions and their consequences are the reader's own responsibility.

Data: US Treasury daily yield curve (2000–2026) · S&P 500 daily closes · FDIC Quarterly Banking Profile · Federal Reserve FOMC statements, minutes and Financial Stability Report · New York Fed SOFR and SRF data.