Theme · past 4 weeksUS

Fed: probably not October, probably December

The Fed’s next hike has slipped from October to December: the front end has priced out the timing, not the hike.

What it did to the curve

Front end

Yields lower

2-year

Long end

Yields little changed

10–30-year

Curve

steeper

long minus short

The 2-year fell about 11–14bp, from 4.85% on 23 Sep to about 4.72–4.74% on 16 Oct, as October hike odds faded, while the 10-year ended near where it started (5.11% to about 5.09%). So 2s10s steepened from about +26bp to about +35–37bp.

October hike odds

about 15%

from about 70%

CME FedWatch on 16 Oct after Warsh, against about 70% after Barr on 23 Sep.

December hike odds

about 89%

from about 83%

FedWatch on 16 Oct against a week earlier: the hike was pushed back, not priced out.

US 2-year

about 4.72%

from 4.85%

16 Oct (CNBC about 4.715%, Treasury par 4.74%) against 4.85% on 23 Sep: the tenor that prices the policy path.

US 2s10s

about +35–37bp

from about +26bp

Derived, 23 Sep to 16 Oct: a steepening driven by the front end, not the long end.

What happened

In late September the question was when the Fed would hike again, not whether. After Governor Michael Barr said on 23 Sep that more tightening would probably be needed, CME FedWatch, which reads odds from , put about 70% on a move at the 27–28 Oct meeting. By 16 Oct that had fallen to about 15%, while a December hike was about 89% priced.

Each release moved the timing, not the destination. Weak payrolls of just +29,000 on 2 Oct cut October odds to about 17–23%. Waller said hikes need not come at consecutive meetings, and Williams said one more 2026 hike may be appropriate, so skipping October would not end the cycle.

September came in hot, at 3.7% with core at 2.6%. That briefly lifted October odds to 44–54%, but Chair Kevin Warsh declined to signal the decision on 16 Oct. Underneath, demand looked softer: retail sales rose 0.8%, but the control group, which strips out volatile items such as petrol, rose only 0.1%.

The 2-year prices the expected policy path, so it fell about 11–14bp as October odds faded. It went from 4.85% to about 4.72%. The 10-year carries inflation, supply and , so it rose to 24-year highs and came back, ending near where it began (5.11% to 5.085%). With the front end down and the long end flat, 2s10s, the gap between them, steepened from about +26bp to about +35–37bp.

Desks trade the timing directly with meeting-dated OIS, which separates October from December, and the curve with a 2s10s . The next tests are the FOMC on 27–28 Oct and September on 29 Oct.

Timeline

23 Sep 2026 · US
Barr says more tightening is probably needed; October hike odds jump to about 70%
1 Oct 2026 · US
Goldman moves its next-hike call to December; FedWatch shows about 38% for October
2 Oct 2026 · US
Payrolls rise just 29,000; October hike odds drop to about 17–23%
8 Oct 2026 · US
Waller: more hikes needed, not necessarily at consecutive meetings; October odds about 20%
14 Oct 2026 · US
September CPI 3.7%, core 2.6%: October odds briefly jump to 44–54%
16 Oct 2026 · US
Warsh gives no signal on 28 Oct; October odds about 15%, December about 89%

What could change it

Base case

The Fed holds on 28 Oct and hikes on 9 Dec, as pricing implies (about 15% for October, about 89% for December). Waller’s point that hikes need not come at consecutive meetings is the template.

Watch: FOMC, 27–28 Oct · US PCE, 29 Oct · FOMC, 9 Dec

Hawkish risk

October comes back. Musalem said he was open to an October move, PPI ran at 6.3% y/y and Brent settled at $109.86 on 16 Oct. A hike on 28 Oct would reprice the front end and flatten 2s10s.

Watch: FOMC, 27–28 Oct · US PCE, 29 Oct

Relief rally

December fades too. Goldman saw a strong chance the FOMC ultimately decides no more hikes are needed, and the retail control group rose only 0.1%. A soft PCE print would pull the 2-year lower and steepen the curve further.

Watch: US PCE, 29 Oct · FOMC, 9 Dec

How desks express it

Fed Oct–Dec meeting-dated OIS spread

Spread

Receive Oct, pay Dec meeting-dated OIS

The hike is delayed, not cancelled: own the gap between a hold in October and a hike by December.

1
Warsh gave no signal for 28 Oct and October odds fell to about 15%, so the receive leg gains if the hold is confirmed.
2
Officials still see one more move: Williams said one more 2026 hike may be appropriate, and December is about 89% priced, so the pay leg gains as that firms up.
3
Most of the repricing has already happened, so this is a trade on confirming the timing, not on a big swing in odds.

Wrong if: A surprise hike on 28 Oct hurts the receive leg. A dovish turn before 9 Dec, the outcome Goldman flagged when it saw a strong chance of no more hikes, hurts the pay leg.

US 2s10s steepener

Curve

Receive 2y, pay 10y SOFR swaps (DV01-neutral)

Fading near-term hikes pull the front end down faster than the long end, which still carries inflation and supply risk.

1
The 2-year prices the policy path: as October odds fell, it dropped from 4.85% to about 4.72%.
2
The 10-year is held up by , with PPI at 6.3% y/y and Brent near $110, so it ended near 5.085%.
3
Front end down, long end flat: 2s10s went from about +26bp to about +35–37bp, and a DV01-neutral steepener gains from the gap, not the level.

Wrong if: A long-end-led rally flattens the curve, as in the week of 12–16 Oct, when the 30-year fell 17bp and the 10-year 15bp against 6bp for the 2-year. A hawkish FOMC on 28 Oct would also lift the front end.

Educational expressions of the theme — how a desk might position, not advice.

Test yourself

Finishing the quiz adds this theme’s key facts to your review deck.

The stories behind it

Sources