Theme · past 6 monthsUS, Euro area, Japan, UK
Hikes arrive: the ECB and BoJ first, then the Fed
Three of the four central banks hiked and the BoE is next, so the front end did the work and curves bear-flattened.
What it did to the curve
Front end
Yields higher
2-year
Long end
Yields higher
10–30-year
Curve
flatter
long minus short
A bear flattening: between 22 Apr and 16 Oct the US 2-year rose from about 3.80% to 4.715% and the 10-year from about 4.305% to 5.085%, so 2s10s narrowed from about +51bp to about +37bp. German 2s10s narrowed from about +47bp in early April to about +35–42bp.
Fed funds upper bound
4.00%
up+25bpHiked to 3.75–4.00% on 16 Sep, the first hike since July 2023; the median dot of 4.1% implies one more.
ECB deposit rate
2.50%
up+50bpHikes on 11 Jun and 10 Sep, the first since 2023, with staff seeing 3.0% inflation in 2026.
BoJ policy rate
1.25%
up+50bpHikes on 16 Jun and 18 Sep, the shortest gap between hikes under Ueda.
US 2s10s
+37bp
down−14bpFrom about +51bp on 22 Apr: the 2-year rose about 92bp, the 10-year about 78bp.
What happened
In mid-April all four central banks were on hold, waiting to see whether the energy shock from the Iran war would fade. It didn’t, inflation forecasts rose, and the hikes came.
The ECB moved first: its deposit rate went to 2.25% on 11 June, its first hike since September 2023, then to 2.50% on 10 September. The BoJ hiked to 1.0% on 16 June and to 1.25% on 18 September. The BoE is still at 3.75%, but votes to hike grew from one in April to three in July and September.
The Fed took longer, through a change of chair. Powell’s last meeting, on 29 April, split 8–4, the most dissents since 1992. At Warsh’s first, in June, the statement dropped its , its hints about future moves, and three officials dissented for a hike in July. On 16 September the Fed hiked 12–0 to 3.75–4.00%, its first hike since July 2023.
The front end trades the expected , so it moved most: the US 2-year rose about 92bp to 4.715%. The 10-year rose less, about 78bp to 5.085%, because it averages over a longer horizon in which hikes are expected to bring inflation back down.
That is a bear flattening of the : yields rose, short ones fastest. US 2s10s narrowed from about +51bp to about +37bp. German 2s10s went from about +47bp in early April to about +35–42bp.
Next come the FOMC on 27–28 October, the ECB on 29 October and the BoJ on 30 October. The BoE follows on 5 November, where a hike is priced at about 83–91%.
Timeline
· US
Powell’s last meeting as chair: hold at 3.50–3.75%, split 8–4, the most dissents since 1992
· Euro area
ECB hikes its deposit rate to 2.25%, its first hike since September 2023
· Japan
BoJ hikes to 1.0%, with Governor Ueda absent in hospital
· US
Warsh’s first FOMC: unanimous hold, forward guidance dropped, 9 officials see a hike by end-2026
· UK
BoE holds at 3.75% by 6–3, with Greene, Pill and Mann voting to hike
· Euro area
ECB hikes to 2.50%; the Schatz yield jumps as much as 12bp to 3.19%, the highest in almost three years
· US
Fed hikes 25bp to 3.75–4.00% (12–0), its first hike since July 2023
What could change it
Base case
The Fed skips October and hikes in December, which Goldman expects to be its second and final hike; the BoE hikes to 4.00% on 5 Nov, as 54 of 63 economists in a Reuters poll expect; the ECB and BoJ pause in October with December in view. Front ends stay high and curves stay flat.
Watch: FOMC, 27–28 Oct · BoE, 5 Nov · FOMC, 9 Dec
Hawkish risk
Inflation forces more: an October Fed hike (the odds briefly jumped to about 44–54% after September CPI), a BoE that signals a series of hikes, or another ECB hike on 29 Oct. The 2-year reprices higher and 2s10s flattens further.
Watch: UK CPI, 21 Oct · US PCE, 29 Oct · ECB, 29 Oct
Relief rally
Growth gives way first (US payrolls rose just 29,000 in September) or oil falls on a Hormuz deal, and central banks stop after one more hike or none. Front ends rally most and curves steepen.
Watch: ECB, 29 Oct · BoJ, 29–30 Oct · FOMC, 9 Dec
How desks express it
US 2s10s flattener
CurvePay 2y, receive 10y SOFR swaps (DV01-neutral)
While the Fed is still hiking, the 2-year has more to reprice than the 10-year.
- 1
The September median dot of 4.1% implies one more hike by end-2026, and a December hike is priced at about 89%.
- 2
Each hike lifts the 2-year, which tracks the expected policy path, more than the 10-year.
- 3
That is how US 2s10s went from about +51bp to about +37bp over the six months.
Wrong if: Long-end-led sell-offs can steepen the curve even in a hiking cycle: 2s10s went from about +26bp on 23 Sep to about +46bp on 1 Oct as the 10-year rose and the 2-year fell. A dovish FOMC on 27–28 Oct or 9 Dec that takes December off the table would steepen it from the front.
BoE catch-up: UK vs US 2-year
Cross-marketPay 2y SONIA, receive 2y SOFR swaps (DV01-neutral)
The Fed is near the end of a short cycle while the BoE is about to start, so UK front-end rates should rise relative to US ones.
- 1
The Fed has hiked once and the dots imply one more; Williams sees no case for 2027 hikes, and Goldman expects December to be the second and final hike.
- 2
The BoE has held at 3.75%, but hawkish dissents grew to three and 54 of 63 economists expect a hike to 4.00% on 5 Nov.
- 3
The spread widens if the BoE keeps hiking after the Fed stops.
Wrong if: Much of the BoE path is already priced (four 25bp rises by end-2027), and Pill pushed back on pricing of four rises over the next 12 months. A soft UK CPI on 21 Oct, a dovish BoE on 5 Nov, or US inflation forcing more Fed hikes (producer prices rose 6.3% y/y) would narrow the spread.
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