September FOMC Recap: The Fed hikes for the first time since 2023 – What should you do now?

• The Fed raised the Fed Funds Rate (FFR) by 25bps to 3.75% - 4.00%, with a unanimous 12-0 vote, ending five consecutive holds and marking the first hike since July 2023.

• A dot plot was provided at this meeting, with median projections skewing higher compared to its last update in June. As usual, Warsh declined to submit his own dot.

• Inflation metrics remain sticky, a trend we expect to persist given the recent resurgence in crude oil prices. A better-than-expected NFP print combined with a steady low unemployment rate means the Fed is likely to stay hawkish.

• Rates are likely to stay higher-for-longer; we maintain our preference for short-duration fixed-income products alongside cheap and profitable equities (Asia over the US).

The Fed hikes and the committee speak with one voice

The Fed raised the target range for the Fed Funds Rate (FFR) to 3.75% - 4.00%. This ends five consecutive holds and delivers the first increase since 2023. This was a unanimous vote by the committee, with all 12 voting members electing to raise the FFR by 25bps.

This decision has been some time in the making. We have argued since spring that a turn in the interest rate cycle was likely, and heading into this meeting, markets assigned a >90% chance of a 25bps hike following the August inflation reports.

The statement was brief once again, but three changes stand out: First, the committee noted that domestic spending has been resilient. Second, and most important, the July line attributing elevated inflation in part to supply shocks in certain sectors, including energy, was dropped entirely – the statement now says simply that inflation remains elevated. In our view, the Fed is no longer looking through the energy shock to justify holding rates. Third, a new sentence frames the hike as supporting "a timelier return to the Committee's 2% inflation target".

The dot plot returns and shifts decisively higher

The projections now point to one more hike this year and no cuts next year. The median end-2026 rate rose to 4.1% from 3.8% in June, and the 2027 median rose to 4.1% from 3.6% – erasing the easing June had pencilled in. Support is broad: 16 of 18 participants see at least one further hike in 2026, against nine in June, and a large minority sees another in 2027. This is not a committee that thinks Wednesday finished the job.

The economic projections explain why. Growth was revised up, unemployment revised down to 4.1% through 2028, and inflation revised higher for 2026. The risk assessments are the clearest tell: in June, a third of participants saw growth risks to the downside and unemployment risks to the upside; in September, none did. The committee has stopped worrying about the labour market and is more concerned about inflation risks to the upside– which is the whole case for hiking into an energy shock.

The case for the hike

Core inflation looks better on an annual basis, but the momentum is going the wrong way. August core CPI eased to 2.4% Y/Y, its lowest since March 2021, yet the monthly prints have accelerated from flat in June to 0.3% in August – a pace that would annualise near 3.7%. The annual rate is falling only because current months remain below the year-ago readings they replace, and that gap is closing. The improvement in core is real, but it is running out.

Headline CPI has stalled at 3.4%, and the gap with core is energy. Gasoline alone drove a third of August's monthly increase. Pump prices have risen further since the survey was taken, with regular petrol around USD 4.31 per gallon and diesel at a record USD 6.23, so September's report is likely to show more of this, not less.

PCE inflation, the measure the Fed targets, has not moved. Core PCE has sat at 3.3% for three months while core CPI improved – a divergence that matters, because the Fed sets policy against the sticky measure, not the improving one. That stickiness underpins both this hike and the second one the committee now projects.

The labour market has stopped weakening, which removed the last reason to wait. August payrolls rose 162,000, roughly three times expectations, and the two prior months were revised up – July from a loss to a gain. We would not read too much into one print: it sits against a 12-month average of about 31,000, and participation remains well below its January level despite August's uptick. But the Fed does not need a strong labour market to hike – it needs one that is no longer deteriorating, and that is what it now has.

Oil: where the inflation is coming from

The most important development since July happened in the Gulf, not in Washington. Brent has risen from around USD 87/bbl at the July meeting to roughly USD 107/bbl as of the September meeting. The latest leg matters because it hit the workarounds rather than the chokepoint itself. Saudi Arabia suspended its East-West pipeline – a route used specifically to bypass the Strait of Hormuz – after drone attacks; talks with Oman on a temporary shipping corridor were postponed, and OPEC+ froze October output.

The sharper squeeze is in refined products, particularly diesel. Ukrainian strikes have disrupted several major Russian refineries, and Moscow has restricted fuel exports to protect domestic supply. Concurrently, Middle East refineries have been hit, and Asian refiners short of crude have cut runs. The US has little slack to fill the gap. Refinery utilisation has climbed from around 89% in late February before the conflict to 98% now and has held above 95% for months as operators defer maintenance to capture refining margins. These refining margins tell the story: the US diesel crack spread (margin between crude oil and the fuel refined from it) has widened from a pre-war norm of roughly USD 15-30/bbl to above USD 100, while distillate inventories (stocks of diesel and heating oil) sit around 13% below their five-year average.

Diesel is embedded in the cost of moving goods. It powers trucks, rail, shipping and agriculture, so it feeds into a far wider range of prices than gasoline alone. That transmission channel is already visible in the data: transportation and warehousing prices rose 2.3% in the August PPI, with truck freight prices up 2.0%. And because the August CPI was largely a snapshot taken before the sharpest part of this move, we expect the September and October reports to show more of it. The Fed is hiking into a shock whose inflationary effects have not yet fully arrived.

Food prices have not yet reacted, but they typically respond last. Food inflation eased to 2.7% Y/Y in August, and grocery prices rose 2.2% and were flat on the month. However, food production is energy-intensive at every stage: Diesel runs farm equipment and freight. Natural gas, much of it shipped through Hormuz, is a key input for fertiliser. Carriers add fuel surcharges once diesel prices pass set thresholds.

Rate hikes cannot fix a supply shock, but they can limit its spread. Higher interest rates do not create refining capacity. However, if energy and transport costs spread into broader prices, a temporary shock could become lasting inflation. That is the risk the Fed is trying to contain.

Press Conference Highlights – broad financial conditions are not yet restrictive

Fed Chair Kevin Warsh framed the hike as the application of a standard rather than a change of mind. He reiterated the Fed must be confident underlying inflation is moving to target clearly and at sufficient speed – and said plainly that the FOMC had decided this standard had not been met. He also answered the supply-shock objection directly: while the Fed cannot restore safe passage through the strait or affect individual prices, it can quell the pass-through into freight, logistics and the wider basket. That is the argument we have made since June, and it is now the Fed's stated rationale for acting against an energy shock rather than looking through it.

He left the door open to more. Warsh said he would be “hard-pressed to describe broad financial conditions as restrictive”; a Fed that does not consider itself tight at 4% has room to go further. On the rise in long-term yields this year, he cited two drivers: economic strength and competition for capital, as the surge in capital expenditure sends hyperscalers to the markets for funding. That is the same diagnosis in our house view below: the pressure on the long end is not solely about inflation, and a rate decision alone will not relieve it.

As usual, he gave no forward guidance – 'I'm not in the forward guidance business' – and submitted no projection of his own to the dot plot.

Market Reaction

Rates: a modest bear-flattening in the yield curve (short-term yields rise faster than long-term yields). The 2-year yield rose to its highest since 2024 as the market priced a further hike in December and no cuts through 2027, while the 10-year closed at 5.016% (+2bps) – its highest since July 2007. The 30-year was little changed. That is the opposite shape to July, when a hold sent the long end sharply higher and the front end lower. This time the front end absorbed the hawkish signal and the long end held.

Equities were mixed across the board. The Dow Jones Industrial Average fell 1.21%, the S&P 500 lost 0.45%, the Nasdaq Composite was flat, and the Russell 2000 held near flat. All had been higher before the decision and gave up their gains during Warsh's press conference as he stressed that inflation risks were not improving.

Precious metals also closed lower. Gold ended down more than 1%, and silver fell 1.7%. Higher yields raise the opportunity cost of holding non-yielding assets, with the 10-year at 5% and the dollar index (DXY) up 0.5% on the day.

Positioning your portfolio in a higher-for-longer world

The Fed has now delivered the hike, pencilled in another, and removed the cuts it had expected for 2027. Our recommendations below are the same ones we have been making (see Table 1 below).

In bonds, keep duration short. Short-dated bonds mature earlier, so they are less exposed to interest rate swings. And the pressure on long yields is not only about inflation: Warsh himself pointed to competition for capital as a driver, with governments and hyperscalers both borrowing heavily. A single rate decision does not fix that.

In equities, pay for profits that arrive soon, not late. A dollar earned today is worth more than a dollar earned in ten years, and higher rates widen that gap. Cheap, profitable companies earning most of their value in the next few years hold up better than expensive ones betting on distant profits.

That is why we continue to favour Asia over the US. Asian equities trade at a meaningful discount to the S&P 500 for broadly comparable earnings growth. Asian semiconductors remain our highest-conviction idea: Samsung and SK Hynix have strong balance sheets, demand underpinned by multi-year contracts, and a valuation gap to US chipmakers that has yet to close. The Global X Asia Semiconductor ETF (HKEX: 3119) is our preferred vehicle.

In Asia, we favour China and Singapore. In China, the GF CSI All-Share Information Technology ETF (SZSE: 159939) captures AI demand already visible in earnings, and the iShares Hang Seng TECH ETF (HKEX: 3067) trades well below its long-term average as a price war fades. Singapore offers dividends – cash paid today, which is exactly what holds up better when rates rise.

Our one exception to US caution is internet names (large-cap profitable companies). Most hyperscalers have underperformed this year, and their valuations have compressed to below historical averages while earnings held up. That does not make them cheap, but less of their price rests on distant profits. The Invesco NASDAQ Internet ETF (NASDAQ: PNQI) is one way in.

Table 1: Recommended Products