Fed set to hold as hot inflation meets the midterms

What the data reveals ahead of the Fed decision

There have been easier weeks to be a central banker, though few easier ones to predict, with the Federal Reserve widely expected to leave rates at 3.75–4.0% on Wednesday despite an inflation backdrop that would, in a quieter month, have made the case for another move all by itself. Headline CPI accelerated to 3.7% YoY in September, core CPI ticked up to 2.6%, wholesale prices jumped by more than 6% and the Fed’s preferred measure, core PCE, is projected to top 3% – hardly the profile of an inflation problem that has been solved.

Markets briefly took the hint. A hotter-than-expected 0.3% MoM rise in core CPI pushed the odds of an October hike above 50%, before a procession of officials – Vice Chairs Philip Jefferson and Michelle Bowman, Governor Chris Waller and New York Fed president John Williams – signalled that there was no particular hurry, which is central bank for ‘please calm down’. By the week before the meeting, pricing had swung back to roughly an 80% chance of a hold.

Chair Kevin Warsh has done little to disturb that view. Speaking at the IMF meetings in Bangkok he noted only that the world had broadly conformed to the Fed’s expectations since September, a notably less excitable performance than his Jackson Hole appearance ahead of the September hike, which was the Fed’s first increase in more than three years.

The decision itself may be the least interesting part of the evening.

Hawkish Case: Why December looks likely

The case for the Fed staying firmly in tightening mode is not hard to make, and several of its own officials have been making it. Minutes from the September meeting showed most participants judged that another increase would likely be appropriate by year-end, while the Summary of Economic Projections had four members pencilling in a further 50bps of hikes before the end of 2026 and nearly all participants expecting at least one more.

That leaves plenty of room for dissent. Beth Hammack, Lorie Logan and Neel Kashkari all voted to hike in July when the rest of the committee preferred to wait, and the trio could well do so again, which would rather undermine any attempt to present a hold as a consensus view.

Then there’s the election. The midterms fall less than a week after the decision, and former IMF chief economist Ken Rogoff argues that the fundamentals point to a hike that only the calendar prevents, while former Fed vice chair Alan Blinder expects a ‘hawkish pause’ and sees the proximity of a vote as one more reason, though far from the only one, to keep heads down. If the election is the main thing holding the Fed back, December could arrive with the excuse removed and the inflation data still uncomfortable. The consensus forecast is a 25bps hike in December that would lift the range to 4–4.25%.

Financial conditions could decide the debate

Warsh appears well aware that the Fed does not operate in a vacuum, and his description of the September move as removing a dose of accommodation looked very much like a nod to the growth boost from easy financial conditions – chiefly the wealth effect from strong equity markets, which has encouraged households to spend freely on the back of their portfolios.

The bond market has since done some of the Fed’s work for it. The 5-year, 5-year-forward real rate has risen by about 50bps since Warsh took over in June to roughly 3%, a level last seen in the early 2000s, which should act as a growing drag on growth over the coming quarters. The 10-year Treasury yield, meanwhile, has slipped back below 5% to 4.986% after an earlier break above 5.25%, helped along by a tech sell-off, wider credit spreads and a battering in the French bond market.

There is a catch. Warsh was criticised after the July meeting for implying that the bond market was doing the tightening, and he may well avoid any suggestion that softer equities and wider spreads mean fewer hikes are needed. A Fed chair who learns from his own press conferences is, at least, a novelty. A plan to shrink the balance sheet, a stated Warsh objective, is likely still some way off, with committee recommendations expected closer to the turn of the year and repo markets behaving far better than they did twelve months ago.

​Dovish Case – why the Fed may not go much further

​None of this means the Fed is embarking on a lengthy hiking campaign, and the outlook looks more like a recalibration – one or two more hikes that would effectively reverse last year’s cuts – than a cycle. AI sits at the heart of that argument, since it is propping up demand through investment and the wealth effect while making the usual policy arithmetic considerably harder to do.

​Much of the AI buildout relies on imported chips and computer equipment, which limits its impact on domestic production and so reduces the rate rise needed to cool it, although the rising cost of those imports is now showing up in US inflation. The labour market also looks looser than the unemployment rate suggests, with retirements, falling participation and AI-related cuts to higher-skilled white-collar roles tilting employment towards lower-paid jobs and keeping wage growth at a benign 3%.

​The more awkward point is that AI capital, dominated by fast-depreciating computer equipment, is relatively insensitive to interest rates, so the burden of tighter policy falls on the parts of the economy that are not driving inflation. Investment outside AI is already contracting and the housing market is feeling the pain. Should AI returns disappoint – and transformative technologies have a habit of doing so in the short run – the wealth effect could reverse quickly and tighten conditions without any help from the Fed.

​The consensus path reflects that caution, with a December hike potentially followed by a prolonged pause and a reversal in December 2027. That view leans heavily on a post-midterm agreement that improves energy flows from the Persian Gulf and, crucially, brings down shipping and insurance costs, which alongside disinflation in housing – 35% of the inflation basket – could bring inflation back to 2% in the second half of 2027. A longer conflict could mean further hikes in the first quarter of 2027, followed by swifter and deeper cuts later in the year as households and borrowers feel the squeeze.

​Historical Parallels and what they tell us

​The Fed has form on both sides of the election question. In November 2018 it held rates two days after the midterms before hiking again in December, a move that coincided with a sharp fourth-quarter sell-off in equities and a swift change of tone in early 2019. In November 2022, by contrast, it raised rates by 75bps just six days before the midterms, which suggests that when inflation is running hot enough the calendar is a consideration rather than a constraint.

​Warsh’s own short record may matter more to traders. Every FOMC meeting he has chaired has set the dollar’s trend for the following four to six weeks, with June’s meeting hawkish, July’s taken as dovish and September delivering a hawkish hike. The dollar rose about 4% through September and into early October before settling into a more consolidative phase, with rate differentials moving slightly against it as the tech sell-off pared back tightening expectations.

​December could signal a turning point

​With no new projections at this meeting, the signals are likely to come from tone and votes rather than the decision itself:

• ​the number of dissents in favour of an immediate hike, with three or more suggesting December is closer to a formality
• ​whether Warsh and the committee continue to describe September’s move as a risk management adjustment
• ​any suggestion that tighter financial conditions reduce the need for further hikes, which Warsh may be keen to avoid after July
• ​the inflation print due the day after the meeting, alongside two jobs reports before December
• ​progress, or the lack of it, on energy flows and shipping costs through the Persian Gulf

​If the Fed holds with dissents and a firm tone, Treasury yields may push back above 5% and the dollar could hold on to its September gains, potentially dragging EUR/USD back towards the year’s lows near 1.11. If Warsh instead sounds relaxed and the next round of data cools, markets may begin to question whether December happens at all – and with the bond market already doing some of the heavy lifting, would anyone at the Fed be in a hurry to argue?