US Inflation Test & Fed Path

The September CPI report lands on Wednesday, and the headline number is already largely known.

Record gasoline prices make a jump in headline inflation close to certain. The Cleveland Fed nowcast points to a 0.53% monthly rise and 3.6% year on year, up from 3.4% in August. The street consensus sits at 3.7%.

That is not the number the Fed will be looking at.

The real test is core inflation. A core print of 0.2% would confirm that the energy shock is still contained. A reading of 0.3% or higher would suggest fuel costs are leaking into services, and that would change the rate path.

The Federal Reserve raised its target range to 3.75%-4.00% on 16 September, the first hike since 2023. The minutes released on 7 October showed most officials expect one more move by year-end. Futures currently price roughly a one-in-five chance of a hike on 28 October and about 70% for December.

Wednesday’s report will decide which of those two numbers moves.

September CPI: what to expect

The headline story is gasoline.

Pump prices averaged $4.33 a gallon in September, according to AAA. That is a record for the month and 50 cents above the previous September high set in 2023. Continuum Economics expects energy prices to rise about 5.5% on the month, the largest contribution since March.

The base-effect arithmetic is straightforward. Headline CPI rose 0.3% in September 2025, so a 0.5% monthly print would lift the annual rate to roughly 3.6%, and 0.6% would take it to 3.7%.

One detail is worth noting. Both the Cleveland Fed nowcast and Kalshi pricing (3.63%) sit slightly below the 3.7% consensus. In other words, the market’s real expectation is a little softer than the published forecast, and a 3.7% print would probably land as a mild upside surprise.

Core is expected at 0.2% month on month and 2.4% year on year.

There are good reasons to expect restraint. Rent and owners’ equivalent rent each rose only 0.2% in August. Used-car prices look due for a pullback after a 0.4% rise. And August’s 5.4% jump in telephone services looks like a one-off.

The risks sit in transport. Airfares rose 2.7% in August and are up 23.4% on the year, which is fuel cost passing straight through to ticket prices. Diesel feeds into delivery and freight charges with a lag. If supercore picks up, that is where it will show first.

Rounding matters more than usual this time. The BLS reports one decimal place, so anything from 0.25% to 0.34% prints as 0.3%. Desks will calculate the unrounded figure within seconds, and a 0.26% reads very differently from a 0.34% for December.

Why a soft core will not calm the Fed

Core CPI at 2.4% is the lowest since March 2021. On its own, that looks almost comfortable.

The problem is that the Fed does not target CPI.

Core PCE, the measure the 2% goal is set in, rose 3.0% in August. CPI normally runs above PCE, so a gap of 0.6 points in the other direction is unusual. In my view the explanation is mostly about weights. Shelter makes up roughly a third of the CPI basket and is cooling at 3.0%. PCE gives it far less weight and gives more to categories that are running hot, such as airfares and financial-services fees, which rise with a record stock market.

Strip out the shelter effect and the underlying picture looks firmer than the CPI headline suggests.

The Fed’s own projections make the point even clearer. The September median sees core PCE at 3.4% in the fourth quarter. The committee hiked in September while expecting core inflation to accelerate into year-end. An in-line Wednesday does not challenge that view. It confirms it.

Momentum is the other issue. Core rose 0.2% in July and 0.3% in August. A 0.2% September would leave the three-month annualised pace around 2.8%. A 0.3% would lift it to about 3.2%, and 0.4% to roughly 3.7%.

There is no plausible print on Wednesday that brings underlying inflation back to target.

Households are drawing their own conclusions. The preliminary Michigan survey for October put year-ahead inflation expectations at 4.7% and long-run expectations at 3.5%, both the highest since May. Sentiment fell to 46.3, close to May’s record low, with the sharpest deterioration among lower-income households. Market breakevens, by contrast, are anchored near 2.4%.

That gap between households and bond traders is exactly what Governor Waller flagged on 8 October, when he warned that more than five years of above-target inflation are putting expectations at risk.

The Fed: pause in October, hike in December?

The committee is not divided on direction. It is divided on timing.

Nobody at the Fed is talking about cuts. Governor Waller said on 8 October that more hikes are needed but do not have to come at consecutive meetings. St. Louis Fed President Musalem said further firming will be required. Dallas Fed President Logan has argued for at least another 50 basis points. On the other side, New York Fed President Williams and Vice Chair Jefferson see little urgency, and Vice Chair for Supervision Bowman would prefer no more hikes this year.

The labour market barely features in this debate.

September payrolls rose by just 29,000 and unemployment climbed to 4.2%. Yet both Waller and Musalem described the job market as stable. Under Kevin Warsh, the Fed is effectively operating with one priority, and that is inflation. Traders who still read weak data as a reason to expect easing are working from an outdated playbook. A soft jobs report now buys a pause, not a pivot.

The bond market is also doing part of the Fed’s work.

The 10-year Treasury touched 5.36% on 7 October, and the 10-year TIPS yield reached 2.91%. With the funds rate at a 3.875% midpoint and core PCE at 3.0%, the real policy rate is only around 0.9%. Long-term real yields are roughly two points higher. Logan made the distinction herself: yields pushed up by a rising term premium tighten financial conditions in a way that can substitute for hikes.

This cuts both ways. If a soft CPI triggers a strong bond rally, conditions ease and the Fed loses its silent partner, which would make a December move more likely, not less.

Politics narrows the October window further. The decision comes six days before the 3 November midterm elections. Hiking on the eve of a vote, without a clear breakout in core inflation, would expose a first-year chair to accusations of political timing whichever way it was read.

My base case is a hawkish hold on 28 October. I expect Warsh to use the statement and the press conference to prepare the ground for December rather than act immediately. Only a core print of 0.4% or higher would make an October hike the more likely outcome.

The market has already paid for much of this path. The 2-year Treasury yield at 4.79% is almost 80 basis points above the upper bound of the target range, and futures imply at least three more hikes by the end of 2027.

U.S. dollar and Treasuries

The front end of the curve is the cleanest way to read the CPI outcome.

A core surprise to the upside should flatten the curve from the short end, with the 2-year rising faster than the 10-year as the next hike is pulled forward. An energy-only print works differently. Breakevens edge higher, the long end cheapens on inflation compensation, and the 2s10s spread, around 48 basis points in early October, widens slightly.

The long end has its own drivers. Oil, fiscal supply and term premium all matter there, which is why the 10-year often reacts less cleanly to CPI than the 2-year. Strong demand at recent Treasury auctions has helped cap yields so far. Whether that demand holds is the main risk for duration.

The dollar is already close to its highs.

The DXY rose to about 102.3 after the September minutes, near its strongest level since April 2025. A hot core print would support it further, but the reaction will not be symmetrical across pairs.

USD/JPY is the clearest example. Higher U.S. yields push the pair towards the 160 area, where Japan intervened in 2024. That makes upside in USD/JPY more limited than downside in EUR/USD, where there is no comparable line in the sand.

In the relief scenario, with core at 0.1% or lower, I would expect the dollar to give back part of its recent gains as December hike odds fall towards 50%.

(EURUSD Weekly Chart, Source: Ultima Markets MT5)

On the weekly chart, EURUSD is testing an extremely strong support zone, although the lower Bollinger Band has already been breached, indicating that the pair is heavily oversold.

• A drop below 1.1180 will push EURUSD further down to 1.0950;
• A rebound will bring the price back up to 1.1400.

Gold: what to expect?

Gold is caught between two opposing forces.

The first is real yields. The 10-year TIPS yield reached 2.91% in early October, while the 10-year breakeven sat near 2.36%. Almost all of the rise in long-term Treasury yields has come from the real component. That raises the opportunity cost of holding a metal that pays no interest, and it usually comes with a firmer dollar.

The result was visible at the end of September, when spot gold dropped to an eight-week low of $4,110.

The second force is the macro mix. Real wages are falling, with average hourly earnings up 3.0% against CPI heading towards 3.7%. Consumer sentiment is near record lows. Energy prices are high because of supply disruption, not strong demand. That combination looks increasingly stagflationary, and stagflation has historically been a supportive environment for gold.

For now, the first force is winning.

The CPI report matters for gold mainly through the front end and real yields. A core print of 0.3% or higher would push the 2-year up, support the dollar and keep pressure on the metal. A soft core would do the opposite: lower real yields, a weaker dollar and room for a rebound.

The more interesting scenario is further out. If the energy shock starts to damage growth more visibly, markets could stop pricing further tightening while inflation expectations remain high. Falling real yields with sticky inflation would remove gold’s main headwind, and that is when I would expect the metal to regain momentum.

(XAUUSD Daily Chart, Source: Ultima Markets MT5)

Gold is testing a downtrend line, while the DeMarker has moved out of the oversold zone, indicating the possibility of a breakout and a rise toward the global trend line.

• If the trend line is broken above 4200, XAUUSD will surge to 4400;
• A pullback could send gold down to 4000.

Oil: main outlook

Oil is less a reaction to the CPI report than its cause.

Brent traded between $102.50 and $105 on 9 October and has spent most of the past week above $100 as the Strait of Hormuz stalemate drags on. WTI settled at $91.49 on 8 October. On 7 October, prices turned lower only after the International Energy Agency agreed to speed up the release of emergency stocks, with diesel as the priority.

That choice of diesel is telling. The pressure on inflation comes increasingly from refined products, not crude alone. In August, fuel oil prices were up 52% year on year and airfares 23.4%. Freight and delivery costs follow with a lag.

The important detail for the next CPI report is that pump prices have not come down. AAA data put the national average at $4.41 on 1 October and $4.37 on 7 October, still above the September average and well above $3.16 a year earlier.

This means October inflation will also carry heavy energy pressure.

Headline CPI is unlikely to roll over before the December meeting unless oil falls sharply. Rate hikes cannot solve that problem. They can slow demand over time, but they cannot reopen shipping lanes or add refinery capacity.

The real bearish catalyst for oil is physical normalisation, not monetary tightening. A credible deal on Hormuz traffic could pull Brent down quickly, and that would ease pressure on inflation expectations at the same time. In my view, a ceasefire headline could matter as much for the Fed’s December decision as the CPI print itself.

(UKOUSD Daily Chart, Source: Ultima Markets MT5)

UKOUSD has formed an ascending triangle pattern on the daily chart. However, %R has exited the overbought zone and begun to fall, indicating a possible correction.

• Consolidation above 110.00 will open the way to 119.00;
• A pullback and a break below the lower trendline will trigger a decline toward 98.00.

Conclusion

Wednesday’s report is an exam in two parts, and markets will grade only one of them.

Headline inflation will rise because of gasoline, and the Fed will not hike on gasoline alone. Core is what counts. Even a clean 0.2% leaves underlying inflation running near 2.8% annualised, with core PCE at 3.0% and household expectations moving higher.

My base case is an energy-only print, a hawkish hold on 28 October and a 25 basis point hike to 4.00%-4.25% on 9 December. Two things could change that call. A core reading of 0.3% or more would bring the hike forward and flatten the curve. A genuine breakthrough on Hormuz would pull oil and inflation expectations lower together and put December in doubt.

The variables worth watching differ by asset: the 2-year yield and USD/JPY for the dollar; real yields for gold; refined products and Gulf shipping for oil; earnings relative to a 5% Treasury yield for equities.

The broader message is that under Warsh, weak growth buys time but not cuts. Until core inflation and long-run expectations turn lower together, the path of least resistance for U.S. rates is still higher.