The Fed Raised Rates Again. Why Gold’s Inflation Story Has Become More Complicated

The Federal Reserve raised its target range a quarter point to 3.75%–4.00% on 16 September 2026, effective the next day, on a unanimous vote. Diesel, tariffs and tighter policy now pull gold in different directions: our Tactical Macro Conditions Score read 38/100, Cautious, on 16 September, while the Long-Term Structural Score sat at 66/100, Bullish. The two horizons are answering different questions, and only one of them changed this month.

Published September 17, 2026 · Updated October 5, 2026 · 12 min read

Key Takeaways

• The Federal Reserve raised its target range for the federal funds rate by a quarter point to 3.75%–4.00% on 16 September 2026, effective 17 September. The vote was unanimous, 12–0, and the statement said inflation “remains elevated”.
• Our Tactical Macro Conditions Score read 38/100 — Cautious, direction Deteriorating Rapidly — for 16 September. Core levels contributed 19 of 50 points, fiscal and Treasury stress 8 of 10, the confirmation basket 5 of 10 and momentum 6 of 30.
• That is 17 points below the 55 reconstructed for 24 August. But the score moved only one point on its latest update, and on the same reconstructed basis it sat between 35 and 40 through July — 38 is a return to the pre-rally range, not a new extreme.
• Diesel and tariffs push the price level up; the policy response pushes real yields up. The 10-year real yield earned 2 of its 20 points. Federal Reserve Board staff put the core PCE price-level effect of tariffs implemented between February and November 2025 at 0.8% through February 2026, and judged pass-through from that cohort effectively complete — a level effect, not a standing rate.
• The Long-Term Structural Score reads 66/100, Bullish, dated 14 September. Central-bank demand belongs there: the People’s Bank of China added 20.2 tonnes in August, a 22nd consecutive month, but that component already sits at its 10-of-10 ceiling and cannot score higher.

The Federal Reserve raised its target range for the federal funds rate by a quarter point to 3.75%–4.00% on 16 September 2026, effective the following day. The vote was unanimous. “Inflation remains elevated,” the statement said; “today’s policy action will support a timelier return to the Committee’s 2 percent goal.” For gold investors the difficult question is what happens when borrowing costs rise while fuel and imported goods stay expensive. In June the rate side won outright — gold fell below US$4,000 even as oil retreated, because the policy outlook had detached from the energy story. This time the two are moving together. All rates in this article are US dollar policy rates; all score readings are dated snapshots of our own dashboard rather than live market quotations.

38 / 100 Tactical Macro Conditions Score, 16 September 2026 — Cautious, direction Deteriorating Rapidly. Down one point on the previous daily update, and 17 points below the 55 reconstructed for 24 August. The direction indicator’s strongest listed drivers were Fed repricing on the 2-year yield, real yields and gold momentum.

That describes a worsening near-term environment. It does not settle whether gold deserves a place in a long-term portfolio — a question our two-horizon framework keeps deliberately separate, and one the same dashboard answers differently on its structural view. The full component breakdown is on our Macro Conditions Score dashboard.

A reversal — measured from a peak

In our August review of the rebuilt tactical engine the reconstructed V2 reading was 52 on 17 August and 55 on 24 August, with conditions improving on both dates. Against the 16 September reading of 38, those checkpoints imply declines of 14 and 17 points respectively. Two things are worth stating before that number is repeated anywhere else. The 55 of 24 August is the local maximum of the reconstructed series, so measuring from it maximises the fall. And on that same reconstructed basis the score sat between 35 and 40 through July, which makes September’s 38 a return to the pre-rally range rather than a move to a new low.

Nor should the full decline be attributed to 16 September’s decision. Markets adjust to expected policy before a meeting, the dashboard combines inputs with different observation dates, and the score itself moved only a single point on its latest daily update. The published reading is not an isolated measurement of the announcement’s effect. It is also worth recording how much the Committee has changed in two months: in July the site described a Federal Reserve split 9-to-8 and deeply divided on whether to move at all. September’s increase carried every vote.

What August’s inflation report actually showed

August’s consumer-price report helps explain why the policy debate became harder rather than easier. Headline US CPI rose 0.4% on the month, seasonally adjusted, and 3.4% over the year, unadjusted. Core CPI rose 0.3% on the month, while its annual rate eased to 2.4%. Gasoline accounted for more than a third of the monthly headline increase. A firmer month alongside a softer annual core rate is mixed evidence rather than proof that every part of inflation is accelerating — a reading we set out at the time in our review of August’s record fund flows, where the same release coincided with gold futures falling on the week.

Policy responses differed across countries. The European Central Bank announced a quarter-point increase on 10 September, taking its deposit facility rate to 2.50% from 16 September. The Bank of Canada held at 2.25% on 2 September. Those decisions reflect different domestic conditions, and neither maps automatically into a fixed number of points in a US-focused tactical score.

What was known, and when

• 2 Sep The Bank of Canada holds: The policy rate stays at 2.25%.
• 10 Sep The ECB raises: A quarter-point increase takes the deposit facility rate to 2.50%, effective 16 September.
• 11 Sep US August CPI: Headline up 0.4% on the month and 3.4% over the year; core up 0.3% on the month with the annual rate easing to 2.4%.
• 14 Sep Two updates, two horizons: The World Gold Council reports a 20.2-tonne PBoC addition for August. The Long-Term Structural Score is refreshed at 66/100, Bullish.
• 16 Sep The Fed raises; the Tactical Score reads 38: A unanimous quarter-point increase to 3.75%–4.00%. The Bank of Canada publishes its 2 September deliberations the same day.
• 17 Sep The new range takes effect: Open market operations are directed to maintain the federal funds rate in the new range from this date — the day this article publishes.

Why diesel is the pressure point

Diesel deserves attention because it carries an energy shock well beyond the fuel pump. Lorries deliver goods, farm equipment helps produce them, and distribution networks move them between businesses. Higher fuel bills can become freight surcharges, thinner margins or higher selling prices, depending on contracts and on how much customers will tolerate.

Crude oil is only part of that bill. The US Energy Information Administration sets out the components — the cost of crude, refining, distribution and marketing, and taxes — and describes a second source of pressure that sits on top of the crude price: when refinery problems or lagging imports pull distillate inventories down, wholesalers and marketers may bid higher for the supply that is available. That is why an oil benchmark alone can understate what businesses are paying. Crude itself is not soft at the moment. The confirmation basket in our own score has WTI’s 13-week trend at plus 34.0%, which is the one place energy enters the tactical framework directly, and it scores as supportive for the inflation thesis rather than as relief.

The Bank of Canada’s deliberations for its 2 September decision, released on 16 September, identified this combination precisely: a protracted Middle East conflict and damage to refinery capacity could keep gasoline and diesel prices higher for longer than the Bank had assumed in July. Officials also reported limited evidence so far of broad spillover into other prices. The risk grows with duration; it should not be presented as an outcome already established across the economy.

Tariffs add a separate cost channel — one whose effect on metal prices we traced across a full year in our Liberation Day retrospective. A tariff raises the landed cost of an affected import unless suppliers, currency movements or margins absorb the difference, and replacing an established supplier takes time. Close trading relationships often involve integrated production networks, so changing their terms can be disruptive even when alternatives eventually emerge.

Federal Reserve Board staff research, published in April 2026 as a FEDS Note rather than as a finding of the Committee, estimated that the tariff changes implemented between February and November 2025 had raised the core PCE price level by 0.8% through February 2026. The authors judged pass-through from those 2025 tariffs effectively complete. The note is explicit about what it does not cover: the effects of the tariff changes arising from the February 2026 Supreme Court ruling against the tariffs imposed under the International Emergency Economic Powers Act. So the research supports the existence of a cost effect from a specific, dated cohort of measures. It does not support a claim that those tariffs keep adding to inflation indefinitely, and it is not a measurement of every tariff in force today.

This is where the price level and the inflation rate have to be separated. A product can jump from US$100 to US$110 and then stop rising. It remains expensive, but its rate of increase has fallen to zero. What would prolong the process is something different: repeated tariff increases, renewed fuel shocks, or wider wage and price adjustments.

Lyn Alden, an independent macro researcher, drew exactly this distinction in her May 2021 newsletter, describing the inflation surge then under way as “somewhat transitory in rate of change terms, but not very transitory in absolute terms”. That framework helps explain today’s affordability problem. It is a 2021 argument about how inflation episodes resolve, not a new forecast from her about September 2026.

Rate increases can restrain spending and reduce the ability of businesses to pass costs on. They cannot repair a refinery or remove a tariff. The Bank of Canada made the other half of the argument in the same deliberations: weaker growth caused by the trade conflict could contain inflationary pressure, provided the economy is running with excess supply and a soft labour market. The unresolved question is which force proves stronger — the increase in costs, or the reduction in demand.

What the score is actually made of

For gold, the bridge between the two channels is the return available elsewhere after expected inflation. If bond yields rise faster than inflation expectations, the income forgone by holding gold increases, and a stronger US dollar can add pressure for buyers using other currencies. These are tendencies rather than pricing rules — investment flows and demand for protection can offset them. The case against metals ownership in our Learning Academy works through that opportunity cost, including the years when it went badly for gold.

The 16 September reading captures the tension. Within the 50-point core-levels block, real yields earned just 2 of their 20 available points; the separate 30-point momentum block earned 6. Fiscal and Treasury stress still earned 8 of 10 — the component we added after the Treasury-market strains set out in our bond trap analysis, and the block closest to its ceiling. Diesel and tariff announcements are not inputs in their own right. Their influence arrives indirectly, through yields, inflation expectations and the currency.

Those are component readings on one day. They are not evidence that a particular headline caused a particular price move, and the site has made the related point before: strong demand over one period does not guarantee a favourable price response to the next inflation release. The market context section of our Learning Academy sets out the same drivers with their current readings.

Central-bank buying belongs to the other horizon

Official-sector demand provides a source of buying that is largely indifferent to a monthly inflation print. The World Gold Council’s China update of 14 September reported that the People’s Bank of China added 20.2 tonnes of gold in August — which the Council describes as the highest monthly increase since October 2023 — taking official holdings to 2,387 tonnes and extending the buying streak to 22 consecutive months. That is one month beyond the July figure discussed in our central-bank custody article, and it supersedes the superlative we recorded there for July. It also fits the pattern set out in our analysis of China’s buying strategy, where much of this year’s accumulation arrived while the price was falling. The Council is a membership organisation funded by gold-mining companies, and its purpose includes developing investment demand for gold. The figure establishes buying during August; it is not a response to a September decision that had not yet been taken.

The same official-sector argument runs through the work of John Reade, the Council’s senior market strategist, whom our central-bank coverage has drawn on before. Our profile records his position as the claim that the conventional macro drivers — real rates, the dollar, risk appetite — are no longer sufficient on their own, because official-sector demand has changed the structure of the market. He publishes within his employer’s mandate, which is to develop demand for gold, and that is context rather than endorsement. It does not establish a guaranteed floor beneath the price.

The Long-Term Structural Score reads 66/100, Bullish, in its 14 September update. Central-bank demand belongs to that slower-moving framework: the dashboard scores the component on a three-year average of annual purchases, and it already earns the full 10 of 10 points. Continued strong buying can sustain that support but cannot add points above the ceiling. A monthly purchase report does not mechanically lift the daily tactical reading, even though buying and the announcements around it can still influence market prices.

The Bigger Picture

The Committee published its Summary of Economic Projections alongside the statement. The median participant put the appropriate federal funds rate at 4.1% at the end of both 2026 and 2027, with median core PCE inflation of 3.4% this year easing to 2.5% next. Those are the individual projections of participants, not a plan the Committee has committed to, and they are revised at every quarterly round. Read alongside the score, they describe an environment in which the opportunity cost of holding a non-yielding asset is not expected to fall away quickly.

The observations worth watching from here are whether diesel costs and refinery pressure ease, whether price increases spread beyond energy, and whether two-year yields, real yields and the dollar stop tightening the environment for gold. Better inflation headlines alone may not be enough; how the market reads the policy response matters as much as the data.

The near-term reading and the reason a portfolio holds gold at all are separate questions, and September’s data bear on the first far more than the second. A worsening tactical score and an unchanged structural one are not a contradiction — they are two instruments measuring different things, which is the distinction between investing and chasing the price. The allocation section of our Learning Academy works through how a target allocation is arrived at; where it lands depends on an investor’s own objectives, horizon and tolerance for risk, not on a monthly score.

What the dashboard read next is in our review of the week to 2 October. The Tactical Score was 35/100, three points below the reading discussed here, in a week when the priced chance of a second increase fell and the 10-year real yield still rose to 2.92%.

A cost shock and the response to it can move gold in opposite directions at the same time. The score measures the balance between them on one day; it does not resolve it.

Sources and Methodology

The rate decision, the target range, the unanimous vote and the quoted statement wording are the Federal Open Market Committee’s own statement and implementation note of 16 September 2026; the effective date of 17 September is the implementation note’s. The projections are the Summary of Economic Projections released with that statement, and are participants’ individual projections rather than a Committee commitment. August consumer-price figures are the US Bureau of Labor Statistics news release for August 2026, published 11 September 2026; monthly changes are seasonally adjusted and annual changes are not. The ECB decision is the Bank’s own 10 September press release; the Bank of Canada decision is its 2 September press release and the characterisations of the Governing Council’s reasoning are its Summary of Deliberations, published 16 September 2026. The description of what makes up a diesel price follows the US Energy Information Administration’s Energy Explained page on the factors affecting diesel prices; the observation that crude is currently rising rather than softening is our own, taken from the WTI trend in our confirmation basket, and is not attributed to the Administration. The tariff estimate is Federal Reserve Board staff work published as a FEDS Note in April 2026 — FEDS Notes carry the views of their authors and are not statements of the Board or the Committee, a distinction that matters in an article about a Committee decision. The Lyn Alden quotation is the May 2021 newsletter recorded against her profile on this site, quoted rather than paraphrased because we hold the original wording. Central-bank figures are the World Gold Council’s China gold market update of 14 September 2026.

All score readings are our own. The Tactical Macro Conditions Score, its four blocks and every component value are the published Engine V2 reading for 16 September 2026, recomputed each weekday from FRED and public market data with every input and its source displayed on the dashboard. The July and August values are a retrospective reconstruction we published on 28 August 2026: Engine V2 went live on 25 August, those readings were calculated afterwards using thresholds calibrated partly on the same period, and they are therefore in-sample and benefit from hindsight. They cannot show that the framework predicted either the rally or the reversal. There are no published readings between 24 August and 16 September. The Long-Term Structural Score is the 14 September 2026 reading; its central-bank component scores a three-year average of annual purchases, which on the World Gold Council’s current full-year figures is about 1,002 tonnes a year across 2023–2025, comfortably above the 800-tonne threshold for full marks. That input was one window out of date on this site until this article was prepared, and was corrected from the 2022–2024 average (about 1,074 tonnes a year on current figures). When this article was first published, the corrected input was about 982 tonnes, calculated from the Council’s first estimates for 2023 and 2024, which it has since revised upwards. The component scores 10 of 10 on any of these figures, so the structural total is unchanged.

Conflicts of interest, stated plainly. The World Gold Council is a membership organisation funded by gold-mining companies and its purpose includes developing investment demand for gold; its central-bank statistics are the best available series of their kind, but its interpretation of why reserves are being accumulated is the analysis of an interested party. John Reade is its senior market strategist and publishes within that mandate. Lyn Alden is an independent researcher with no disclosed commercial interest in gold recorded on her profile; the asymmetry between the two attributions here is deliberate. GoldSilverPortfolio may earn a commission through two links elsewhere on this site: an affiliate link to SilverGoldBull, a bullion dealer, and a referral link to Wealthsimple, a brokerage. No product, fund or dealer is recommended here, and the Macro Conditions Score is our own framework — we publish its historical misses alongside its readings, and it is not a signal to buy or sell anything.

Primary sources

• Board of Governors of the Federal Reserve System — FOMC statement and implementation note, 16 September 2026: the target range raised by 1/4 percentage point to 3-3/4 to 4 percent, effective 17 September 2026, approved 12–0 with no dissents; “Inflation remains elevated. Today’s policy action will support a timelier return to the Committee’s 2 percent goal.”
• Board of Governors of the Federal Reserve System — Summary of Economic Projections, 16 September 2026: median projected federal funds rate of 4.1% at the end of 2026 and 2027 and 3.9% at the end of 2028, with a longer-run median of 3.2%; median PCE inflation of 3.7% in 2026 and 2.3% in 2027, and median core PCE inflation of 3.4% in 2026 and 2.5% in 2027
• US Bureau of Labor Statistics — Consumer Price Index news release for August 2026 (USDL-26-1496), released 11 September 2026: all-items CPI up 0.4% on the month seasonally adjusted and 3.4% over 12 months unadjusted, with gasoline accounting for more than a third of the monthly increase; core CPI up 0.3% on the month and 2.4% over 12 months
• European Central Bank — monetary policy decisions, 10 September 2026: the three key rates raised by 25 basis points, deposit facility to 2.50%, effective 16 September 2026
• Bank of Canada — interest rate announcement, 2 September 2026: the policy rate held at 2.25%
• Bank of Canada — Summary of Governing Council deliberations for the fixed announcement date of 2 September 2026, published 16 September 2026: a protracted conflict and damaged refinery capacity could keep gasoline and diesel prices higher for longer than assumed in the July Report; limited evidence so far of broad spillover into other prices; weaker growth from the trade conflict could contain inflationary pressure given excess supply and a soft labour market
• US Energy Information Administration — “Factors affecting diesel prices”, Energy Explained (undated; accessed 16 September 2026): the components of the retail diesel price, and the effect of refinery problems or lagging imports on distillate inventories and on what wholesalers and marketers bid for available supply
• Robert Minton, Madeleine Ray and Mariano Somale — “Detecting Tariff Effects on Consumer Prices in Real Time – Part II”, FEDS Notes, Board of Governors of the Federal Reserve System, 8 April 2026: tariff changes implemented from February to November 2025 raised the core PCE price level by 0.8% through February 2026, with pass-through from the 2025 tariffs effectively complete; the analysis excludes the effects of tariff changes arising from the February 2026 Supreme Court ruling against the IEEPA tariffs
• World Gold Council — “China gold market update: Official buying accelerated in August”, 14 September 2026: a 20.2-tonne PBoC gold reserve addition in August, described as the highest monthly increase since October 2023; official holdings up for 22 consecutive months, reaching 2,387 tonnes at end-August and 9% of total foreign exchange reserves, up from 8% in July
• Lyn Alden — May 2021 newsletter: the inflation surge described as “somewhat transitory in rate of change terms, but not very transitory in absolute terms”
• GoldSilverPortfolio — Macro Conditions Score dashboard, Engine V2: the 16 September 2026 tactical reading of 38/100 with its four blocks and fourteen component signals, and the 14 September 2026 Long-Term Structural Score of 66/100. The July and August 2026 tactical values are the retrospective reconstruction published on 28 August 2026