Fed's Collins Inflation Warning Comes True as Rate Hike Tests CPI

From conditional warning to policy reality

In late August, Boston Federal Reserve President Susan Collins said the Fed would need to raise rates soon unless incoming data showed inflation was still falling. At the time it read as a conditional warning. Within weeks it described what the central bank actually did. Her case was that high prices had become a "pervasive" worry for New England businesses and households, and that policy should tighten soon if lasting progress on inflation did not show up.23 She allowed that the policy rate of 3.5% to 3.75%, unchanged since December, might still deliver what she called "gradual disinflation," helped by rising longer-term bond yields.24 She also argued that the Fed could not wait indefinitely. After more than five years above target, she said, more time missing the goal risked shifting consumers' expectations in ways that would make 2% harder to reach.23

The evidence she wanted did not come. On September 16 the Fed raised its target range by a quarter point to 3.75% to 4%, its first increase in more than three years.511 A week later Collins wrote on LinkedIn that she had supported the move. She said she now saw "an increased likelihood" of outcomes in which inflation stays notably above 2%, and that a somewhat more restrictive funds rate would help bring inflation back to target for good.2 As she put it, "I did not see the inflation progress I was hoping to see." She pointed to renewed fighting in the Middle East since August and to companies in her district preparing to pass higher costs on to customers.8

The question now is whether the consumer price index will keep the Fed tightening. The September CPI report, due October 14, is the last inflation reading before policymakers meet on October 27–28.1112

What the consumer price index is saying

The 2026 inflation data have been unusually volatile. Annual headline CPI was 2.4% in January and February. It jumped to 3.3% in March, 3.8% in April and 4.2% in May, then eased to 3.5% in June and 3.4% in both July and August.20 The May peak came as the US-Israeli war with Iran pushed energy costs up. Coverage ties the June easing partly to a temporary US-Iran ceasefire that briefly lowered energy prices, and that truce has since broken down.82

The August report shows how uneven the picture is. Headline CPI was up 3.4% from a year earlier. Core inflation, which leaves out food and energy, was 2.4%, with energy prices up 16.3% over the year.11 Goods excluding food and energy rose only 0.7% from a year earlier, while non-energy services rose 3.0%.11 Accounts differ on the monthly headline figure: one tracker shows a 0.1% August gain16, while several other previews give 0.4%.1217 Every account agrees on the annual 3.4% and the 0.3% monthly core gain.

Forecasts for September broadly agree. The Cleveland Fed's nowcast puts headline CPI at 3.60% year over year and core at 2.39%.15 The FactSet consensus expects a 0.5% monthly rise, taking the annual rate to 3.6%, with core slowing to 0.2% for the month and holding at 2.4% for the year.19 Some Street estimates are slightly higher at 3.7%.17 Bettors on Kalshi expect about 3.6% and give less than a one-in-three chance that inflation tops its May high before year-end.1314

Gasoline accounts for most of the expected rise. The average price of regular gas hit $4.36 a gallon in September, 7.3% more than in August.11 Gasoline makes up only about 3.8% of the CPI basket, but a move that size can shift the overall index by roughly a quarter of a percentage point.11

Headline versus core: why the Fed isn't relaxing

Core CPI near 2.4% looks close to the Fed's goal, so a reader could fairly ask why officials sound so hawkish. Part of the answer is which measure the Fed watches. Its preferred gauge, the personal consumption expenditures index, runs well above CPI on the core measure. The Cleveland Fed estimates September core PCE at about 3.02% year over year.15 Officials' own projections put core PCE at 3.4% for 2026.17 When Collins first made her warning, economists expected July core PCE at a 3.3% annual rate, unchanged from June.23

The second reason is that officials no longer treat energy price spikes as temporary. Reuters reports that because inflation has stayed above 2% for so long, Fed officials have stopped looking past supply shocks they once would have ignored. That is true even though war-driven price pressure is hard to fight with interest rates.2 Richmond Fed President Tom Barkin has said price pressures now go beyond energy and tariffs.9 Kansas City Fed President Jeff Schmid said directly that the inflation problem is "not just about energy."10 There are also specific worries that higher costs will spread into other prices. Airline fares rose 23.4% over the year to August, diesel and delivery costs can feed into services prices with a delay, and University of Michigan respondents expect 4.6% inflation over the next year.17

Collins's own explanation fits this view. She has said firms are preparing to push costs onto consumers.8 She has also said a firmer labor market means policy can concentrate on "a timely return to price stability."1

A committee drifting hawkish, but not rushing

Collins is far from alone. Governor Michael Barr said inflation was "not clearly trending toward target in a timely way" and that further rate increases were likely needed.10 Philadelphia Fed President Anna Paulson said "some modest further tightening may be warranted."10 On October 9, St. Louis Fed President Alberto Musalem said that reaching 2% within about 18 months would require higher rates over the next six to nine months.7 In the September dot plot, 16 of 18 participants expected one more increase in 2026. The number projecting rates at 4% or higher by the end of 2027 rose to 14, from three in June.10

The tone has hardened, but there is no consensus on timing. Markets had priced in an October hike until New York Fed President John Williams and Vice Chair Philip Jefferson said there was no urgency. Expectations have since moved toward December.7 In late September, CME FedWatch put the odds of an October increase at about 53%.1 Chair Kevin Warsh describes the September move as removing "a dose of accommodation." He has not endorsed the projected path, consistent with his reluctance to give forward guidance.210 Collins herself, who does not vote on the FOMC this year, did not say in her LinkedIn post whether she wants another increase.2 Quartz, however, reported that she expects one more hike before year-end.8

Reading the stakes for October 14

The expected September report is not the clear evidence of sustained progress that Collins said would let the Fed hold off. It is also not the broad acceleration that would force a quick second hike. If core CPI comes in at 0.2% and headline rises about 0.5% because of gasoline, most forecasters consider that already priced in.18 Morningstar's preview argues that such a reading would give officials room to be patient, though Bank of America economists warn that a hot report could revive talk of an immediate hike.19 One preview counts a core reading of 0.3% or more, or a headline above 0.6%, as the upside surprise that would matter most for Fed pricing.18

The softer side of the data gets less attention. August PCE came in below expectations.10 The Cleveland Fed estimates third-quarter core CPI at an annualized 2.16%.15 Under the approach Collins set out in August, that would look like real progress if the oil shock fades.

The Fed's actions point to a different reading. Collins argued in August that the price of waiting was rising, and the committee acted as though she was right. Its dots, its language and its broad shift away from looking past supply shocks all suggest the burden of proof has moved. In this assessment, a core-only improvement is unlikely to be enough unless headline inflation and expectations also fall. If September CPI comes in near 3.6% as expected, the realistic choice is between hiking in October and hiking in December, with no cut on the table. That is broadly in line with the base case at J.P. Morgan of one more quarter-point increase before year-end.10

Collins's August warning has turned into the Fed's operating rule: policymakers will tighten unless the data show inflation is falling, and they will not assume it is falling without that evidence. Wednesday's CPI report is the next test.