Warsh Faces a 59.4 Percent Hike Bet and a 120 Basis Point Gap to the Target Index
Interest rate futures put the probability of a Federal Reserve increase at the 16 September decision at 59.4 percent on 5 September, against 54.4 percent a month earlier. The argument now running through Washington turns on which price index is being read. Administration officials have pointed to core consumer prices running at a 1.6 percent annualised pace over three months. The index in which the committee writes its own 2 percent objective is the headline personal consumption expenditures measure, and that stood at 3.7 percent in July. Over twelve months the two series are 120 basis points apart, and only one of them is the target.
The 1.6 percent is right and it belongs to the wrong index
The figure holds up. The Bureau of Labor Statistics published core consumer prices, seasonally adjusted, rising 0.2 percent in May, flat in June and 0.2 percent in July. Compounded and annualised, that run comes to 1.61 percent on our calculation. Nobody has invented anything.
Put the same three months through the personal consumption measure and the answer moves. Core personal consumption prices rose 0.4 percent in May, 0.1 percent in June and 0.2 percent in July. Annualised the same way, that is about 2.8 percent on our calculation, 122 basis points above the consumer reading over an identical window and on the equivalent core basis.
Neither of those is the objective. The committee’s Statement on Longer-Run Goals defines 2 percent inflation “as measured by the annual change in the price index for personal consumption expenditures”, the headline series, and the word core does not appear in it. Kevin Warsh put the same point in his own words at Jackson Hole on 28 August, calling the 2 percent objective, as measured by that index, a firm and fixed target. Over twelve months to July that index reads 3.7 percent.
| Measure, 12 months to July | Percent |
|---|---|
| Headline personal consumption expenditures | 3.7 |
| Headline consumer price index | 3.4 |
| Core personal consumption expenditures | 3.3 |
| Core consumer price index | 2.5 |
Consumer price figures from the Bureau of Labor Statistics release of 12 August 2026, personal consumption figures from the Bureau of Economic Analysis release of 26 August 2026. Both cover July 2026. Ranked high to low. The top row is the series the 2 percent objective is written in. The monthly changes behind the three month calculations above are published to a single decimal, and May core personal consumption stands at 0.4 percent in this vintage after being revised up from 0.3 percent at the August release.
Read across the same twelve months, the target index sits 120 basis points above core consumer prices on our calculation. The two core measures are 80 basis points apart on that window and 122 basis points apart over the latest three. Whichever pairing is chosen, the series the administration cites is the lowest number on the page and the series the objective is written in is the highest.
Warsh gave a second number at Jackson Hole that cuts the same way. He put the personal consumption index at 3.7 percent over twelve months and 4.1 percent over six, so the recent half year is running 40 basis points hotter than the year behind it on our calculation. He also disaggregated the index into its 199 components and found 54 percent of the basket rising more than 3 percent over the year, roughly 107 components on our calculation, against a post pandemic peak near 77 percent and a pre pandemic norm of 32 percent. Measured against that norm, the excess breadth has fallen from 45 points to 22, so slightly more than half the distance back has been covered and the rest has not.
A jobs print more than five times the recent average
Payrolls rose 162,000 in August against an average monthly gain of 31,000 over the prior twelve months, which is 5.2 times that average on our calculation. June was revised up by 11,000 and July by 44,000, leaving the two months 55,000 higher than first reported, so part of the weak summer the market had priced turns out to have been a measurement problem. Unemployment held at 4.1 percent.
Pay is the part that complicates both cases. Average hourly earnings rose 10 cents in the month to 37.75 dollars, up 0.3 percent, and 3.1 percent over the year. Set against headline consumer prices at 3.4 percent, hourly pay is going backwards by 0.3 points on our calculation, though the price reading is a month behind the earnings reading and that comparison will not settle until the August consumer price report on 11 September. Set against core consumer prices at 2.5 percent, the same pay growth is positive by 0.6 points.
The chairman has already discounted that channel in any case. Wage growth, he said at Jackson Hole, “has not proven a reliable indicator of future inflation for a very long time”. A committee that takes him at his word is not going to hold because pay looks contained.
| Hike probability for 16 September | Percent |
|---|---|
| 5 September | 59.4 |
| 28 August | 57.0 |
| 5 August | 54.4 |
CME Group FedWatch, captured at 14:19 GMT on 5 September 2026, with the two earlier readings as the tool itself dates them. The week containing the jobs report added 2.4 points and the month added 5 points, on our calculation. On the 5 September reading the residual is a hold at the current range at 40.6 percent and the probability of a cut reads zero. The measure moves with the futures tape, so each row is a timestamp rather than a close.
The committee was already split before any of this
On 29 July the committee held its target range at 3.50 to 3.75 percent by nine votes to three. Beth Hammack, Neel Kashkari and Lorie Logan each preferred to raise by a quarter point. A quarter of the votes cast at a meeting that changed nothing were votes to tighten, and that dissent predates the August data by a month.
The range explains the discomfort. Its midpoint of 3.625 percent sits 0.075 points below the index the objective is written in, so on that comparison the policy rate is fractionally beneath inflation rather than above it. Warsh reached a similar conclusion by a different route at Jackson Hole, saying credit and loan markets were showing few signs of policy restraint and that he would be hard pressed to describe broad financial conditions as restrictive. A simple subtraction of current inflation from the funds rate is not a measure of policy stance, which also depends on expected inflation and on where neutral sits, but the chairman’s own reading of financial conditions points the same way.
What the administration has said
President Donald Trump wrote on Truth Social on Friday 4 September that the United States should have the lowest rate of any country in the world, and that unless rates came down he would stop trading with countries with which the United States runs a deficit. He called that course better than tariffs. The instrument named in the post is the trade relationship itself rather than a tariff schedule.
Vice President JD Vance said at a White House briefing on 3 September that the central bank should be lowering rates, and that it would be nice to have some help. Peter Navarro, a senior economic counselor, said in an interview on 4 September that a hike would be careless and would hit precisely the sectors the country most needs to prosper. Treasury Secretary Scott Bessent argued that the bank does not usually raise rates during a supply shock until second or third order inflationary effects appear. The week’s remarks were reported by CNBC.
Warsh has been chairman since 22 May 2026. No approved source carried a response from him to any of them, and this report does not supply one.
Why it matters: The September decision rests on a measurement question rather than a growth question, which is an unusual place for a central bank to be ten days out. The series the administration cites has slowed to 1.6 percent annualised. The series its own objective is written in stands at 3.7 percent, and by the chairman’s arithmetic it is running hotter over six months than over twelve. Holding means accepting that the target series is running at close to double the target while the economy adds 162,000 jobs in a month. Raising means acting against the stated position of the President, the Vice President, the Treasury Secretary and the President’s senior economic counselor, all put on the record in the same week. Either way the committee has to choose in public, and the choice will be read as an answer to more than the data.
Outlook: The August consumer price report lands on Friday 11 September, four days before the meeting opens on 15 September. Personal consumption figures for August do not arrive until 30 September, after the decision, so the committee goes into the room with a fresh reading of the series it does not target and a stale one of the series it does. Warsh has set his own bar for easing: confidence that underlying inflation is moving to the objective clearly and at sufficient speed. Nothing published this week met it.
Sources: Federal Reserve, Bureau of Labor Statistics, Bureau of Economic Analysis, CME Group, CNBC.
