July's FOMC Minutes Land Wednesday. The Forecast Is Stale; the Conditions Are Not.
At 2:00 p.m. Eastern on Wednesday, Aug. 19, the Federal Reserve will publish the minutes of the Federal Open Market Committee meeting held on July 28-29. The Fed's own August events calendar lists the release. By the time it lands, the meeting it describes will be three weeks old, and five significant data releases will have intervened. On the arithmetic alone, the minutes are a document about an economy that no longer quite exists.
That is the easy observation, and it is largely correct about one half of the document. It is wrong about the other half, and the half it is wrong about is the half that moves markets.
Start with what the committee actually did. The July statement, published on the Fed's website, kept the target range for the federal funds rate at 3-1/2 to 3-3/4 percent. The vote was 9-3. All three dissenters — Beth M. Hammack of Cleveland, Neel Kashkari of Minneapolis and Lorie K. Logan of Dallas — are regional Reserve Bank presidents, not Board governors, and all three preferred to raise the target range by a quarter point. This cycle's contested move is a hike. It was the first time since September 2016 that three dissents had come in the same policy direction — a point made independently by The Motley Fool's account of the meeting, by the Wall Street Journal's Nick Timiraos and by a market brief published by Kraken.
The statement's economic assessment is the part that has aged. It said that economic activity was expanding at a solid pace despite elevated uncertainty owing in part to the conflict in the Middle East, and that job gains had kept pace with the workforce while the unemployment rate had changed little. On inflation it said that price growth remained elevated relative to the committee's 2 percent goal, in part reflecting supply shocks that had driven price increases in certain sectors, including energy. It then closed with an unusually flat declarative, one of the few phrases short enough to survive the new format intact: "The Committee will deliver price stability."
Nine days later, the Bureau of Labor Statistics reported that both nonfarm payroll employment, at -23,000, and the unemployment rate, at 4.1 percent, changed little in July. Payrolls fell. The BLS also revised May down by 66,000 and June down by 37,000, a combined 103,000 reduction. Average hourly earnings were up 3.2 percent from a year earlier and the labor force participation rate was 61.4 percent, which the BLS noted had declined seven-tenths of a percentage point since January. The July CPI, released Aug. 12, showed headline prices up 0.1 percent on the month and 3.4 percent on the year, with core up 0.2 percent and 2.5 percent, figures reported by Kiplinger among others. The Producer Price Index for final demand was unchanged in July on a seasonally adjusted basis, the BLS said on Aug. 13. Census Bureau advance figures reported on Aug. 14 showed retail sales down 0.6 percent on the month while still running about 5 percent above a year earlier, and the University of Michigan's preliminary August sentiment index fell to 51.0 from a July final of 55.2.
The repricing followed, and it ran one way. According to CME Group's FedWatch tool as cited in press reports, the implied probability of a quarter-point September hike stood at 67 percent on July 31, was in the mid-50s by Aug. 5, and had fallen to roughly 44 percent by Aug. 7, after the jobs report. Kiplinger, citing the same tool, reported that odds of a hold rose to 64 percent after the Aug. 12 CPI print from 52 percent the prior day; Benzinga, also citing FedWatch, put the hike probability at about 38 percent that day, down from 48.4 percent on Aug. 11. These are secondary citations of a futures-derived measure rather than a primary exchange reading, and the levels vary by a few points from outlet to outlet and by more than that against prediction-market quotes. They are best read as a direction of travel — consistently and substantially lower across every source — rather than as a precise level on any given afternoon.
So the committee's July picture of a labor market keeping pace has been contradicted, and the market has acted accordingly. Here is the part that has not been contradicted, because it was never a picture in the first place.
All three dissenters published individual statements explaining their votes on July 31, and those statements are the best available preview of the reasoning the minutes will render in aggregate. Read carefully, each contains not a forecast but a condition — a stated circumstance under which the author would change position. Conditions do not go stale when data arrives. They get tested.
Logan's statement, published by the Dallas Fed, is the one the subsequent data hit hardest. She wrote that inflation appeared to be trending toward the mid-2's rather than all the way to 2 percent even after accounting for productivity gains and temporary supply shocks, with risks to the upside. She also wrote that "the job market is solid and perhaps strengthening a bit." That sentence was seven days old when payrolls printed negative and the prior two months were revised down by more than 100,000. Her policy conclusion rested on a two-legged argument — inflation drifting above target, employment strong enough to absorb restraint — and one leg has been kicked. Her own framing of the trade-off was that "Modest action in the near term would reduce the likelihood of needing to take sharper action later," and that the committee "cannot count on unanticipated shocks to achieve its goals and can always adjust policy if unanticipated shocks occur."
Kashkari's statement, published by the Minneapolis Fed, is more explicitly conditional and contains its own exit. He argued that a series of successive supply shocks, unlike an individual one, can entrench higher inflation, drawing a comparison to the 1970s while noting that unemployment is lower and inflation much lower today. His operative sentence was that to manage the risk of entrenchment, "I would rather tighten policy incrementally as we gather more data on the path of inflation and employment." The final clause is the one that matters now: employment is precisely the variable that has since deteriorated, and by his own construction it is one of the two inputs he said he was waiting on. He then supplied the off-ramp himself: "On the other hand, if inflation durably fades, a strategy of small policy steps would allow the FOMC to slow or pause subsequent adjustments without unnecessary impact on the real economy."
Hammack's statement, published by the Cleveland Fed, rests on breadth rather than level. She wrote that inflation "has remained stubbornly above 2 percent for more than five years, and I am not confident it will return to our objective on its own," and that "Businesses describe pricing pressures as broadening rather than fading, and consumers are expressing despair over persistently higher prices." Crucially, she addressed the supply-shock objection directly: "Supply-side factors, including energy prices, have boosted inflation this year, but I see inflationary pressures coming from the demand side of the economy, as well." Her conclusion, hedged to her own judgment: "In my view, now is the time for the FOMC to act to speed the return of PCE inflation to our 2 percent objective and deliver on our commitment to price stability for the American people."
That framing matters because the disinflation of the past week was not broad. The BLS reported that the flat headline PPI reflected offsetting moves: final demand services rose and construction prices jumped, while final demand goods fell 0.7 percent on a 3.1 percent drop in energy, with gasoline down 5.7 percent. Final demand less foods, energy and trade services rose 0.4 percent on the month. On the CPI side, headline at 3.4 percent sits well above core at 2.5 percent. Energy did the cooling. Whether that counts as progress or as exactly the supply-driven noise the dissenters said they would look past is the argument the minutes will document — and it is an argument the past week's data sharpens rather than settles.
There is a second reason Wednesday's document carries more weight than a routine minutes release. Chair Kevin Warsh, sworn in on May 22, has substantially narrowed the Fed's other communication channels. As American Banker has reported, the policy statement has been cut to roughly 130 words from a typical 300 to 400, and forward guidance has been dropped outright — "We dropped forward guidance," Warsh said. Writing for PIMCO after Warsh's first meeting in June, Tiffany Wilding noted that he did not submit his own projections for the Summary of Economic Projections and emphasized that the committee is not bound by its projections. Jerome Powell, it is worth noting, remains on the Board as a governor.
Strip out forward guidance, shorten the statement, and remove the chair's dot, and the minutes become one of the few remaining places where the committee's reaction function is written down at all. Whether they will do that job is a live question. Axios reported ahead of the June minutes that JPMorgan's chief U.S. economist Michael Feroli did not expect material format changes but saw the risks tilting toward less transparency in Fed communications. Don Kohn, a former Fed vice chair, told American Banker: "Markets, the public and the legislature need to understand how the Fed is doing things, why it's doing what it's doing, how it might react to incoming data. Being completely opaque on that is not a good idea and it's not conducive to accountability." The June 16-17 minutes, when they appeared on July 8, ran to roughly a dozen pages and did retain the traditional structure, including a section on participants' views that laid out branching conditions — broadly, that further firming would likely be warranted if inflation stayed elevated while employment held up, and that holding or eventually lowering the range would be appropriate if price pressures dissipated. That is the template to check Wednesday's document against.
The June projections also quantify how close the internal argument is. Of the 18 dots submitted — Warsh abstained — nine participants forecast at least one rate increase in 2026, eight expected the range to be held steady and one projected a cut, according to Wilding's account, lifting the 2026 median policy rate to 3.8 percent from 3.4 percent in March. A committee that splits nine-eight-one is not one where three dissents can be dismissed as fringe. The specific number the minutes may reveal is how many participants who voted with the majority in July nonetheless indicated they were close to the dissenters — the distinction between a comfortable hold and a reluctant one.
What the minutes cannot do is decide September. The committee next meets Sept. 15-16, and the Fed's calendar marks it as a meeting associated with a Summary of Economic Projections — a fresh dot plot, minus the chair's. Between now and then, PCE inflation is scheduled for Aug. 26, and another employment report and CPI release will arrive. Those are the inputs. The minutes are the record of how the committee said it would weigh them.
Markets go into the week close to highs and with a front end that has not fully abandoned the tightening case. The Russell 2000 closed at a record 3,068.42 on Friday, Aug. 14, up 0.51 percent, while the S&P 500 slipped 0.17 percent to 7,785.76 and the Nasdaq Composite fell 0.28 percent to 26,729.16. Advisor Perspectives' Treasury yield snapshot put the two-year note at 4.17 percent and the ten-year at 4.68 percent at Friday's close — with the two-year still sitting above the top of the current 3.50-3.75 percent target range. Trading Economics data showed Brent crude roughly a third higher than a year earlier as of Aug. 14 — the page carried inconsistent readings of the level itself, so no precise figure is printed here — with Strait of Hormuz disruption the standing overhang behind the energy line in every one of these prints.
The week around the minutes is busy in its own right. Monday brings the Empire State manufacturing survey and the NAHB housing market index; Tuesday, housing starts and building permits, import and export prices, and industrial production; Wednesday, a 20-year bond auction at 1:00 p.m. before the minutes an hour later; Thursday, jobless claims, the Philadelphia Fed manufacturing index and the Conference Board's leading index; Friday, S&P Global's flash US PMIs. Reddit joins the S&P 500 before Tuesday's open, according to S&P Dow Jones Indices. Traders looking for the September answer will find more of it in those releases than in Wednesday's transcript of an old argument — but the transcript is where they will learn the terms on which the argument gets settled.
