Three Weeks That Repriced September — and the Fed Minutes That Could Reprice Them Back
The Russell 2000 finished Friday, August 14, at 3,068.42, up 0.51% and at a record close — the only major US index higher on the day. The S&P 500 slipped 13.23 points to 7,785.76 (-0.17%), the Nasdaq Composite fell about 74 points to 26,729.16 (-0.28%), and the Dow Jones Industrial Average lost about 0.20%. Over the five sessions, the small-cap benchmark gained 1.11% against 0.36% for the S&P 500, 0.14% for the Nasdaq and a 0.56% decline for the Dow, according to the session wrap published by InvestingLive on August 14.
Those weekly figures reconcile against the prior Friday's closes reported in a week-in-review published by Hustlin' on August 7 — S&P 500 7,757.64, Nasdaq Composite 26,690.62, Dow 54,036.93 — to within a rounding error on all three, which is a useful check at a moment when several outlets published mutually inconsistent Friday numbers. TheStreet's own August 14 wrap, for instance, printed a Russell gain of 0.54%, a figure that does not follow from the index's Thursday close. It was the small-cap index's second record close in as many sessions. Schaeffer's Investment Research reported on August 13 that both the S&P 500 and the Russell 2000 notched record closes that day, attributing the move to producer-price data.
The tempting story writes itself: over the same three weeks the Russell was grinding to records, the market's assessment of whether the Federal Reserve will raise rates in September collapsed. That trail is worth laying out with dates attached, because it is genuinely dramatic — and then it is worth being careful about what it does and does not prove.
Start in late July. J.P. Morgan Wealth Management, in a piece by Sergei Klebnikov published on the Chase site on Wednesday, August 5, wrote that "Markets are also leaning in that direction, with futures pricing implying a roughly 65% chance of a September hike." In the same piece, J.P. Morgan Wealth Management's chief investment strategist, Phil Camporeale, described the setup this way: "The combination of a slower-than-expected normalization of supply chains around the Strait of Hormuz and market questioning of inflation-fighting credibility after the July FOMC meeting has lowered the bar for a rate hike in September." Writing for The Motley Fool on August 12, Sean Williams put the CME FedWatch reading at 67% for a quarter-point September hike as of Friday, July 31.
The hinge was Friday, August 7. The Bureau of Labor Statistics reported that nonfarm payroll employment fell by 23,000 in July and the unemployment rate was 4.1%, with 6.9 million people unemployed. The release's opening line was blunt about how little it saw: "Both nonfarm payroll employment (-23,000) and the unemployment rate (4.1 percent) changed little in July." Employment declined in local government education and retail trade; health care continued to trend up; average hourly earnings rose two cents.
The revisions did more damage than the headline. BLS marked May down by 66,000, to a gain of 63,000 from 129,000, and June down by 37,000, to a gain of 20,000 from 57,000 — 103,000 jobs erased from the prior two months. Quartz reported the same day that economists had been looking for something in the range of 83,000 to 95,000 new jobs; Williams's Motley Fool piece put the consensus at 85,000. By August 7, Williams wrote, the FedWatch probability of a September hike "had dipped to 44.4%," while on Polymarket "the odds of a rate hike in September fell from roughly 60% to 40%" over the July 31 to August 7 stretch.
Then came inflation. The July Consumer Price Index was released Wednesday, August 12, at 8:30 a.m. ET — not Tuesday, as some calendars circulating this week had it. BLS reported that the CPI-U "increased 0.1 percent on a seasonally adjusted basis in July after falling 0.4 percent in June," with the all-items index up 3.4% over 12 months. Core CPI rose 0.2% on the month and 2.5% over the year. Energy fell 1.5% on the month but remained up 14.7% year over year; shelter rose 0.1% and 3.2% respectively.
Markets moved on it immediately. The Motley Fool, citing CME FedWatch as of 11:32 a.m. ET on August 12, reported hike odds had "declined from over 48% to roughly 38%," leaving "nearly a 62% chance the Fed holds rates within its current 3.50%-3.75% range in September," up from about 52% the previous day. 24/7 Wall St., writing on August 13, tracked the same repricing with different numbers — 55% heading into the week, 42% after CPI, 32% after Thursday's producer-price data. Two outlets, one event, a 10-point gap in the readings; the discrepancy is a reminder that FedWatch is a live snapshot and the timestamp matters as much as the number.
Producer prices were flat month over month in July, released Thursday, August 13. 24/7 Wall St. put the 12-month PPI rate at 4.7%, down from 5.5% in June; the Bureau of Labor Statistics release itself carries the same two figures, so the number rests on the primary document rather than on a single aggregator. Then on Friday, August 14, the Census Bureau reported that retail and food services sales fell 0.6% in July to $763.6 billion from $768.1 billion in June, still up about 5% from a year earlier, per Quartz's account of the release. The Washington Post headlined it as the biggest monthly drop in more than a year.
By Friday there was no single number for September, which is itself the honest finding. Charles Schwab's pre-open note, timestamped 9:13 a.m. ET on August 14, said of the FedWatch-implied probability of a September increase that "Chances were just above 30% as of this morning." Dow Jones Newswires, in a week-ahead piece carried by MarketScreener on August 14, reported that "U.S. money markets price only a 27% chance of a 25 basis-point rate increase next month, with a rate increase not fully priced until early next year, LSEG data showed." Benzinga's midday market summary at 1:22 p.m. ET on August 14 reported Polymarket odds of a 25-basis-point hike down to 24%, with unchanged at 76%. And IG's Tony Sycamore, writing August 14, framed it in basis points: "The US rates market is set to end the week pricing in 9 bp of hikes for the September meeting and a total of 23 bp of Fed tightening by year-end."
So: a repricing from roughly two-thirds to roughly a quarter-to-a-third, over three weeks, alongside small-cap leadership. What that co-movement does not establish is causation. Small caps' rate sensitivity is a widely argued mechanism, not a measured one, and the honest version of this story names who argues it and on what evidence rather than asserting it in a newsroom's own voice.
The most specific version on the record this year came from Goldman Sachs. In a note reported by Business Insider on July 2 in a piece by Samuel O'Brient, Goldman's Ben Snider wrote: "A Fed tightening cycle would pose a bigger risk to small-caps than large-caps ... While our economists believe that the Fed will most likely remain on hold for the next few quarters, the market is pricing roughly 35 bp of Fed tightening through year-end. About 30% of Russell 2000 debt is floating rate, compared with 7% for the S&P 500." The ellipsis is Business Insider's, and what follows it is a hedge worth keeping: Snider's own economists expected a hold.
The floating-rate statistic is the load-bearing claim, and estimates of it vary. Writing for Investing.com on April 20, Peace Longe put it at "Roughly 32% of Russell 2000 debt is tied to floating rates, compared with just 6% for the S&P 500," and argued small caps face "a $368 billion maturity wall in 2026" with debt issued in the low-rate era rolling over near 6.5%. The gap between those two estimates is small, but they come from different desks with different definitions, and neither is an index-provider statistic. Profitability estimates diverge much more sharply. Snider's note said that "While analyst estimates point to strong EPS growth, roughly a quarter of the Russell 2000 is unprofitable," specifying that "Unprofitable stocks represent 29% of Russell 2000 constituents and 23% of market cap." Sami Khan, writing for IBTimes UK on June 4, cited an analysis attributed to Morningstar putting the share of Russell 2000 companies generating no earnings at about 40%, up from 17% two decades ago. Longe's Investing.com piece went further still, reporting that "Analysts now estimate that 41% to 46% of Russell 2000 companies cannot cover their interest expense with operating profits." Somewhere between a quarter and nearly half of the index is fragile, depending on whose screen and whose definition — a spread wide enough that anyone leaning on it should say which one they mean.
There is also a competing explanation for small-cap leadership that has nothing to do with the Fed. The Russell 2000 rose about 21% in the first half of 2026 — its best first half since 1991, a comparison drawn at the time by CNBC, by Briefs and by Schaeffer's Investment Research, whose own tally put the gain at "more than 20%" — and the attribution at the time was artificial intelligence, not interest rates. Amy Zhang, a portfolio manager at Alger, said in comments reported on June 30 that "I think a significant part of the small cap story is tied to AI. The impact of AI investment trickles down from large-cap leaders to small-cap companies." Adam Turnquist, chief technical strategist at LPL Financial, added in the same coverage that "Small-cap leadership has been notable amid the mega-cap-driven bull market, although small caps have meaningful exposure to semiconductors and technology hardware. Building fundamental strength has also helped offset headwinds from higher rates" — an argument that small-cap gains have come despite the rate backdrop rather than because of it, and one that ties the leadership back to chip and hardware exposure rather than to the Fed. Snider's July note flagged a complication for that thesis too: the June index reconstitution cut the weight of AI infrastructure names in the Russell 2000 from 15% to 7%.
Which brings us to Wednesday. The Federal Reserve will release the minutes of the July 28-29 FOMC meeting at 2:00 p.m. ET on Wednesday, August 19, per the Board's own August calendar. The July statement held the target range at 3-1/2 to 3-3/4 percent and recorded the vote as 9-3, with the dissent line reading: "Voting against the monetary policy action were Beth M. Hammack, Neel Kashkari, and Lorie K. Logan, who preferred to raise the target range for the federal funds rate by 1/4 percentage point at this meeting." The Motley Fool noted on July 31 that this was "the first time three dissents have occurred in the same policy direction since September 2016."
The vote tells you three officials wanted a hike. The minutes tell you something the vote cannot: how many more were close to it and chose not to break ranks. That is the variable the past three weeks of repricing has had no way to observe. Daniela Hathorn, senior market analyst at Capital.com, put the point plainly in a note reported by Dow Jones Newswires: "Wednesday's FOMC minutes will be closely examined for the extent of disagreement within the Fed." In comments carried by TheStreet on August 14 she also cautioned that "the inflation story isn't entirely comfortable. CPI is moving in the right direction, but remains above target, while longer-dated Treasury yields remain elevated amid concerns around fiscal borrowing and structurally higher demand for capital."
There is a documented reason to think the hawkish bloc is wider than three. In the Summary of Economic Projections published with the June 16-17 meeting, the median participant put the federal funds rate at 3.8% at the end of 2026 — above the midpoint of the range the Committee has since held. The central tendency ran 3.6% to 4.1% and the full range of the 18 participants' projections ran 3.4% to 4.4%, against a longer-run median of 3.1%. Those same projections put 2026 median core PCE inflation at 3.3% and headline PCE at 3.6%, with unemployment at 4.3% and real GDP growth at 2.2%. July's actual 4.1% jobless rate came in below that projection; the inflation medians remain far above the Committee's 2% goal. That matters for September specifically, because the September 15-16 meeting is one of the four in 2026 carrying a Summary of Economic Projections, per the Board's meeting calendar. The dots get refreshed. If Wednesday's minutes reveal a bloc that was closer to hiking than the 9-3 tally suggests, the September dot plot is the instrument that would make that visible in numbers rather than in prose.
The obvious caveat: the minutes describe a meeting that this month's data has already overtaken. July's CPI landed on August 12, PPI on August 13 and retail sales on August 14 — all after the July 29 decision, and all after the discussion the minutes record. Nothing in Wednesday's document reflects a payroll print of -23,000, a 0.6% drop in retail sales, or two consecutive benign inflation reports. Minutes are a record of deliberation, not a forecast, and they carry no update.
One current the disinflation story does not cover: the Committee's own July statement said inflation "remains elevated relative to the Committee's 2 percent goal, in part reflecting supply shocks that have driven price increases in certain sectors, including energy," and described economic activity as "expanding at a solid pace despite elevated uncertainty that owes, in part, to the conflict in the Middle East." July CPI energy was down 1.5% on the month but still up 14.7% over 12 months. Brent and WTI both traded up roughly 5% on the week, on renewed US pressure on Iran and the continued effective closure of the Strait of Hormuz — the same channel Camporeale identified on August 5 as having "lowered the bar" for a September move. A benign July does not settle an energy shock still in progress.
For small-cap investors the structural risks sit underneath all of this and do not move with the odds screen. Companies in this segment are more likely to carry floating-rate borrowings, to face near-term refinancing at rates well above what they originally issued at, and — on every estimate cited above, from Goldman's quarter to Investing.com's 41%-46% — to include a large cohort that does not currently earn enough to cover interest. Index-level records say nothing about the dispersion beneath them. Goldman's Snider, for what it is worth, wrote in early July that "The combination of elevated valuations and near-trend US economic growth points to low single-digit Russell 2000 returns in the next 12 months" — a house forecast, reported here as one, from a note published before the last six weeks of data.
Markets closed Friday at 4:00 p.m. ET and nothing has traded since. The week ahead brings the NAHB housing index Monday, housing starts and permits plus industrial production Tuesday, the minutes Wednesday at 2:00 p.m. ET, jobless claims and the Philadelphia Fed survey Thursday, and S&P Global's flash PMIs Friday. Of those, only one speaks directly to the question that has moved small caps for three weeks — and it will do so by describing a room whose information set is now three weeks out of date.
This article is news and analysis, not investment advice. Nothing here is a recommendation to buy, sell or hold any security. Small-capitalization stocks carry heightened risks including illiquidity, dilution, refinancing and going-concern exposure, and index-level performance is not indicative of individual holdings. Market-implied rate probabilities are point-in-time readings that change continuously; every figure above is dated and attributed to the outlet and venue that reported it.
