Week Ahead: Wednesday's CPI lands on a Fed with no forward guidance | Week-Ahead Brief #1

Wednesday, 8:30am ET. That is the week.

First issue, so here is the deal: one email a week with what is on the calendar, when it lands, and what is at stake. No stock tips, no filler, and no number in here I have not checked against a primary source.

July CPI lands Wednesday morning. Three things have stacked up to make it matter more than a normal inflation print.

Friday's jobs report broke the hiking case. Payrolls fell 23,000 against about +83,000 expected. Unemployment fell to 4.1%, which sounds fine until the reason lands: participation dropped to 61.4%, the lowest in more than five years. People left the count. May and June were both revised down, leaving a three-month average of +20,000.

On CME's FedWatch, odds of a September hold jumped to about 60%, from 45% on Thursday and roughly one in three a week earlier. Kalshi's contract has hold nearer 65%. The two do not agree on the split. They agree completely on the direction.

Worth keeping in view, because most coverage dropped it: CME still prices a 55% chance of a hike in October and almost 75% by December. The market is pausing the cycle, not calling it off.

The committee is already split. July 29 held at 3.50-3.75% with three dissents in favour of a hike, from Beth Hammack, Neel Kashkari and Lorie Logan. Most since September 2016.

And there is no forward guidance to lean on. Chair Kevin Warsh has taken it out of the post-meeting statements. Nothing steers expectations between meetings now, so the data does all of it. That is why a single CPI print is carrying this much.

What the forecasts actually say

The Cleveland Fed runs a nowcasting model on daily oil and weekly gasoline prices. Its August 7 update puts July CPI at 3.42% year over year against June's 3.5%, with core at 2.52% against 2.6%.

So Wednesday looks dull on the front page. Core is the number the Fed reacts to, and a dull-looking print can still settle September.

A syndicated economist consensus of 2.8% has also been circulating. It sits nearly a full point below both the nowcast and Friday's prediction-market pricing of 3.5-3.6%. Two of the three cluster and that one does not. My read is that it is the odd one out. I cannot reconcile it either way, and a forecast range this wide going into a print is worth knowing about on its own.

Now look at what the same model says about August: 3.45% headline, but a monthly jump of 0.38% against July's 0.09%. Brent settled at $83.55 on Friday, having dipped below $80 earlier in the week on hopes of a deal to reopen the Strait of Hormuz. It rose back above $84 on Monday after Iran's foreign minister said the waterway stays shut until Washington eases sanctions and pays reparations.

The energy shock lands in the August print, reported in September. Wednesday is the calm month.

Monday already taught us something

Two reporters landed before the open, and the first one is the more useful.

monday.com beat every line. Revenue of $364.6 million, up 22%, past both its own guide and consensus. Non-GAAP EPS of $1.48 against $1.11. Then it held the full-year revenue range at $1,466-1,474 million and guided Q3 below the street. The stock was quoted down about 9% pre-market.

Options had been charging 23.4% for that move, the widest single-name implied on the board this week. A clean beat, a cautious guide, and the realised move came in at well under half what the options wanted. Three more names carry double-digit implied moves this week. That is the calibration to hold in mind for all of them.

Barrick beat its production guide and missed on cash. 796,000 ounces against a 730,000-770,000 guide, but adjusted EPS of $0.82 against $0.84, and attributable free cash flow of $141 million against $1.21 billion in Q1. The realised gold price fell to $4,417 an ounce and capex rose to $978 million. Quoted down about 5.9%.

The largest thing in that release is not in the earnings at all. Barrick and Newmont settled every dispute over Nevada Gold Mines. Newmont pays a $1.95 billion cash top-up and has consented to Barrick's planned North American IPO, targeted for year end.

Both pre-market quotes above are snapshots. Neither is a close.

The calendar

Monday 10

6:00am ET, done: Barrick Q2
Before the US open, done: monday.com
After the close: Simon Property Group. Street looking for FFO near $3.21.
After the close: Rumble, now RUM Group, call 5:00pm ET

Tuesday 11, the heavy day

7:00am ET: Tencent Music, carrying a 10% implied move
Pre-open: Sea Limited, call 7:30am ET
Pre-open: Cardinal Health, closing fiscal 2026
Pre-open: eToro, webcast 8:30am ET
Around 4:10pm: Cava, call 5:00pm ET
After the close: CoreWeave, call 5:00pm ET
After the close: Super Micro, fiscal Q4

Wednesday 12

8:30am ET: JULY CPI
After the close: Cisco fiscal Q4, call 4:30pm ET, covering the quarter ended 25 July

Thursday 13

8:30am ET: July PPI
After the close: Applied Materials fiscal Q3, call 4:30pm ET

Friday 14

8:30am ET: July retail sales

Two calendar corrections, because getting these right is most of what this email is for. PPI is Thursday, not Friday. Our own week-ahead piece had it on Friday when it published and has been corrected. Several calendars elsewhere still show CPI on Tuesday. The BLS release schedule settles both.

The implied moves, and one that is not real

Sea Limited, 19.1%, struck against the $114.91 close on 6 August. One of the largest of the season. Five analysts sit in the EPS consensus and the spread runs $0.74 to $1.00, which covers most of it.
Super Micro, about 18%, against the $28.40 close on 31 July.
CoreWeave: no usable number. The roughly 12.5% quote in circulation traces to the prior quarter at a different share price. Anyone quoting a CoreWeave implied move this week is recycling a stale one, and I would rather say nothing than repeat it.

Both live figures were struck last week and will have moved.

One thing to watch for on Tuesday night. Super Micro's EPS consensus exists in two irreconcilable versions, $0.59 and $0.92, on different adjustment bases. Any headline calling it a beat or a miss against a single number is unreliable. Most of the quarter is public anyway, from a 21 July preannouncement claiming more than $60 billion of new orders. The margin behind that figure is the real question.

Also on the radar

Home Depot reports 18 August, Walmart and Deere on 20 August. If Friday's retail sales come in soft, those three are the confirmation or the refutation.
The SpaceX day-70 tranche lands around 21 August, roughly 319 million shares on our own derivation against an approximate date. Everyone expected the first unlock to crash the stock and it rallied 6.1%. The fear has been discharged. The supply is still scheduled.
Jackson Hole runs 27-29 August. In a year where the argument is whether to hike, it matters more than this year's payments-themed agenda suggests.

The one-line read

A market at a record high, a labour force that is shrinking, and one number on Wednesday morning. If core keeps easing the way the Cleveland Fed's model says, the September hike stays off the table and the three dissenters stay outvoted. If it does not, a Fed with no forward guidance and three votes already in the hawkish column has to move against a jobs market losing 23,000 a month. That is the outcome nothing on this calendar is positioned for.

Read further

The full week-ahead piece, with every figure sourced
Previews for this week: Sea Limited | CoreWeave | Super Micro | Applied Materials

Regards of Wall Street | regardsofwallstreet.com

The author holds no position in any security mentioned, and does not trade the securities covered on this site.

This email is published commentary sent to every subscriber at once. It is not investment advice and not a personal recommendation, and it cannot take account of anyone's individual circumstances. Anyone who needs advice on their own position should seek it from someone authorised to give it.