The Week Ahead: Everybody Says They've Cut Back

Last week the government asked Americans how they feel about money and got 51.0 on a scale where 100 means fine. Then it counted the actual receipts and found retail sales fell 0.6%, the first decline since October. Two data points, one obvious conclusion: the consumer is done.

Except the companies that run the actual registers all report this week. Home Depot Tuesday, Target and Lowe's Wednesday, Walmart Thursday. A survey is what people say into a phone. Same-store sales, meaning what an existing store rang up versus a year ago, is what people did with a card in their hand. Those two have disagreed before, and the register has a better record.

Wednesday afternoon the Fed publishes the minutes of a meeting it held three weeks ago, which is the institutional equivalent of texting an explanation the following month. If you need catching up on how we got here, last week's Week In Review covered the light that turned green while nobody moved.

Everything dated lives here. The rest of this newsletter is why it matters.

The market has already picked a winner, which is usually the fastest way to get embarrassed. Walmart rose 3.06% last week while Home Depot fell 4.71%, and the retail sector ETF dropped 2.06% in a week the S&P 500 managed to gain 0.40%. Traders have decided the discounter eats and the home-improvement guy starves.

You can see the logic. When money gets tight, people still buy groceries and skip the new deck. Home Depot and Lowe's sell projects, and projects are the first thing to get postponed when a mortgage costs 6.67%, and the kitchen can wait another year. Walmart sells dinner, and dinner does not wait.

The number that matters in all four reports is not earnings per share. It is same-store sales and, more importantly, what management says about the back half of the year, because a retailer cutting its own forecast in August is telling you what it sees in the Christmas order book. Home Depot closed Friday at the lowest price of its week, so somebody has already read the tea leaves and did not like the flavor.

What could go wrong: all four beat, the "consumer is dead" trade unwinds in a single session, and last week's ugly retail sales print gets revised into a rounding error. Or Walmart guides down and takes the whole tape with it.

Wednesday at 1:00 PM CT, the Fed releases the minutes from its July 28-29 meeting. We already know what happened: rates were held within the target range of 3.50% to 3.75% on a 9-3 vote, with all three dissenters voting to raise rates. That is the unusual part. Dissents normally come from the doves who want cheaper money. These came from the other direction, which is why last week's edition focused on the hike nobody wants to talk about.

So the minutes are not a spoiler; they are a temperature reading. The question is how close the nine were to becoming eight or seven, and whether the phrase "further tightening may be warranted" appears in a form the bond market must take seriously. Cleveland Fed President Beth Hammack, who votes this year, said on August 10 that she expects more than one hike will be needed and does not see the current range as meaningfully restricting anything. She has not exactly buried the lede.

Markets are pricing in a roughly 1-in-4 chance of a September hike, with about a 73.5% chance of no change at all when the committee meets September 15-16. Yet the odds of at least one hike landing in 2026 are nearly a coin flip. Traders have effectively decided the hike is coming to a meeting near you, just not the next one, which is the monetary-policy version of agreeing to go to the gym in the fall.

There is also a bond auction at noon that same day, an hour before the minutes. Last week's 30-year sale went badly. If the 20-year goes the same way and then the minutes read hawkish, Wednesday afternoon gets loud.

What could go wrong: the minutes show the three dissenters nearly had company, September repricing starts immediately, and every rate-sensitive thing you own has a bad Wednesday.

Monday's builder confidence survey is expected at 33. Anything under 50 means more builders describe conditions as poor than good, so 33 is not a soft patch; it is a mood. Builders have been living below 50 for a long stretch now and have mostly stopped pretending otherwise.

Tuesday brings the harder numbers. Housing starts are forecast to fall 4.7% after a 19% jump in June, which sounds alarming until you remember that a 19% monthly leap in construction was never the new normal; it was a scheduling quirk with good PR. Building permits, the better forward signal because a permit is a builder betting his own money on next spring, are expected roughly flat. Pending home sales are seen up 0.5% after a 5.4% drop, which counts as a recovery only in the sense that falling down the stairs and then sitting up counts as standing.

Toll Brothers reports Tuesday after the close and Hovnanian Thursday morning, and both will be asked the same question: how much are you paying in incentives to move a house? Builders have been buying down mortgage rates for buyers, which is a discount that never shows up in the sale price and quietly eats the margin instead. Thursday's mortgage reading comes in at around 6.67%, and until that number starts with a 5, the math on the median American house does not work.

Here is a quiet pattern nobody has put on a chart yet. Monday's Empire State factory survey is expected to slip to 10.2 from 15.6. Thursday's Philadelphia Fed index is forecast at 25.3, down from 41.4, a 16-point drop. Friday's flash PMIs, the first fast estimate of business activity, are expected to be lower across the board: composite 53.2 (from 54.5), services 53.9 (from 54.6), manufacturing 53.7 (from 53.9).

Every one of these is a survey. They ask executives whether things got better or worse and count the answers, which makes them fast, useful, and about as emotionally stable as a group chat. None of them measures a dollar. All of them are expected to fall in the same week that four retailers report what actually got sold.

Note the readings are still positive. Above 50 on a PMI and above zero on a factory index both mean growth, so a slowdown from very good to merely good is not a recession; it is a haircut. The problem is that markets trade the direction, not the level, and the direction on every single one of these this week points down.

The whole week is one argument. Surveys on Monday, Thursday and Friday say the mood is cooling. Registers on Tuesday, Wednesday and Thursday say what people actually bought. When those two split, the money is usually right, and the mood is usually early. Your job is to notice which one you are trading.

Ranked by how much of your portfolio it can reach.

Walmart, Thursday morning, before the open.

One company, roughly one in four American grocery dollars, and the single clearest window into whether households are trading down, holding steady, or quietly falling behind. It reports after the Fed minutes, which means Thursday's number lands on a market that has just spent a day arguing about rate hikes and is in no mood for nuance.

Two ways to read it, and be honest with yourself about which one you are hoping for. If Walmart posts solid same-store sales and holds its full-year forecast, last week's 0.6% retail sales drop looks like one bad month, the "consumer is finished" trade gets unwound, and the tape probably likes it. If Walmart guides down, the survey was right, the receipts agree, and every consumer name on your account reprices before lunch.

The honest posture: this is not a print to be a hero in front of. The stock has already run 3.06% over the past week while its sector fell 2.06%, suggesting the good news is partly priced in, and the bad news is not. Let it report. Thursday afternoon is plenty early to have an opinion.

The week's whole argument is between what people say and what people spend, and it gets settled at a checkout counter rather than in a Fed conference room. Monday and Friday, executives fill in surveys about how they feel. Tuesday through Thursday, four companies read out what actually got scanned. One of those is a vibe. The other is a receipt.

Everybody says they've cut back. Thursday morning, Walmart tells us.

Follow us on X and Instagram for daily market updates

Markets don't wait, and neither do we. Follow Tracking the Trade for fast takes, the numbers that matter, and the occasional meme that hits a little too close to your portfolio.