Week In Review: The Economy Lost 23,000 Jobs. Wall Street Threw a Party.

Week of Monday, August 3 through Friday, August 7, 2026

Had $1,000 riding the Nasdaq from last Friday's close? You have $1,050.93 now. Had it in energy instead? $965.58. One week, one decision, an 8.5-point spread between them, and neither choice had anything to do with how good the companies were.

Friday at 7:30 AM Central, the Bureau of Labor Statistics said US employers cut 23,000 jobs in July. The consensus guess was plus 80,000. That is not a miss; that is a wrong-direction miss, the kind where you check the sign twice before you believe it. It is the first monthly decline since February.

It got worse on the second read. June's number, originally reported as a respectable +57,000, was revised down to +20,000. So the "solid June" everybody built their summer thesis on lost 37,000 jobs, retroactively and quietly, in a footnote.

Then the unemployment rate went down, from 4.2% to 4.1%. That sounds like the good news finally arriving. It is not. The unemployment rate only counts you if you are actively looking for work, so when people give up and stop looking, the rate improves. Participation, the share of adults working or seeking work, fell from 61.5% to 61.4%. The rate improved because the denominator decreased. That is the statistical equivalent of winning a race because everyone else went home.

Wage growth cooled too: average hourly earnings rose just 0.1% on the month and 3.2% over the year, against expectations of 3.5%. For a Fed that has spent three months insisting inflation is the enemy, a cooling wage number is the one piece of evidence it cannot argue with.

And Wall Street stood up and cheered. The S&P added 0.61% on the day. Because a labor market this soft means the Fed cannot raise rates, and for a market that spent all summer terrified of a hike, 23,000 lost jobs read as a hall pass.

Monday and Tuesday delivered 3.25 of the week's 3.51 percentage points on the S&P. The engine was a reported breakthrough on reopening the Strait of Hormuz, the shipping channel that carries a large share of the world's seaborne oil.

Here is the part that matters: as of Friday night, nothing has been signed. A joint Iran-Oman statement is described as "under review." The route being discussed would run for two to four months, and Iranian officials say plainly it does not amount to a full reopening. Iran's semi-official Fars agency reports the proposed terms would bar vessels from the US, Israel, and any nation that has "caused damage" to Iran. Tehran says normalization depends on Washington lifting its blockade of Iranian ports first.

So the market bought a temporary, partial reopening with a guest list. And it bought it before the invitations went out.

Oil did what oil does when it hears the word peace. Crude's ETF proxy USO fell 8.66% on the week, including back-to-back drops of 5.46% and 5.19% on Monday and Tuesday. Energy stocks also fell, and XLE finished dead last among eleven sectors at -3.44%. Somewhere an energy executive is on a call explaining that the fundamentals are unchanged, which is true, and irrelevant because nobody is trading fundamentals this week.

We have now had six straight editions in which some version of "the deal is done" had to be cut before printing. Not because we are contrarians, but because every time we go to verify it, one of the two governments involved has refused to agree to it.

Palantir gained 39.78% in five trading days, including a 29.46% single-day jump on Tuesday after earnings. A thousand dollars in Palantir last Friday is $1,397.77 today. Companies this size are not supposed to move like this. That is a lottery-ticket chart bolted onto a firm with government contracts and an investor day.

It was not alone, and Friday was the giveaway. Atlassian closed +35.31% on a strong revenue forecast. Airbnb finished +17.43%, the best in the S&P. Meanwhile, Trade Desk fell 21.90% after guiding next quarter's revenue nearly 20% below analysts' expectations.

The backdrop is a genuinely excellent earnings season. Of the 440 S&P 500 companies that had reported by Friday, 86% beat estimates, and Q2 profits are tracking up roughly 23% year over year. AI infrastructure is expected to account for close to 60% of the index's earnings growth this quarter. So no, this rally is not purely a story about a fictional peace deal. Real companies posted real numbers and got paid for them.

But look at the payout shape. Beat slightly, gain 35%. Miss slightly, lose 22%. Those are not the reactions of investors calmly repricing cash flows. That is a crowd with too much money and not enough patience, standing over the table waiting to see which way to jump.

Two weeks ago, we wrote that the miners went broke while the guy selling shovels bought a mansion. Then last week it flipped, and the shovel-sellers got sold. This week it flipped back.

The silicon led everything. Nvidia +11.56%, Broadcom +9.89%, Micron +6.63%. Three of the top four slots on our ten-name megacap board went to companies that make the picks and shovels of artificial intelligence rather than the ones digging with them. The buyers cooled: Microsoft held third place but decelerated from last week's +21.75% to +7.59%, while Amazon fell from +17.00% to +1.07%.

And then there was Alphabet, -0.51%, the only red name on a board of ten in a week when the Nasdaq gained more than 5%. It fell 4.03% on Wednesday alone, the session after chief scientist Jeff Dean left the company along with three senior colleagues to start a public benefit corporation called Discovery Loop. In the same 48 hours, Demis Hassabis moved from running DeepMind day to day to a chair role and then to Alphabet chief scientist. Losing one legendary engineer is a headline. Losing four in a week, while reshuffling who runs your AI lab, is a pattern, and the market priced it as one.

Technology finished the week up 7.20%, best of the eleven sectors and more than double the S&P's own gain. Capital spending, meaning the money a company lays out for buildings and machines rather than salaries, is still the only story the market wants to hear. It just changed its mind about who gets to collect it.

For four days, this looked like a completely different story. Kansas City's Jeff Schmid, who holds no vote this year, said Tuesday he does not consider current policy restrictive and that getting inflation to 2% "will require tighter policy." Governor Lisa Cook, who does vote at every meeting, said Wednesday in Anchorage that she is "prepared to act by raising rates, if necessary." Thursday evening, St. Louis's Alberto Musalem added that policymakers "cannot afford to tolerate higher inflation."

That is a coordinated hawkish drumbeat under Chair Kevin Warsh, who has stripped forward guidance out of the Fed's statements entirely, so speeches are now the only signal anyone gets.

Then Friday's payroll print landed and none of it mattered.

Rate futures went from pricing roughly a 57% to 58% chance of a September hike before the report to about 44% after it, according to LSEG data. On CME's FedWatch, the odds the Fed simply holds rates jumped to 60%, up from 45% on Thursday and about one-in-three a week ago. The 10-year Treasury yield, the number that quietly sets your mortgage, fell to 4.651% from 4.740% the prior Friday.

There is a joke in here at our own expense, and it is worth telling. Our Week Ahead on Monday was published under the headline "Everyone Wanted a Cut. Friday Prices the Hike." Friday priced the exact opposite. On Thursday we noted that Wall Street's payroll guesses spanned 65,000 people. With consensus at +80,000 and a spread that wide, the gloomiest forecaster on the street was still calling for job growth. The economy came in at minus 23,000. Not one of them had the direction right, and neither did we.

Read those top three together, and you get the most honest summary of the week available anywhere: no cuts this year at 86%, no move in September at 63%, but a hike sometime in 2026 is still favored at 55%. The market is not betting the danger has passed. It is betting that the danger got postponed. Note also that $48 million is riding on the Fed doing nothing all year, and $66,600 is riding on whether the economy is actually okay. That ratio tells you exactly what this market thinks its job is.

Gold closed Friday at $4,265 an ounce, up 5.28% on the week and 25.8% over the past year. A thousand dollars in gold last Friday is $1,052.79 today, which beats the Nasdaq, and gold does not have an earnings call.

Now the detail almost nobody caught. Practically the entire move happened on Wednesday, a single 5.05% session. Wednesday was the day ADP's private payroll count came in at 44,000, below the 70,000 expected, and the ISM services employment gauge fell to 47.4, below the 50 line that separates growth from contraction. Gold read the labor market correctly forty-eight hours before the government confirmed it. The speculative crowd piled in behind: CFTC net long positioning in gold rose 8.5% on the week to 197,600 contracts.

Grandma's coin jar is having a better year than most hedge funds, and it did it by paying attention to a payroll processor on a Wednesday.

The 30-year mortgage rate rose to 6.69% from 6.66%, and the Mortgage Bankers Association's own 30-year measure climbed to 6.81% from 6.76%. Purchase applications fell to 154 from 159.8.

Sit with that. In the week the bond market decided rate hikes were off the table and the 10-year yield fell nearly nine basis points (a basis point is one hundredth of a percentage point), the rate on an actual house went up. Mortgage pricing runs on a lag and on lender margins, not on Friday's headline, so the relief everyone spent the week celebrating has not yet reached a single closing table. On a $400,000 loan, this week's three-basis-point move costs you about $8 a month. Small, but it is going the wrong way while the stock market throws a parade.

Here is the week's best contradiction. While the Nasdaq gained 5.09%, the professional futures crowd got shorter. CFTC positioning showed Nasdaq 100 speculative net positions flipping to -14,600 contracts from +4,900 the week before, and S&P 500 net shorts deepening to -27,300 from -17,200.

So the tape ripped while the people who trade it for a living leaned the other way. Somebody is very wrong here, and the size of the gap says the resolution will not be gentle.

News sentiment on the S&P itself cooled to 0.40 on Friday from 0.95 on Thursday, even as prices rose, suggesting a press corps writing "stocks rally" and "payrolls collapse" in the same paragraph and not enjoying it. NVIDIA, meanwhile, drew the loudest coverage of any name we track, peaking at 328 stories on Wednesday with sentiment holding near 0.84. Retail is loud and long. The institutions are quiet and hedged. Pick your team carefully.

The appetizer arrived Monday and Tuesday: a peace deal, served raw, unsigned, and absolutely delicious, 3.25 points of S&P gain before anyone asked to see the paperwork.

The main course was Palantir at +39.78%, plated alongside a full spread of AI silicon, with Nvidia and Broadcom carved at the table.

The palate cleanser was Wednesday and Thursday, two flat sessions where the room went quiet and a few people checked their phones.

Dessert was Friday's jobs report, which was on fire when it reached the table, and which everyone ate anyway while insisting the smoke was part of the presentation.

The bill comes Wednesday, and it is written in CPI.

The S&P 500 rose 3.51% this week on three engines. One of them was real: a genuinely strong earnings season where 86% of reporters beat. One of them was imaginary: a Hormuz agreement that six consecutive editions have now failed to verify. And one of them was actively bad news: an economy that stopped creating jobs.

Two out of three is not a rally. It is a bet that the Fed blinks before the damage shows up in anything you can feel.

And that is what makes Wednesday interesting. The July CPI is scheduled for 7:30 AM Central on August 12. This entire week's move rests on the idea that a weak labor market forces the Fed to stand down. A hot inflation print takes that away instantly, and puts the Fed back exactly where it was on Thursday, when three officials in four days were talking openly about raising rates. The jobs report brought the market a reprieve. CPI decides whether it was a pardon or a postponement.

Here is the question that decides your next month: if it takes 23,000 lost jobs to keep the Fed off your back, what exactly are you rooting for?

The family in the kitchen is still cheering. Nobody has looked at the paycheck yet.

Markets don't wait, and neither do we. Get the charts, the jokes, and the occasional existential crisis in real time.

Everyone’s hunting for the next Unicorn.

The type of “category disruptor” that grows fast and turns early believers into big winners.

59,000+ investors think that Mode Mobile could be one of those rare finds.

Americans spend 4 ½ hours on their phones daily, and Mode Mobile is monetizing that screentime. With $1B+ earned by over 490M customers and 32,481% revenue growth, Mode’s EarnPhone is turning smartphones into income generating assets.

Their previous raises sold out, and the company is now offering pre-IPO shares at $0.52/share with up to 20% bonus, exclusive to early investors.

Being early is everything, and this window is still open.

>> Review the offer before it closes

^{*Please read the offering circular and related risks at invest.modemobile.com.}

^{Mode Mobile recently received their ticker reservation with Nasdaq ($MODE), indicating an intent to IPO in the next 24 months. An intent to IPO is no guarantee that an actual IPO will occur.}

^{The Deloitte rankings are based on submitted applications and public company database research, with winners selected based on their fiscal-year revenue growth percentage over a three-year period.}