Weekend Update & Market Outlook 8/29/26

Prefer to watch? Here's this weekend's video update.

The Week in Review

The week had two acts, and they pointed in opposite directions.

Act one was Wednesday. July PCE landed at 8:30 with the Q2 GDP revision, and the news was not good for the doves. Core PCE came in at +0.2% month-over-month and 3.3% year-over-year — in line with consensus, but unchanged from June. Headline PCE was worse: +0.2% monthly against +0.1% expected, and 3.7% annual where economists had looked for 3.6%. Personal income doubled expectations at +0.4%; real spending was flat at 0.0%. GDP held at 1.5% on the second print, but the GDP price index rose to 6.4%, the highest in four years.

Remember what the FOMC minutes said two weeks ago: many participants judged that tightening “would likely be necessary if inflation did not decline.” Core PCE at 3.3% is the definition of not declining.

Then Nvidia reported after the close and beat. Data center revenue cleared expectations, and the AI complex rallied Thursday on the relief.

Act two was Friday, and it belonged to Kevin Warsh. In his first Jackson Hole keynote as Chair, he was more hawkish than the market expected. The key line: “While this summer's PCE and CPI readings were better than expected, they do not tell me that underlying trends have meaningfully improved.” He added that the Fed must be confident inflation is moving to target “clearly and at sufficient speed. Otherwise, we have work to do.” And plainly: “The Fed's predominant focus right now should be on prices.” He noted rates show little evidence of being restrictive, and called the breadth of inflation concerning. Chicago's Austan Goolsbee backed him from the symposium, saying a couple of benign months “doesn't feel like out of the woods.”

Warsh also announced Fed task forces to study AI's long-run effects on productivity, labor, token pricing models, and the neutral rate — a genuinely new development that got buried under the rate story.

The repricing was violent at the front end. September hike odds went from 35.4% to 57.5% in a day, with two hikes priced by March. The 2-year jumped 12 basis points to 4.38%, its highest close since January 2025 and the largest post-Jackson-Hole Fed Chair reaction this century. It now sits roughly 73 basis points above the effective fed funds rate — the bond market is openly demanding hikes.

But the long end fell. The 30-year eased 2bps to 5.168% and the 10-year barely moved near 4.70%. That is the week's most instructive detail. A week ago the 30-year was at 5.31%, a 19-year high, and the story was fiscal risk and a market that would not absorb supply. This week the Fed Chair said inflation is the priority and the long bond rallied. BMO's rates strategist called it “a deliberately hawkish speech that will put to rest any concerns about the Fed's willingness to raise rates to restore price stability.” Hawkish credibility at the front end buys lower inflation expectations at the back.

Equities split along exactly the line you would predict. Large-cap growth held up; the Russell 2000 fell 1.51% on the week and 1.37% on Friday alone, breaking below 3,000 for the first time all summer. Higher-for-longer means financing conditions stay restrictive, and that lands hardest on smaller, credit-dependent companies. Two weeks ago small caps were the healthiest part of this rally. Now they are the tell.

Nvidia lost 4.45% Friday despite Wednesday's beat, and Marvell fell 10.3% after beating estimates and raising its outlook. That is now four consecutive weeks in which bellwethers have delivered and been sold.

Weekly Market Scorecard — Week Ending August 28, 2026

QQQ and IWM are derived from their underlying indexes, which are exact: Nasdaq Composite 26,180.46 → 26,402.42 (+0.85%); Russell 2000 3,017.87 → 2,972.37 (-1.51%). Other closes: Dow 53,559.99 (+0.53%, first winning week in three), VIX ~14.4 (down from 15.13), 2-year 4.38%, 10-year ~4.70%, 30-year 5.168%. Month-to-date through Thursday: Dow +3.41%, Nasdaq +2.76%, SPX +2.56% — August finished green despite the volatility.

SPX / SPY

SPX closed at 7,711.76 and reclaimed 7,700, which had been overhead resistance after breaking down through it last week. That is a genuine improvement in structure. Resistance sits at 7,816 first, then the record closing zone near 7,799, then 7,850 and 8,000. Support is 7,650 (Monday's low was 7,652.86), then 7,635, then 7,570–7,580, with the 50-day near 7,500 as the level where the uptrend would take real damage.

Volatility (VIX)

VIX at ~14.4, down from 15.13 and back near the floor of its range. That is complacent positioning going into a payrolls print that decides a live Fed meeting.

QQQ

Around 719.5, up roughly 0.85%, with the Nasdaq Composite at 26,402.42. Tech held up better than small caps because large-cap growth carries durable earnings and cash rather than floating-rate debt. But the AI complex has its own problem: Nvidia beat on Wednesday and still fell 4.45% on Friday.

IWM

Around 295.5, down roughly 1.51%, with the Russell 2000 at 2,972.37 — below 3,000 for the first time all summer. Small caps are the most rate-sensitive corner of the market, and a September hike moving to 58% priced hits floating-rate borrowers directly. Two weeks ago this was the leadership group at +23% year to date.

Next Week's Economic Calendar — August 31 to September 4

Monday August 31 — Chicago PMI (Aug), 9:45 AM ET; Dallas Fed Manufacturing, 10:30 AM ET; monthly auto sales.

Tuesday September 1 — S&P Global Manufacturing PMI final, 9:45 AM ET; ISM Manufacturing PMI (Aug) and JOLTS Job Openings (Jul), both 10:00 AM ET; Construction Spending.

Wednesday September 2 — ADP Employment Change (Aug); EIA crude inventories.

Thursday September 3 — ISM Services PMI (Aug), 10:00 AM ET; initial jobless claims.

Friday September 4 — August Employment Situation: nonfarm payrolls and unemployment rate, 8:30 AM ET.

Key watch: Friday's jobs report, and it isn't close. The September 16 FOMC meeting now carries a 58% probability of a hike, and payrolls is the last major release before it — CPI does not arrive until September 11. Warsh said Friday he is “impressed by the overall performance of the economy, which appears to have strengthened.” A strong August print confirms that read and effectively locks in September. Genuine labor-market weakness is the only argument left against a hike, and after July's -23,000 payrolls with heavy back-revisions, that outcome is not far-fetched. Watch the front end rather than the S&P for the reaction: the 2-year at 4.38% is already 73 basis points above the effective funds rate, so a great deal of hiking is priced, and a soft number would unwind it fast.

Major Earnings — September 1 to 4

Tuesday — Medtronic (MDT) and NIO before the open. After the close: Dell (DELL), Palo Alto Networks (PANW), MongoDB (MDB), GitLab (GTLB), Credo (CRDO).

Wednesday — Five Below (FIVE), NetApp (NTAP), Ollie's Bargain Outlet (OLLI), PVH, Victoria's Secret (VSXY).

Thursday — Broadcom (AVGO) after the close. Also lululemon (LULU), Hewlett Packard Enterprise (HPE), Zscaler (ZS), Samsara (IOT).

Key watch: Broadcom Thursday. It is the last major AI semiconductor report of the cycle and a direct test of whether the pattern holds. Four straight weeks now, bellwethers have beaten and been sold — Cisco on margin, Applied Materials on nothing in particular, Walmart despite a raise, and this week Nvidia and Marvell, with Marvell down 10.3% after raising guidance. If Broadcom beats and sells off too, that is a five-week pattern and it stops being noise. Watch custom-silicon commentary and margins specifically, since memory and supply-chain costs are what killed Cisco.

30-Day Market Outlook

Overall bias: Constructive on large caps, cautious on everything rate-sensitive. The S&P is back above 7,700 and roughly 90 points off its record, August closed green, and the bond market's long end has calmed. But a September hike is now more likely than not, and the market has begun sorting winners from losers on exactly that basis.

Macro narrative: The picture flipped this week in a way that is easy to misread. Last week's story was a bond market in revolt — the 30-year at a 19-year high on fiscal deficits, $40 trillion of debt, and AI-related corporate issuance. The obvious extrapolation was that long rates keep climbing and equity multiples compress.

Friday complicated that. Warsh delivered enough hawkishness that the long end came in while the short end spiked. What that says is that the long-bond selloff was partly about inflation credibility, not purely about supply — and a Fed Chair willing to hike restores some of it. That is, on balance, healthier for equities than the alternative, because long rates are what discount future earnings.

The cost is a genuine growth-versus-financing split. Large-cap growth with durable earnings and cash on hand can absorb a higher policy rate. Small caps carrying floating-rate debt cannot, and the Russell breaking 3,000 says the market is pricing that distinction explicitly. Meanwhile the AI complex has an unrelated problem: four weeks of beats being sold suggests positioning, not fundamentals, is the binding constraint.

Primary risk: A strong August payrolls print Friday. It would confirm Warsh's read, push September odds well past 58%, and drive the 2-year toward 4.50%. Small caps and unprofitable growth take that worst, and with VIX at 14.4 there is essentially no cushion priced for it.

Secondary risk: Broadcom continuing the beat-and-sell pattern Thursday. Four weeks is a trend; five with the AI complex's last major report would suggest the sector's premium is being repriced rather than just consolidating.

What would change the picture: A soft payrolls number. September hike odds are 58%, so there is real room for a dovish surprise to unwind the front end quickly. That is the scenario where small caps snap back hardest, since they have absorbed the most damage from the higher-for-longer repricing.