For three days this week the market got exactly what it wanted from the data. On Friday, it got a reminder that cooling inflation and a cooling consumer are often the same story told from different ends.
July retail sales fell 0.6%, the first decline in nine months and the largest in fourteen, against expectations for a modest gain. The University of Michigan’s preliminary August consumer sentiment index dropped to 51.0 from 55.2 in July, well below the 54.5 consensus and reversing two consecutive months of improvement. Stocks slipped from Thursday’s record: the Dow Jones Industrial Average fell 108 points, or 0.20%, to 53,732, the S&P 500 eased 0.2%, and the Nasdaq Composite gave back roughly 0.4%.
The pullback was orderly rather than fearful. The S&P 500 still logged its third consecutive weekly gain — its longest streak since early April — and remains up about 14% on the year. But the character of the week changed on its final day: the disinflation that powered Wednesday and Thursday’s rally now carries a question mark about what is causing it.
Index Performance
Rates, Dollar, Commodities
The Consumer Data: Real Weakness, Real Caveats
The retail sales miss was broad. Headline sales fell 0.6% against consensus for a gain of roughly 0.1–0.2%. Excluding autos, sales dropped 0.3% versus expectations of a 0.2% increase. The control group — the measure that feeds most directly into GDP’s consumer spending component — fell 0.4% against forecasts for a 0.3% rise, the worst reading since January 2025.
There are genuine mitigating factors. Amazon moved Prime Day to June this year from July last year, and most major e-tailers shifted their own promotional calendars to match, pulling sales forward out of the month being measured. Nonstore retailers fell 2.2% as a direct result. Lower gasoline prices and falling car sales also mechanically depressed the dollar figures. And critically, roughly two-thirds of US consumer spending goes to services, which this report largely does not capture.
The less comfortable interpretation is that the generous spring tax refunds that supported first-half spending have simply been spent.
The sentiment survey offered less room for benign readings. Survey director Joanne Hsu noted that sentiment fell about 8% in August, ending two months of improvement, and while views of personal finances declined only modestly, expected business conditions sank 11% for the short run and 17% for the long run. One-year inflation expectations ticked up to 4.3% from 4.2% — still far above the 3.4% that prevailed before the Iran war.
That combination is genuinely awkward for the Fed. Weak confidence with sticky inflation expectations is the configuration that gives neither hawks nor doves a clean argument. As Charles Schwab framed the market’s dilemma: investors want an economy weak enough to tame inflation but strong enough to keep earnings growing, and Friday’s data pushed uncomfortably toward the first without guaranteeing the second.
Oil Rises as Washington Prepares “Unprecedented” Sanctions
Crude reversed higher after Treasury Secretary Scott Bessent said the administration will announce “unprecedented” new economic measures next week aimed at the economic isolation of Iran. The pivot is notable in itself: the administration appears to be shifting from active military pressure toward economic strangulation, a change in tactics that markets are still learning to price.
The rebound in energy prices worked against the disinflation narrative that had driven the week’s rally, and it compounds the consumer problem — elevated fuel costs are precisely what the Michigan survey identified as weighing on household budgets.
Gold traded near $4,376.50, up about 0.6%, extending a run that has it up more than 30% over twelve months. The 10-year Treasury yield rose modestly. The VIX, having touched new 2026 lows below 14.4 this week, settled near 14.5.
Memory Stands Alone
The semiconductor complex was broadly soft, but memory was conspicuously not.
Applied Materials fell about 5.1% after Thursday evening’s report, where solid quarterly results failed to satisfy expectations inflated by a roughly 190% gain over the past year. Yet Micron, SK Hynix, Seagate, and Western Digital all advanced, and SanDisk rose again after Thursday’s double-digit surge, helped by a JPMorgan upgrade to overweight from neutral with a $2,250 price target — implying 47% upside from Thursday’s close. Analyst Harlan Sur argued the company is uniquely positioned to capture the structural inflection in NAND demand driven by AI. SanDisk has risen roughly sixfold in 2026.
The divergence within semiconductors is now the sector’s defining feature. Equipment makers and logic names are being sold on valuation and guidance; memory is being bought on scarcity. Elon Musk crystallized the thesis on X, replying “Few realize this” to Peter Diamandis’s post arguing that memory, not compute, is the rate limiter of the Agentic Era.
That framing — if it holds — explains why memory keeps outperforming even on days the broader chip complex retreats. It is the one part of the AI supply chain where demand visibly exceeds supply, with contracts locked years out and no quick capacity fix available.
The Burry Trade and Its Followers
An uncomfortable side story circulated through the market this week. Michael Burry — who wound down Scion Asset Management at the end of 2025 and now publishes his strategy on Substack for $39 a month to roughly 300,000 subscribers — has held short positions against SOXX, Micron, Nebius, and Palantir. All have surged since late July.
The arithmetic is worth noting: at that subscriber count, the subscription revenue alone runs to roughly $11.7 million a month. Reports have circulated of retail investors who followed the trade into severe losses. The lesson isn’t about whether Burry’s thesis is eventually right — it may well be. It’s that a publicized short position and a leveraged personal position in the same trade are entirely different risk propositions, and the person publishing is not the person facing the margin call.
Around the Market
Reddit surged 12.69% on news it will join the S&P 500, with the change effective for trading on August 18. The stock had been down roughly 31% year-to-date through Thursday, making the inclusion a well-timed reversal catalyst.
Drone manufacturers rallied after President Trump announced plans to impose tariffs on imported drones, a policy that would advantage domestic producers.
Within the Dow, Salesforce (-2.25%), Cisco (-1.67%), and Amgen (-1.28%) led decliners, while Walt Disney (+2.05%), Chevron (+1.16%), and UnitedHealth (+0.61%) supported the index. Valneva jumped 22% after earnings and news that European regulators validated its Lyme disease vaccine application. T-Mobile slipped on a brokerage downgrade citing revenue risks.
Reading the Volume: Complacency, Not Distribution
Total composite volume came in at about 9.6 billion shares, far below the 17.4 billion average of the prior twenty sessions — a striking gap that deserves interpretation rather than alarm.
The mechanical explanations come first. Mid-August is the deepest stretch of the Wall Street vacation calendar, when senior traders are away and institutional desks run thin. Earnings season is essentially finished, removing the single largest driver of repositioning. And the week’s entire data slate — CPI, PPI, jobless claims, retail sales — had been cleared by Friday’s open, leaving nothing left to trade. Schwab had flagged exactly this in its morning note, expecting summer trading patterns to return and keep volume light.
So how should the light volume be read? Conventional technical analysis is reasonably clear that a decline on thin volume is less concerning than a decline on heavy volume: falling prices with institutional participation suggests distribution, while falling prices without it suggests an absence of buyers rather than an abundance of sellers. On that reading, Friday looks like drift, not damage.
The more useful caution is different. Light volume combined with a VIX at 2026 lows and indices at records describes a market in a state of low-conviction calm. That is not bearish in itself — most of the time, calm markets stay calm. But thin liquidity reduces shock absorption. When participation is this sparse, an unexpected headline moves prices further than it otherwise would, because there is less capital standing between the news and the tape. The setup argues for respecting position sizes rather than for reducing exposure.
As for whether the rally since late July is a dead cat bounce: the evidence doesn’t support that framing. The move has been underwritten by an exceptional earnings season, genuinely cooling inflation across three consecutive data points, and September rate-hike odds falling from 50% to 30%. Those are fundamental improvements, not a mechanical rebound. The legitimate concern is narrower — the rally’s leadership has been concentrated, the consumer is now visibly softening, and record highs on record-light volume mean fewer investors have actually participated in the last leg than the index level implies.
The Week Ahead
The economic calendar thins considerably. July’s FOMC minutes arrive, but their informational value is limited — the meeting predates both the weak July payroll report and this week’s benign inflation data, making the minutes something of a historical document.
The real test comes from retail earnings, which will directly interrogate Friday’s consumer scare: Home Depot on the 18th, TJX, Lowe’s, and Target on the 19th, and Walmart on the 20th. If the largest retailers in the country describe a healthy consumer, the Prime Day timing explanation gains credibility. If they don’t, July’s retail sales decline stops looking like a calendar quirk and starts looking like a trend.
Bottom Line
Friday ended a genuinely good week on a cautionary note. Three inflation-friendly data points pushed the S&P 500 to a record and cut September hike odds to 30%; a fourth data point suggested part of the reason inflation is cooling may be that consumers have stopped spending. Those are not contradictory findings, but they lead to very different investment conclusions depending on which dominates.
Memory’s continued outperformance — Musk’s “Few realize this” moment included — remains the market’s clearest conviction trade, and the one place where scarcity is doing the work rather than sentiment. Everything else now waits on Walmart, Target, and Home Depot to tell us whether the American consumer took a break in July or started closing tabs for good.
Leave a Reply