Saturday, September 26, 2026

The week opened Monday September 21 in the first post-FOMC relief window of any consequence since Warsh's unanimous 12-0 hike to 3.75–4.00% delivered exactly the outcome that seven weeks of analysis had been positioning toward — and within five trading sessions the week had managed simultaneously to be the best single-session for the Nasdaq since early August (Monday's +2.26% to a record 27,122.09, with Intel +12%, AMD +10%, Meta Platforms +11%, and Qualcomm +9% in a single AI-driven session that produced the largest single-day concentration of semiconductor and megacap gains since the early-August relief rally), the most significant diplomatic development of the seven-month Iran war (a three-hour meeting between Special Envoy Steve Witkoff and Iranian Foreign Minister Abbas Araghchi on the margins of the UN General Assembly Tuesday September 22, with Qatar facilitating, the first direct senior-level US-Iran contact of the conflict, producing by September 24 a formal written Iranian roadmap for peace including a 60-day regionwide ceasefire, phased Strait of Hormuz reopening, end to the US naval blockade, halt to Iranian attacks on Arab neighbors, and an agreed negotiating timeline — then revealed by the Wall Street Journal on Saturday September 26 to have been privately rejected by President Trump, who doubts Tehran will meet US demands and believes renewed bombing is likely after the November midterms, a private-public divergence that is the week's defining geopolitical fact in retrospect), the most confounding pair of economic data releases of 2026 (Wednesday September 23's S&P Global Flash PMI — Manufacturing 57.0 versus 53.6 expected, the strongest manufacturing print since May 2022 and the largest single-month beat against consensus in 2026; Services 58.7; Composite 58.4 — arriving on the same morning as Fed Governor Michael Barr's statement that "further policy adjustments are likely to be needed" and the 10-year Treasury yield's decisive break above 5.10% to the 5.12–5.18% range, the highest sustained level since 2007, implying that the PMI data designed to celebrate economic strength instead became the instrument by which yields rose sufficiently to pressure the equity gains Monday's session had accumulated), a Swiss National Bank hold at 0% on Thursday September 24 that confirmed the most pronounced central bank divergence of 2026 (the SNB is the one major central bank that did not hike in the September round — the Fed hiked to 3.75–4.00% on September 16, the BoJ hiked to 1.25% on September 18, and the ECB hiked to 2.50% on its September schedule, while the SNB stood pat with Swiss CPI below 0.8%, producing a 4.00% spread between the US and Swiss policy rates, the widest in modern memory), a Costco Q4 FY2026 beat (EPS $6.75 versus $6.48 estimate, $95.72 billion revenue, +11.2% year-over-year, boosted $0.15/share by IEEPA tariff refunds) that validated the consumer-staples resilience thesis even as University of Michigan Consumer Sentiment finalized September at 48.1 — above the 47.6 consensus but below August's 51.7, erasing August's recovery and sustaining the year-ahead inflation expectations at 4.6%, up sharply from 4.0% in August — the highest since June 2026; Jobless Claims for the week ending September 20 fell to 197,000 (below the 210,000 consensus), confirming the labor market's continued imperviousness to $99+ oil and a federal funds rate now at 3.75–4.00%; and the week's trading configuration — S&P 500 +1.21% (7,650.50 → 7,743.41), Nasdaq +2.06% (26,522.55 → 27,068.72), Dow +0.28% (51,682.64 → 51,828.62), Russell 2000 approximately −1.05% (~2,874 → ~2,844, its fourth consecutive week of losses) — confirmed that AI infrastructure's decoupling from rate-sensitive small caps has now extended from a three-week pattern into a structural divergence: the Nasdaq's second consecutive weekly advance versus the Russell's fourth consecutive weekly loss represents a spread that is not cyclical positioning but a measurable re-rating of which corner of the market rate-hike transmission mechanisms can reach.

Week in Review

The Numbers

*YTD figures carry forward from prior weeks compounded with this week's confirmed move. Weekly % reflects prior Friday September 18 close to Friday September 25 close: S&P 500 from 7,650.50, Nasdaq from 26,522.55, Dow from 51,682.64, Russell 2000 from ~2,874.63. This was a full five-day trading week: Monday September 21 through Friday September 25.

The Story Arc

Monday September 21 delivered the week's defining session in its first hour. AI and semiconductor stocks opened higher across the board in what would become the best single day for the Nasdaq since early August: Intel surged approximately 12%, AMD approximately 10%, Meta Platforms approximately 11%, Qualcomm approximately 9%, in a broad technology-sector rally driven by a combination of factors that were each individually modest but collectively overwhelmed the prior week's post-FOMC selling. Treasury yields declined modestly from Friday's near-5.00% closing levels as the immediate post-FOMC repricing completed; WTI crude retreated slightly from $99.53 as the Trump Iran direct-talks claim from Wednesday September 16 began to be credited for the first time by oil traders who had initially dismissed it. The Nasdaq closed at 27,122.09 (+2.26%), a record closing high; the S&P 500 at 7,764.70 (+1.49%); the Dow at 52,048.83 (+0.71%). Goolsbee delivered the first post-FOMC-blackout speech in London, telling the Official Monetary and Financial Institutions Forum that "my threshold for cutting is a little higher than it was at the last two meetings" and that the road to 2% inflation "may not be painless" — a hawkish first post-blackout statement that, delivered on the same day as a +2.26% Nasdaq session, illustrated the market's decision to price the relief of decision-certainty over the hawkishness of forward guidance.

Tuesday September 22 was the week's geopolitical pivot. At some point during the UN General Assembly opening sessions in New York, Special Envoy Steve Witkoff and Jared Kushner sat down with Iranian Foreign Minister Abbas Araghchi for a three-hour meeting — the longest and most senior direct US-Iran conversation since the seven-month conflict began on February 28, 2026. Qatar facilitated the meeting; Oman's Foreign Minister Badr Albusaidi was also present in New York for parallel consultations. The three-hour duration — compared against the prior week's "direct contact" claim that Iran had refused to confirm — was itself the most significant diplomatic fact: both sides stayed in the room for three hours. Markets received partial news of the meeting during Tuesday's session, with WTI declining. The Richmond Fed Manufacturing Index came in at -2 for September (prior month +4, consensus -2), confirming the regional factory softening signal that August's flat Industrial Production had introduced. The 2-Year Treasury Note auction ($69 billion at 4.787%, bid-to-cover 2.63 versus the 2.60 recent average) priced in-line with no tail — adequate post-hike demand at the new rate level.

Wednesday September 23 was the week's most consequential data day and the session that revealed the week's central tension. At 9:45 AM ET, S&P Global released its preliminary September Flash PMI data: Manufacturing PMI 57.0 (versus 53.6 consensus and 53.9 prior, the strongest manufacturing print since May 2022 and the single largest beat against consensus in 2026), Services PMI 58.7 (versus 56.8 prior month), Composite 58.4 (versus 56.0 prior). Every component was above 55 — the historical threshold for "strong expansion" rather than mere growth. New orders accelerated to the fastest pace in four and a half years. Employment growth was the highest since February 2021. The data was the clearest signal yet that Warsh's two months of rate hikes had not slowed the US economy's output side. The market's response was paradoxically negative: bond yields surged as traders repriced October FOMC probability upward, with the 10-year Treasury yield breaking decisively above 5.10% to trade in the 5.12–5.18% range — the highest sustained close since approximately mid-2007. S&P 500 fell back from Monday's highs. Fed Governor Michael Barr, speaking at the Housing Affordability 2026 Summit in Chicago, confirmed the implication: "in my base case, further policy adjustments are likely to be needed." The Chicago Fed National Activity Index (August) came in modestly negative (prior +0.08 revised), providing the week's one data point that cut against the PMI euphoria: August's composite economic activity modestly contracted while September's forward-looking PMI reached a four-year high, a divergence that is consistent with businesses committing to expansion (PMI measures future orders) while August's completed activity was still compressed under $100 oil and the prior months' rate hike transmission.

Thursday September 24 compressed three significant events into a single session. The Swiss National Bank held its policy rate at 0% — standing entirely apart from the global tightening cycle that had produced three central bank hikes (Fed, BoJ, ECB) in September alone. Swiss CPI is projected at 0.7% for 2026, 0.8% for 2027, and the SNB has now cut rates by 175 basis points cumulatively since March 2024 — it is hiking in the wrong direction relative to the Fed. Costco Wholesale reported Q4 FY2026 results (16 weeks ending August 30): non-GAAP EPS $6.75 versus $6.48 estimate (4.2% beat), revenue $95.72 billion (+11.2% year-over-year), boosted by $0.15/share from IEEPA tariff refunds and by 42.3 million paid executive members (+9.4% year-over-year). The result confirmed that the Costco consumer — household income typically above $100,000, membership-fee committed, bulk-buying oriented — is not showing the distress that UMich's 48.1 headline reading implies. Jobless Claims for the week ending September 20 fell to 197,000 (below the 210,000 consensus, a near multi-month low), maintaining the labor market's resilience signal for the second straight week sub-200,000. Iran formally presented the United States with a written roadmap to end the war: a 60-day regionwide ceasefire, phased reopening of the Strait of Hormuz, end to the US naval blockade, a halt to Iranian attacks on Arab neighbors, and an agreed timeline for substantive negotiations — the most complete Iranian diplomatic proposal since the conflict began.

Friday September 25 closed the week on a recovery note, with the Dow gaining 478.64 points (+0.93%), the S&P 500 +0.51% to 7,743.41, and the Nasdaq +0.50% to 27,068.72. WTI crude declined toward $92.44 — a weekly loss of approximately 7% from $99.53, the largest weekly crude decline since the August Oman-corridor announcement — as Iran's formal ceasefire proposal, combined with news that Saudi Arabia had restarted the East-West pipeline from Abqaiq to Yanbu (though tanker loadings had not yet resumed), provided the clearest peace-premium oil compression since August. University of Michigan Consumer Sentiment finalized September at 48.1 (versus 47.6 consensus, 47.8 preliminary, and 51.7 in August) — year-ahead inflation expectations at 4.6%, up sharply from 4.0% in August — the highest since June 2026, and the current conditions index (50.9) and expectations index (45.8) both remained well below year-ago levels. Darden Restaurants and Paychex reported results without market-moving implications. The week closed with the S&P 500 having recovered its entire post-FOMC loss and added 92.91 points (+1.21%) — but on Saturday September 26, the Wall Street Journal reported that Trump had privately rejected Iran's ceasefire proposal, expressing skepticism that Tehran will comply and privately believing renewed bombing is likely after the midterms. The oil market that had closed Friday at $92.44 (WTI) will open Monday with a significantly different peace-probability priced in than Friday's close implied.

Biggest Movers

Winners

Losers

Market Scoreboard

Weekly Index Performance

Note: S&P 500, Nasdaq, Dow Friday September 25 closes confirmed from TheStreet, CNBC, and Washington Post market data. S&P 500 close 7,743.41 (+0.51% on the day); Dow close 51,828.62 (+478.64 pts, +0.93%); Nasdaq close 27,068.72 (+0.50%). Russell 2000 approximate close ~2,844 based on -1.05% weekly from 2,874.63, provisional. Nikkei September 25 close ~66,364 from Yahoo Finance/Trading Economics; Chuseok holiday confirmed for KOSPI. Hang Seng Thursday/Friday September 25 close ~24,761 from Investing.com.

Monday September 21's Nasdaq +2.26% — the largest single-day gain since early August — and the week's +2.06% closing advance illustrate the AI infrastructure thesis's endurance across the rate cycle's most aggressive phase yet. The immediate catalyst was familiar (post-event relief, sector rotation into beaten-down AI names after the FOMC week's selling) but the magnitude was significant: Intel's +12% was among its strongest single-day performances in several years; AMD's +10% reflected the market's re-rating of AI accelerator demand beyond NVIDIA's lead position; Meta's +11% confirmed that AI-driven advertising infrastructure is the highest-monetized layer of the AI application stack. The Nasdaq's record close (27,122.09 on Monday) arrived exactly five trading sessions after Warsh's unanimously hawkish press conference produced the index's worst two-day run since July. The implied message: the market now separates the AI infrastructure cycle (multi-year order books, rate-insensitive capex commitments) from the rate cycle (each 25-basis-point move reprices equity multiples immediately). Relevant: ai_infra_picks_shovels, ai_mega_ecosystem.

The Russell 2000's fourth consecutive weekly loss — now at approximately 2,844, approximately 7.3% below its August 14 all-time intraday high — is no longer a rate-hike reaction; it is a structural repricing of small-cap capital costs in an economy where the fed funds rate is 3.75–4.00% and moving higher. Barr's Wednesday September 23 confirmation that "further policy adjustments are likely needed" is the operative signal: the dot plot's December hike is not a tail scenario but the base case of at least 14 of 18 FOMC participants. For the 35–40% of Russell 2000 companies with variable-rate debt, the September hike raised interest expense immediately; the October hike (if confirmed at the October 27–28 FOMC) will compound it. At ~2,844, the Russell is approaching the year's lowest sustained support levels, last tested in June 2026. The path to 2,800 requires only the confirmation of October's hike at current CME FedWatch probability levels — no new escalation, no surprise macro weakness, just the base-case Fed path. Relevant: small_cap_value, fomc_announcement.

The 10-year Treasury yield's decisive break above 5.10% — closing the week in the 5.12–5.18% range, approximately 12–18 basis points above Friday September 18's approximately 5.00% close — is the rate market's interpretation of September 23's Flash PMI data, and it is the most structurally consequential number of the week. PMI Manufacturing at 57.0 (versus 53.6 expected) is the bond market's permission slip to price an October hike as the operative scenario rather than the December hike as the next move. The sequence: strong PMI → economy not slowing → Fed can tighten more → October FOMC (October 27–28) probability rises → 10Y reprices against a higher expected path of short rates → 5.12% is the market's updated estimate of where the 10Y should trade given a fed funds rate that may reach 4.25–4.50% by year-end 2026. At 5.12–5.18%, the 10-year is now well above its October 2023 intraday high of 5.008% and approaching levels last sustained in mid-2007. The S&P 500's earnings yield at 7,743 (approximately $400 2026 forward EPS / 7,743 = 5.17%) is now essentially at parity with the 10-year yield — an equity risk premium near zero that has historically resolved either through equity price decline or rapid earnings upgrade. Relevant: bond_duration_trade, yield_curve_inversion.

Commodities & Rates

WTI September 18 prior close $99.53; September 25 close ~$92.44 confirmed from Trading Economics (prior day Sep 25 $92.44/barrel, down 2.29% on the day). Brent September 25 close ~$104.37 (Sep 26 early Brent trading at $104.32 confirmed by Trading Economics). Gold September 25 close approximately $4,285 ($4,284.20/oz confirmed from APMEX Sep 25, 6:08 PM ET). Bitcoin approximately $84,000 (Sep 23 reference $86,214; current market snapshot ~$83,779; provisional ±$1,000). 2Y Treasury ~4.81% (Sep 25 close; prior Friday September 18 close ~4.90–4.95%). 10Y ~5.12–5.18% from FRED/Forbes Advisor September 23; flagged provisional by ±5 bps for Sep 25 close. 30Y ~5.49% (Sep 25 close, +15 bps week). Copper Sep 25 close $6.6955/lb futures settled (Barchart). Uranium ~$90/lb (MetalCharts Sep 26) with long-term contract prices ~$96.50/lb (nominal all-time high September 2026).

The week's most anomalous commodity development is the WTI-Brent divergence: WTI crude fell approximately 7.1% (from $99.53 to ~$92.44) while Brent crude rose approximately 1.1% (from $103.21 to ~$104.37), a roughly 820-basis-point spread between two benchmarks of the same underlying commodity in the same week. The mechanism is structural rather than technical: WTI and Brent normally track within $3–5/barrel of each other, but the Yanbu pipeline attack (September 8–10, 2026) disrupted Saudi Aramco's primary export route for Arabian Light crude — a Brent-grade crude that flows from the Eastern Province to Yanbu on the Red Sea for European and Asian export. Saudi Aramco officially halted crude supplies to European refiners starting October, and Brent-grade supply tightness in Atlantic basin markets kept Brent pricing elevated even as Iran's formal ceasefire proposal on September 24 reduced the generalized geopolitical risk premium that had been sustaining WTI. WTI's larger decline reflects US domestic supply dynamics and forward expectations: if the Strait of Hormuz eventually reopens, US crude exports currently priced at $92 will face increased global supply competition. The WTI-Brent spread widening (WTI now trading approximately $12–13 below Brent, compared to a normalized $3–5 spread) is the oil market's simultaneous expression of two contradictory realities: Brent is tight because the Yanbu bypass is damaged; WTI is falling because peace talks are advancing. Relevant: warflation_hedge, commodity_supercycle.

Gold's approximately -2.3% decline (from ~$4,385 to ~$4,285) continues the precious metal's directionally consistent but sometimes volatile relationship with the Warsh hiking cycle. The three mechanisms compressing gold are each operating at maximum intensity simultaneously this week: the 10Y at 5.12–5.18% provides genuine fixed-income yield competition that makes holding zero-yield bullion more expensive; the PMI Manufacturing at 57.0 removes the recession fear premium that had been providing gold with a macro-uncertainty bid; and Iran's formal ceasefire proposal (regardless of Trump's private rejection, which was not publicly known until Saturday September 26) modestly reduced the geopolitical safe-haven bid. Gold's decline of approximately 2.3% this week brings it to approximately $4,285 — still extraordinarily elevated relative to pre-conflict levels (gold was approximately $5,300–$5,600 in early 2026 before the Iran war's additional premium) but now approximately $450 below the war-premium peak. The practical question is whether the war premium decompresses in a sequence that mirrors the 2022 Black Sea Grain Initiative dynamic (wheat -5% day of signing but -40%+ premium required three specific milestones to clear): gold's full decompression requires Hormuz traffic restoration, Lloyd's reclassification of the Persian Gulf as a standard-risk zone, and US sanctions lifting — none of which is priced for this quarter. Relevant: gold_bug, crisis_alpha.

Bitcoin's approximately +4.5% (from ~$80,400 to ~$84,000) in a week when the 10-year yield rose 16–22 basis points is the week's second most structurally counterintuitive asset move. The conventional playbook: rising yields are negative for speculative assets, and Bitcoin has been negatively correlated with rate moves in 2022–2024. This week's positive correlation between Bitcoin and yields — both rose simultaneously — suggests the market is currently treating Bitcoin as a "strong economy" risk asset rather than a "inflation hedge" store of value. Monday's AI-driven risk-on session pulled Bitcoin higher alongside Intel and AMD; the peace talks optimism that pushed oil lower also pushed risk appetite higher broadly; and the continued spot ETF inflows (partially recovered from the CLARITY Act failure week) provided mechanical demand. Bitcoin at ~$84,000 is approximately 5% above the $80,000 level it recovered after the CLARITY Act failure short squeeze — the question is whether this second $4,000 advance has the same short-squeeze mechanics or reflects genuine incremental institutional demand. Relevant: crypto_ecosystem.

Earnings Recap

The Costco Beat and the Thin-Docket Week

Note: AutoZone, General Mills, Darden, and Paychex provisional; confirmed results from earnings reports not fully reflected in available data at publication time. Costco Q4 FY2026 results confirmed from SEC 8-K filing and Yahoo Finance/Quartz reporting September 24–25, 2026. Micron Q4 FY2026 earnings (fiscal year ending August 2026) reporting September 30 — next week's most significant earnings event. Nike Q1 FY2027 (quarter ended August 2026) reporting approximately October 1.

Costco's $6.75 EPS beat on a week when University of Michigan Consumer Sentiment finalized at 48.1 is the clearest illustration of the bifurcated consumer economy that the week's data presented. A UMich reading of 48.1 is, in any historical context, a recession-zone reading: the index average during the 2001 recession was approximately 87–88; during 2008–2009 it fell to a trough of 55.3 (November 2008); at 48.1, September 2026 is producing consumer sentiment levels below those of the 2020 COVID shock (which bottomed near 71 briefly) and approaching levels seen only during the 2009 financial crisis trough. Yet Costco's Q4 showed 84.1 million paid members, +9.4% executive tier growth, and $95.72 billion in quarterly revenue. The reconciliation: Costco's member base is predominantly households with income above $75,000 — the segment whose primary asset (equity portfolio up +12.9% YTD) has outperformed their primary liability (mortgage originated before 2024 at sub-4%) and whose response to $4/gallon gas is to buy in 55-gallon drums at lower per-unit cost. The segment that is driving UMich's 48.1 reading is the bottom 40% of income distribution — uninsured against oil price shocks, no stock portfolio to offset the income erosion, mortgage refinancing at 7%+ if they need to move. Costco beats and UMich cracks because they measure different Americas. Relevant: global_consumer_staples, k_shape_economy.

The week's most consequential non-reporting event was Micron's September 30 fiscal Q4 announcement — the earnings catalyst that will either confirm or complicate the AI memory supercycle thesis that Applied Materials' record quarter established the prior week. Micron's pre-announced guidance for fiscal Q4 is extraordinary: revenue $50.0 billion (±$1 billion), non-GAAP gross margin approximately 86%, non-GAAP EPS approximately $31.00. If confirmed, fiscal Q4 revenue would represent approximately 342% year-over-year growth and gross margins that approach software-like levels for a capital-intensive memory manufacturer. The driver is HBM4 (high-bandwidth memory): Micron's HBM4 shipments for NVIDIA's Vera Rubin platform are fully booked for 2026, DRAM spot prices have risen approximately 52% year-to-date, and the AI data center supercycle is converting standard DRAM manufacturing capacity into premium HBM allocation. The September 30 report will provide the most complete picture of AI infrastructure demand at the memory layer — one level below the GPU (NVIDIA) and semiconductor equipment (Applied Materials, which reported record Q3 FY2026 results on August 13, 2026) layers that have already reported record results. Relevant: semiconductor_value, nvidia_supply_chain.

Geopolitical Update

UNGA Week: The Three-Hour Meeting, the Written Roadmap, and the Private Rejection

The week of September 21–26 was the most diplomatically active of the seven-month Iran war — and closed with the deepest disconnect yet between public and private US positioning.

On Tuesday September 22, Special Envoy Steve Witkoff and Jared Kushner met with Iranian Foreign Minister Abbas Araghchi for three hours on the margins of the UN General Assembly in New York, with Qatar facilitating. This was the first direct US-Iran senior-level meeting since the February 28 conflict began. The three-hour duration — compared against Trump's September 16 "direct contact" claim that Iran had refused to acknowledge — was itself the most significant diplomatic signal: both sides stayed in the room through three hours without walkout or public breakdown. Oman's Foreign Minister Badr Albusaidi, the traditional Iran-US back-channel intermediary, was also present in New York for parallel consultations with Omani officials.

On September 23, President Trump addressed the UN General Assembly in the most binary framing of the war's diplomatic arc: the United States offered Iran two paths — "rebuild through agreement" or face "annihilation" and be driven "into hell." The address simultaneously established Trump's negotiating ceiling and his negotiating floor. Following the speech, Trump told reporters that the three-hour meeting with Iran was "productive" and predicted "oil and other prices could fall if the war ends" — the most direct Trump linkage between Iran diplomacy and domestic US inflation of the conflict's duration.

On September 24, Iran formally presented the US with a written roadmap: a 60-day regionwide ceasefire; a phased reopening of the Strait of Hormuz; an end to the US naval blockade of Iranian ports; a halt to Iranian-backed attacks on Arab neighbors (including Houthi strikes on Saudi Arabia); and an agreed timeline for substantive nuclear and regional security negotiations. Iranian Foreign Minister Araghchi described the proposal as "comprehensive" and said Iran's nuclear program would be on the negotiating table.

On September 25, Araghchi revealed to journalists that the proposal included a one-week immediate ceasefire to reopen the Strait of Hormuz and allow further talks. Iranian President Pezeshkian's UN address had framed Iranians as "victims of terrorism" by the US while calling for reviving diplomatic frameworks before the November midterm elections.

The Wall Street Journal reported on Saturday September 26 that Trump has privately rejected Iran's ceasefire proposals and told advisors he doubts Tehran will meet US demands. Trump's private assessment, per the Journal: renewed bombing is likely after the November midterms. The public diplomatic theater of the week — the three-hour meeting, the "productive" characterization, the acknowledgment that prices "could fall" — was, on Saturday's evidence, a midterm management exercise rather than a genuine peace signal. The oil market that priced WTI at $92.44 on Friday's close will open Monday with this information fully incorporated.

Yanbu Pipeline: Restart Without Resumption

Saudi Aramco officially restarted the East-West pipeline from Abqaiq to Yanbu on September 22 — less than two weeks after the September 10 Houthi attack destroyed pumping stations across two regions of Saudi Arabia. The restart announcement was the week's most important supply-side oil development, and it partially explains WTI's decline toward $92.

The critical caveat: tanker loadings from Yanbu had not resumed as of Saturday September 26 despite the pipeline restart. Saudi Aramco has been hunting for fuel supply in the Mediterranean after the attack and had halted crude supplies to European refiners effective October. The pipeline is operating at reduced capacity (targeting approximately 4 million barrels/day versus the pre-attack 4.5–5.5 million bpd), and Saudi Aramco is building a bypass for full repairs estimated at 4–6 weeks. Saudi Arabia intercepted two Houthi-launched drones headed for Riyadh on September 25 and two ballistic missiles and two additional drones on September 26. The military chiefs of Saudi Arabia, Pakistan, and Turkey met in Riyadh on September 25 under the Mecca Joint Defence Agreement. France confirmed the deployment of troops to guard the Yanbu oil terminal.

The Yanbu situation is the structural reason Brent has not declined despite WTI's -7% week: Brent-priced crude from the Arabian Peninsula now has a damaged bypass route, and the Hormuz Strait remains at approximately 1% of pre-conflict traffic levels (one transit recorded September 20 versus the 85-per-day pre-conflict baseline). The peace talks reduce the Hormuz risk premium; the Yanbu damage maintains the Brent supply premium.

UN Week: Netanyahu, Iraq, and the Regional Architecture

Israeli Prime Minister Netanyahu delivered a combative UN address on September 24, defending Israel's February nuclear strikes on Iran as "one of the easiest decisions" and framing Israel as engaged in a "seven-front war." Netanyahu attacked Qatar and Turkey over alleged influence campaigns in the US and Europe; Qatar rejected the accusations. The Israeli UN posture confirmed that any Iran peace framework must account for Israeli security requirements that are not part of the current US-Iran direct talks — a negotiating complexity that did not exist in the August Oman corridor attempt.

Iraqi Prime Minister al-Zaidi proposed a revised US military withdrawal timeline: completion by September 30, followed by a 90-day freeze, weapons handover beginning January 2027, and final deadline June 2027. Two hundred Ukrainian military specialists deployed to Qatar, Saudi Arabia, and the UAE to assist with air defense coordination — confirming that the Iran war's theater has expanded to include European-trained air defense expertise in the Gulf.

The UN General Assembly's Security Council scheduled a 90-day briefing on the 1737 Sanctions Committee's work on Iran, following the UN fact-finding mission's September 17 finding of grounds to believe both US war crimes and Iranian gross human rights violations (221 children killed, mass arrests, expedited executions during protest crackdowns).

Strategy Scorecard

Winners

Mixed

Losers

MVP of the Week

ai_infra_picks_shovels — for the Monday September 21 single-session validation of the AI infrastructure thesis at maximum scale: Intel +12%, AMD +10%, Meta +11%, Qualcomm +9%, and Nasdaq record close 27,122.09, arriving in the first trading session after Warsh's unanimously hawkish press conference, confirming that the market is now pricing AI infrastructure capex as a separate and non-overlapping economic cycle from the rate-hike transmission cycle. The mechanism that produces a Nasdaq record in the same week that the 10-year yield breaks above 5.10% is the same mechanism that produced Applied Materials' record Q3 FY2026 (announced August 13, 2026): AI infrastructure companies operate on order books that are committed 2–5 years forward by hyperscalers' capital expenditure cycles, and those commitments do not get repriced by a single 25-basis-point FOMC decision. Intel's +12% in a week when the 10-year yield rose 12–18 basis points is the week's most concentrated expression of the AI decoupling thesis — a company whose foundry business is receiving multi-billion-dollar AI-related orders from TSMC competitors and government-sponsored programs has a demand curve that is immune to short-term rate fluctuations. The Nasdaq's +2.06% weekly advance against the Russell's fourth consecutive weekly decline (now -1.05%) is the durable version of the rate-cycle divergence that has defined 2026's second half: not a week-to-week oscillation but a structural re-rating that compounds with each additional 25-basis-point move.

Next Week Preview: September 28 – October 3, 2026

Economic Calendar

Earnings (Key)

Political / Central Bank

Geopolitical Watchlist

• Trump's Iran Private Rejection: The Wall Street Journal's Saturday September 26 report that Trump privately rejected Iran's formal ceasefire proposal is the week's most market-moving unreleased information. Watch for: (a) any official US response to Iran's September 24 written proposal — the absence of response is itself a signal; (b) whether Trump issues any public statement directly contradicting the WSJ characterization (as he did with prior leak reports about policy positions); (c) Brent crude's Monday open — if the WSJ report is believed, Brent should reverse Friday's decline toward $103–105; (d) Araghchi's response to the US non-response — if Iran withdraws the proposal, the peace trade is over; if Iran extends the proposal's validity, the negotiating clock resets.
• Yanbu Tanker Loading Resumption: Saudi Aramco restarted the East-West pipeline (September 22) but tanker loadings from Yanbu had not resumed as of September 26. The first Yanbu cargo loading after the pipeline restart is the physical event that begins closing the WTI-Brent spread (currently ~$12–13, versus normalized $3–5). Watch for Saudi Aramco operational updates and Kpler tanker tracking data for Yanbu terminal activity.
• Houthi Campaign Continuation: Saudi Arabia intercepted ballistic missiles and drones on September 25–26. The Saudi-Pakistan-Turkey military chief meeting (September 25, Mecca Joint Defence Agreement) signals Gulf defense architecture hardening regardless of US-Iran talks. France deployed troops to Yanbu; watch for any NATO member deployments that change the conflict's escalation calculus.
• Micron September 30 and October FOMC: These two events — one corporate, one monetary — will dominate market narrative from September 30 through October 2 (NFP). If Micron beats guidance by confirming the AI memory supercycle, AI infrastructure trades extend; if the PCE is hot and NFP is strong, October becomes a live FOMC meeting that the dot plot had not explicitly signaled as such.

Monday Setup (September 28 Open)

Scenario A: Trump confirms Iran rejection publicly; PCE above expectations (~35% probability)

Trump issues a statement before Monday's open (or over the weekend) that the US has not accepted Iran's ceasefire proposal and that conditions for negotiation have not been met; Brent spikes toward $107–110 on Monday open as Friday's -7% WTI move partially reverses. August PCE (Tuesday September 30) prints core above 0.3% monthly as services inflation remains sticky under $100+ oil. October FOMC probability rises to 65–70% by Friday. Action: restore warflation_hedge to 13–15%; add crisis_alpha; reduce oil_down_tech_up to 30% weight (oil up thesis dominates over tech-up); reduce crypto_ecosystem to 70% weight (risk-off on oil spike); maintain fomc_announcement hawkish positioning at full weight; raise cash to 20–22%.

Scenario B: US-Iran talks continue without resolution; PCE in-line; Micron beats (~45% probability)

The weekend and Monday pass without a public US statement on Iran's proposal; the "private rejection" remains a WSJ characterization rather than an operational fact; Brent consolidates at $104–107; WTI partially recovers toward $94–96. August PCE core 0.2–0.3% monthly (in-line), headline PCE modestly below expectations as July energy prices (Brent averaging ~$88–92 in July before the August escalation) reduce the energy component; October FOMC probability stays at 55–60%. Micron's September 30 earnings confirm guidance (revenue $50B, EPS ~$31), validating AI memory supercycle. Action: maintain warflation_hedge at 12–14% (restore from prior week's trim given Saturday WSJ); hold fomc_announcement hawkish positioning through October FOMC; add to semiconductor_value on Micron beat confirmation; hold ai_infra_picks_shovels at full weight; keep small_cap_value at 20–25% weight until October FOMC probability falls below 35%.

Scenario C: Iran peace framework advances; PCE soft; Micron guides above (~20% probability)

Oman Foreign Minister Albusaidi confirms a technical ceasefire framework taking effect within 72 hours; Iran and US agree on a one-week Hormuz opening as a confidence-building measure; Brent falls toward $97–100. August PCE core below 0.2% monthly (energy drag, services moderation); October FOMC probability falls to 35–40% as disinflation signal overrides PMI 57.0's hawkish impulse. Micron reports above guidance and guides Q1 FY2027 above $55 billion. Action: restore small_cap_value to 45–50% on rate relief; trim warflation_hedge to 8–9% (operational framework is worth more oil premium compression than a statement); restore gold_bug to 13% (dollar weakens on oil decline and reduced hike probability); S&P 500 target 7,850–7,950; add v_shape_recovery positioning.

Position Sizing

• Warflation hedge (warflation_hedge): restore to 12–14% from prior week's 10–12% on Saturday WSJ Trump-private-rejection reporting; the formal Iranian proposal (the most complete of the conflict) has been privately rejected, which means the war is closer to the Trump-expects-post-midterm-resumption scenario than the market priced on Friday; do NOT exit — Yanbu tanker loadings not resumed, Hormuz at 1% baseline, and Trump's post-midterm bombing scenario is now the base case per WSJ
• Rate positioning (fomc_announcement): hold maximum hawkish positioning through October 27–28; PMI 57.0 + Barr + Goolsbee in one week = the October hike is now a live scenario at >55% probability; the September 30 PCE will set the terminal probability before the blackout begins
• Gold (gold_bug): maintain 10–11%; the Saturday WSJ Iran rejection is a modest gold tailwind (geopolitical risk restored); restore to 13% if Trump publicly rejects Iran proposal and Brent spikes above $108
• Small caps (small_cap_value): reduce to 20–25% (from 30–35% prior week); fourth consecutive weekly loss, PMI 57.0 confirmed October as live, 10Y at 5.12–5.18%; restore to 40% only on confirmed PCE core below 0.2% monthly with October CME FedWatch falling below 35%
• AI infrastructure (ai_infra_picks_shovels): hold at full weight; Intel +12%, Nasdaq record — the AI capex confirmation from Monday's session and PMI 57.0 both confirm the demand environment; Micron September 30 is the next data point
• Cash: 17–20%; raise toward 22% in Scenario A; deploy 5% of cash into small caps on confirmed October hold signal

---

The 57–48 Paradox — When the Factory Floor and the Kitchen Table See Different Economies

On Wednesday September 23, 2026, two gauges of the American economy published their September readings within hours of each other and pointed in exactly opposite directions. S&P Global's Flash PMI Manufacturing index hit 57.0 — the strongest manufacturing reading since May 2022, a full 3.4 points above consensus, and the largest single-month beat against expectation in 2026. The same morning, Federal Reserve Governor Michael Barr confirmed at a housing summit in Chicago that "further policy adjustments are likely to be needed," citing an economy that remains too strong for the Fed's inflation target. On Friday September 25, University of Michigan released its final September Consumer Sentiment reading: 48.1. The gap between these two numbers — 57.0 for manufacturing activity, 48.1 for consumer sentiment — is not a rounding error or a measurement coincidence. It is the most precisely quantified expression of the two-speed American economy that 2026 has produced, and understanding why they are diverging, which one is right about where the economy is heading, and what the divergence means for every major asset class is the most important analytical exercise of the current market environment.

The measurements themselves

Before explaining the divergence, it is worth being precise about what each number measures, because imprecision here generates false interpretations.

The S&P Global Flash PMI Manufacturing Index is a diffusion index derived from monthly surveys of purchasing managers at approximately 800 manufacturing companies across the United States. The specific question each respondent answers is not "how did your factory do this month" but "how do you expect conditions to be next month relative to this month" — in other words, PMI is a forward-looking signal from the people who place orders. A reading above 50 indicates that more purchasing managers are expecting improvement than deterioration; a reading of 57.0 means that the margin of those expecting improvement is the widest since May 2022. The PMI's composition in September 2026 included Manufacturing at 57.0 (new orders accelerating at the fastest pace in 4.5 years), Services at 58.7 (fastest services improvement since 2021), and Composite at 58.4. These readings describe a business sector that is placing the largest volume of forward orders in years.

The University of Michigan Consumer Sentiment Index is a diffusion index derived from monthly surveys of approximately 500 US households, asking respondents to rate their personal financial situation (current and expected), their views on business conditions (current and five-year ahead), and their assessment of buying conditions for durables, homes, and vehicles. A reading of 48.1 means that the household sector is deeply pessimistic: more respondents are describing their financial situation as worse than better, are pessimistic about future business conditions, and believe that the current moment is a bad time to make major purchases. In September 2026, the Current Economic Conditions Index was 50.9 and the Consumer Expectations Index was 45.8 — meaning the forward-looking component of household sentiment was more pessimistic than the current assessment, a configuration that historically precedes spending reductions.

The critical asymmetry: PMI purchasing managers are primarily employed at large and mid-sized manufacturing companies, are forecasting their companies' future orders, and are insulated from consumer-facing price pressures by salary and benefits packages. UMich respondents are a representative sample of all US households, including the 60% of Americans who hold no equity assets, including the 43 million households who are net gasoline consumers spending approximately 7–12% of post-tax income on transportation fuel at $4+ per gallon, and including the 65+ million homeowners whose refinancing cost has risen from sub-3% to 7%+ for any new mortgage.

The gap's historical rarity

The simultaneous combination of Manufacturing PMI above 55 and UMich Consumer Sentiment below 50 is historically rare in a non-recessionary environment. Running the data backward through PMI and UMich records since 1990, the combination of PMI above 55 with sharply declining UMich has appeared in approximately three prior episodes, though the current 2026 episode — with UMich clearly below 50 — is the first clearly confirmed instance in modern data of PMI above 55 and UMich simultaneously below 50:

2005–2006: The housing-gas paradox. During the height of the housing boom, manufacturing PMI readings regularly exceeded 55 (driven by residential construction equipment, building materials, and consumer goods demand). UMich plunged sharply to 76.9 in September 2005 (from 89.1 in August — the largest single-month drop since 9/11) after Hurricane Katrina drove gas prices above $3/gallon nationally — a record at the time. Purchasing managers at manufacturing companies were experiencing the housing boom's order wave; households were experiencing $50 fill-ups for the first time. The resolution: UMich recovered above 80 by mid-2006 as gas prices normalized and housing wealth effects reasserted themselves. The economy was not in recession; manufacturing led, and consumer sentiment eventually followed.

2011–2012: The euro-crisis-US-recovery split. US Manufacturing PMI exceeded 55 in early 2011 as the domestic economy recovered from 2009, but UMich fell to a trough of 55.8 in August 2011 during the euro zone sovereign debt crisis (when S&P downgraded US debt on August 5, 2011, and the Dow fell approximately 6% in the week of the downgrade, including a 634-point single-day drop on August 8; a 10% cumulative decline in the S&P 500 took roughly 8 weeks). US purchasing managers were ordering based on domestic recovery; US households were watching European bank contagion risk on CNBC and reducing major purchases. The resolution: the ECB's Draghi "whatever it takes" (July 2012) resolved the euro crisis, UMich recovered, and manufacturing PMI normalized downward as the expansion matured.

2022: The post-lockdown supply chain phase. PMI remained above 55 in early 2022 as supply chain normalization produced order surges, while UMich fell to 50.0 in June 2022 (its lowest reading since 2008) as CPI hit 9.1%. This episode resolved in the negative direction: manufacturing PMI fell below 50 by late 2022 as rate hikes eventually slowed orders, and UMich eventually recovered as inflation decelerated in 2023. In the 2022 case, UMich was right about the economy's direction — manufacturing PMI was too optimistic about the durability of the post-lockdown order surge.

The current 2026 episode — PMI at 57.0, UMich at 48.1 — is the only instance in this series in which UMich is clearly below 50 simultaneously with PMI above 55 (the prior episodes saw UMich decline sharply but not below 50). The spread between the two indicators is the widest of any of the episodes reviewed, and the cause is categorically different from each prior episode.

Why they are diverging in 2026

The 2026 divergence is not driven by one shock but by three simultaneous structural forces that affect purchasing managers and consumer households differently.

Force 1: AI-driven manufacturing demand versus consumer energy costs. Manufacturing PMI's 57.0 reading is predominantly driven by AI infrastructure orders: data center construction equipment, semiconductor manufacturing equipment, power grid expansion components, network hardware, and the entire supply chain of industrial goods that feed into hyperscaler capital expenditure programs. Meta's $125–145 billion 2026 AI capex plan, Microsoft's $190 billion, Amazon's $200 billion — these commitments flow directly into manufacturing order books across hundreds of supplier categories. A purchasing manager at a industrial motor company whose largest new customer is a data center cooling equipment buyer is experiencing PMI 57.0 in real-time. Meanwhile, a US household paying $4.47/gallon for gasoline (derived from Brent crude above $100, itself a direct consequence of the Iran war's Hormuz blockade) is experiencing a persistent energy tax that the PMI does not measure. The AI capex cycle and the oil price inflation are occurring simultaneously, but they fall on completely non-overlapping populations.

Force 2: The rate transmission asymmetry. The Federal Reserve's rate hikes at 3.75–4.00% affect purchasing managers and households through entirely different channels. A manufacturing purchasing manager whose company's AI-infrastructure customers have pre-committed capital expenditure contracts through 2028–2029 does not reduce orders because the 10-year yield rose to 5.12%; their buyers' capital allocation decisions were made before the September 16 hike and will not be revisited until the next annual budget cycle. A household whose mortgage rate is 7.2% (because their fixed rate is up for renewal, or they are trying to buy a first home), whose auto loan is 8.5%, and whose credit card carries a variable rate that rose approximately 500+ basis points from early 2022 through mid-2023 is experiencing the rate hikes at the individual-transaction layer immediately. The PMI does not feel the hike; the household feels every basis point.

Force 3: The wealth distribution effect. The S&P 500 is up +12.9% year-to-date, which means the median stock-owning US household's equity portfolio has gained approximately $6,000–$7,000 in 2026 alone (the $30,000–$60,000 range applies to upper-quartile investors with $230,000–$465,000 invested). Approximately 60% of US households own some equities; approximately 10% of households own enough equities that the 2026 YTD gain constitutes a meaningful fraction of annual income. The purchasing managers who contribute to PMI disproportionately belong to the households with equity wealth — they are employed at manufacturing companies that are themselves contributing to the S&P 500's gains. The 40% of US households with no equity ownership — and the 50% who have less than $10,000 in equities — experience 2026 as a year of $4/gallon gas, 7%+ mortgage rates, grocery price increases, and no compensating asset appreciation. UMich's 48.1 is an average: the top-equity-owning households are pulling it toward 60–65; the bottom-no-equity households are pulling it toward 35–40. The reported 48.1 is not a monolithic reading but an averaging of two populations who are living in demonstrably different economic realities.

The historical question: which indicator is correct?

The fundamental analytical question is whether a Manufacturing PMI at 57.0 signals an economy that is too strong to produce the demand destruction UMich's 48.1 warns of, or whether UMich's 48.1 signals a household sector whose spending weakness will eventually reach the factory orders that the PMI is celebrating.

The academic literature on the lead-lag relationship between PMI and consumer sentiment provides a partial answer. In normal expansion cycles — non-oil, non-war, non-supply-shock expansions — Manufacturing PMI tends to lead UMich by approximately 2–4 months, because business orders reflect management's forward expectations, and those expectations eventually transmit to employment (which households experience as income security) and consumer confidence. In 2003–2007, PMI recoveries consistently preceded UMich recoveries by 2–3 quarters.

But the current episode is not a normal expansion cycle. It is a supply-side geopolitical shock (Iran war → Hormuz blockade → $100 Brent) superimposed on a technology-demand boom (AI → hyperscaler capex → manufacturing order surge). In supply-side energy shock episodes, the lead-lag relationship breaks down: PMI can stay elevated indefinitely on AI demand while UMich stays depressed indefinitely on oil prices, because the causal chain from PMI to UMich (business orders → employment → income → spending → household confidence) requires labor market strength to translate — but the labor market has already been fully employed since 2021, and the current round of confidence deterioration is driven not by unemployment risk but by real-income compression from energy and food prices that are immune to further employment growth. You cannot employ your way out of a $4/gallon gasoline problem.

The 2005–2006 precedent (where PMI led UMich's recovery) resolved because gas prices normalized within 12 months of the Katrina shock. The 2022 precedent (where UMich led PMI lower) resolved because the order surge underlying the PMI was driven by the same inflationary impulse that was destroying UMich sentiment — when CPI fell, both PMI and UMich moved together. The 2026 precedent is categorically different because PMI is driven by a technology investment cycle (AI capex) that operates independently of oil prices, while UMich is driven by oil prices that operate independently of the technology investment cycle. They can coexist in this configuration for as long as Hormuz remains at 1% of baseline traffic and AI hyperscaler capex commitments remain funded.

The Fed's impossible position

The 57.0–48.1 divergence places the Federal Reserve in the most structurally difficult position of Warsh's chairmanship. PMI Manufacturing at 57.0 provides perfect macroeconomic ammunition to hike in October: the economy is expanding at the fastest manufacturing pace in 4.5 years, employment is rising, new orders are accelerating. There is no economic justification in the PMI data for a pause. Governor Barr explicitly cited the PMI data in his Wednesday September 23 speech as supporting "further adjustments."

But UMich at 48.1 — sustained at near-crisis levels for the third consecutive reading — contains within it a leading indicator of consumer spending reduction that the PMI does not capture. Consumer sentiment at 48.1 has historically preceded consumer spending slowdowns by 2–4 quarters. The Consumer Expectations Index at 45.8 is particularly ominous: households do not expect their financial situation to improve, and expectations-driven sentiment weakness has historically produced self-fulfilling spending reductions as households pre-cut discretionary expenditure in anticipation of further deterioration.

Warsh is hiking into a split economy: one half is roaring (PMI 57.0), one half is fracturing (UMich 48.1). The rate hike compresses the half that is already fracturing (it makes mortgages, auto loans, and credit card rates more expensive) without touching the half that is roaring (AI capex order books are denominated in long-dated contracts, not floating rates). The Fed is the wrong instrument for this divergence — supply-side oil inflation from a foreign military conflict and AI demand-pull inflation from hyperscaler capex require different policy responses, but the Fed only has one tool.

The practical implication: the gap closes, but slowly and painfully

The 57.0–48.1 gap will eventually close. The historical record is clear that sustained divergences between PMI and consumer sentiment resolve; the uncertainty is the direction. Two resolution paths exist:

Path A — PMI falls to meet UMich: Consumer spending declines (UMich 48.1's warning materializes), household retail sales weaken, AI hyperscaler quarterly earnings begin to show revenue growth moderation, and the order book commitments that are driving PMI to 57.0 get deferred into later quarters. This is the 2022 resolution path — the strong PMI was temporary; sentiment was right. This path implies an S&P 500 correction.

Path B — UMich rises to meet PMI: The Strait of Hormuz reopens (Iran deal), Brent falls from $100+ to $70–80, gasoline prices at the pump return to $3/gallon, household real income improves, and consumer confidence recovers as the energy tax is lifted. PMI stays elevated on AI demand; sentiment recovers on oil relief. This is the 2005–2006 resolution path. This path implies a significant broad-market rally.

The Saturday September 26 WSJ report that Trump privately rejected Iran's ceasefire proposal raises the probability of Path A (PMI falls to meet UMich) and lowers the probability of Path B (UMich rises to meet PMI). If the war continues through the midterms and Trump's post-midterm bombing scenario materializes, the $100+ Brent environment extends indefinitely, UMich stays at 48–50, and the rate hikes required to fight the resulting inflation eventually reach the AI capex cycle. Path B requires a Hormuz deal; Path A only requires patience.

The 57.0–48.1 gap is not a prediction. It is the market's unresolved question, quantified. Every week it persists is a week in which the factory floor and the kitchen table are operating in separate economies — and every basis point of additional tightening is pressing harder on the kitchen table than the factory floor. Relevant: economic_indicators, fomc_announcement, k_shape_economy, warflation_hedge.

---

Sources:
- CNBC — Stock market news for Sept. 21, 2026
- CNBC — Stock market news for Sept. 25, 2026
- TheStreet — Stock Market Today Sept. 21, 2026
- TheStreet — Stock Market Today Sept. 25, 2026
- Washington Post — US stocks gain ground toward finish of a winning week
- Washington Post — How major US stock indexes fared Monday 9/21/2026
- Bloomberg — Fed's Goolsbee Says Road to 2% Inflation May Not Be Painless
- Washington Post — Federal Reserve official says inflation persistence led her to support interest rate hike
- Yahoo Finance — Fed's Michael Barr backs more rate hikes as PMI hits 5-year high
- Kitco News — Fed's Barr says further rate hikes will likely be needed
- FXStreet — Breaking: US S&P Global Composite PMI came in at 58.4 in September
- TradingView — Chicago Fed Activity Index Down in August
- CryptoBriefing — Richmond Fed manufacturing index slips to -2 in September as orders and shipments retreat
- Helious — 2-Year Treasury Note Auction Results September 22, 2026
- US News — Claims for Unemployment Benefits Drop to 197,000, the Lowest Since Mid-July
- CNBC — Swiss National Bank holds rates at 0% as inflation stays low
- RTE — Swiss National Bank holds rates - as expected
- GuruFocus — Swiss National Bank Holds Rates Steady Amid Global Tightening
- MarketScreener — September Final University of Michigan Consumer Sentiment 48.1
- Quartz — University of Michigan consumer sentiment index, September 2026
- Yahoo Finance — Costco Q4 2026 earnings beat expectations on tariff refund
- Quartz — Costco Q4 2026 earnings beat expectations on tariff refund
- SEC — COSTCO WHOLESALE CORP 8-K Filing Q4 FY2026
- Al Jazeera — Iran war live: Witkoff, Iran's FM Araghchi meet in New York
- Washington Post — Iran says it has given Trump administration new proposal for peace talks
- The National — Trump reportedly rejects Iranian ceasefire proposal as conflicting signals persist
- NBC News — Trump tells U.N. he could 'annihilate' Iran but expects a deal after the midterm elections
- Washington Times — Trump says U.S. and Iranian officials met Tuesday, shortly after warning he may 'annihilate' Iran
- Eurasia Review — Trump says U.S. and Iranian officials held three-hour talks at UN
- Straits.live — Strait of Hormuz Status: September 26, 2026
- GlobalSecurity.org — Iran War 2026, Day 209 Update, September 24, 2026
- Wikipedia — 2026 Strait of Hormuz crisis
- US News — Saudi Arabia Restarts East-West Oil Pipeline, to Resume Exports from Yanbu
- Baird Maritime — Yanbu tanker loadings yet to resume despite Saudi pipeline restart
- Kpler — Saudi Yanbu Outage Shakes Crude, Freight, LNG Markets
- Bloomberg — Saudi Aramco Halts Crude Oil Supply to European Refiners After Pipeline Attack
- US News — Saudi Arabia Says It Intercepts Another Houthi-Launched Missile
- Al Jazeera — Saudi Arabia allies line up support as Houthi attacks mount
- Wikipedia — September 2026 Houthi strikes on Saudi Arabia
- NPR — Houthi attacks on Saudi Arabia ignite fires at oil facilities and wound 73 people
- Trading Economics — WTI, Brent Crude (September 25–26, 2026)
- Barchart — Copper Futures September 25, 2026
- APMEX — Gold Price September 25, 2026
- MetalCharts — Uranium Price September 26, 2026
- FRED — 2-Year, 10-Year, 30-Year Treasury Yields September 23, 2026
- Micron Investor Relations — Fiscal Q4 FY2026 Earnings Scheduled September 30, 2026
- Yahoo Finance — KOSPI Drops Below 8,000
- RBC Economics — September Executive Briefing: What to watch in the final quarter of 2026

---

Disclaimer

This report is produced for informational and educational purposes only and does not constitute investment advice, a solicitation, or a recommendation to buy or sell any security. All data cited reflects information available as of the publication time noted above. Market conditions and geopolitical developments may change materially before or during the trading session. Futures and pre-market levels are indicative only and are not guaranteed opening prices. Past performance of any strategy referenced is not indicative of future results. Consult a qualified financial advisor before making investment decisions.