Issue 01: Where capital is rotating

Yield Theory Market Brief, Issue 01

As of the US cash close, Tuesday 30 June 2026. Reference session is New York. Written the morning of 2 July, so a few post-close developments (the June jobs report, oil headlines out of Hormuz) are flagged as live. Every figure is sourced inline and listed at the end. Where a number rests on a single source or an aggregator rather than a primary filing, we say so. That honesty is the product.

This is the half-year issue, and the theme picked itself. The question we exist to answer, where is the money actually going, has rarely had a cleaner answer than it did in H1 2026.

The one-paragraph version

The headline indexes finished the first half up high-single to low-double digits. That number is almost useless. Underneath it, capital did three very specific things, all of them driven by a single macro shock (a war), and all of them still live going into the back half. First, money moved down the AI stack, out of the crowded hyperscalers and application software and into the physical bottleneck: memory and the companies that build the picture. Second, it moved out of the top of the market, down the cap ladder into small caps and across the ocean into Asia, in the broadest rotation in years. Third, it rode the war trade into energy and straight back out as oil round-tripped. Sitting over all of it, the Federal Reserve quietly changed regimes: a new chair, a hawkish hold, and rate cuts erased from the Fed's own forecast. Most retail positioning is still built for the market of 2024. It should not be.

TL;DR

H1 close: S&P 500 +9.6% to 7,499.36, Nasdaq Composite +12.8% to 26,213.72, Dow +8.9% to 52,319.20, and the star, the Russell 2000 up roughly 22%, its best first half since 1991 (CNBC, 30 Jun 2026). Q2 alone was the best quarter for the S&P (+14.9%) since Q2 2020.
The trade of the year was memory. DRAM contract prices rose an unprecedented 93 to 98% quarter-on-quarter in Q1 (TrendForce, 1 Jun 2026). Micron (MU) printed $41.46bn of revenue in its May quarter, up 346% year-on-year, guided next quarter to $50bn at an ~86% gross margin, and its stock rose about 240% in Q2 alone (CNBC, 25 & 30 Jun 2026).
The AI trade changed hands. Capital rotated out of hyperscalers and out of software (Salesforce (CRM) -28% YTD, Microsoft (MSFT) -17% YTD despite AI leadership) and into the "enablers": memory, foundry, equipment, and power (Morningstar, 2026).
The real leadership was overseas. South Korea's KOSPI rose about 101% in H1, the best major index in the world, on the same memory story. Taiwan's TSMC (TSM) +55.5%, ASML +86.8%, MSCI Emerging Markets +24% (CNBC, 1 Jul 2026).
The Fed changed regimes. Kevin Warsh, sworn in 22 May, held at 3.50-3.75% in his first meeting, erased 2026 rate cuts from the dot plot (year-end median moved to 3.8%, implying a hike), scrapped forward guidance, and even withheld his own dot (Federal Reserve; CNBC, 17 Jun 2026).
Inflation is back, but it is an oil story. Headline CPI hit 4.2% in May, a three-year high, with energy driving more than 60% of the monthly rise. The Dallas Fed's trimmed-mean PCE was just 2.4%, telling you the spike is narrow, not broad (BLS; Dallas Fed, 2026).
Energy was the round-trip of the decade. The 2026 Iran war sent Brent to about $126 and regional Dubai crude to $166, then a June ceasefire unwound the entire premium: Brent closed 29 June at $73.21, barely above its pre-war level (CNBC, 29 Jun 2026).
The bubble question is now institutional. The BIS used its annual report to warn that the five largest hyperscalers will spend over $1 trillion on AI in 2025-26, outpacing their earnings and forcing debt issuance, with "circular" vendor financing it compared to the canal and railway manias (BIS, 29 Jun 2026).
What to watch: the 28-29 July FOMC (roughly 30% priced for a hike, not a cut), peak Q2 earnings, and the clock on the Hormuz ceasefire. The two things that break this market are the same two that made it: the war coming back, or the AI capex bill arriving before the revenue.

Market snapshot

Asset	Level (30 Jun close)	H1 2026	Note
S&P 500 (SPX)	7,499.36	+9.6%	Q2 +14.9%, best quarter since Q2 2020
Nasdaq Composite	26,213.72	+12.8%	Q2 +21.4%
Dow (DJIA)	52,319.20	+8.9%	Best H1 since 2021
Russell 2000 (RUT)	~3,030 (ATH 3,033.75)	~+22%	Best H1 since 1991
KOSPI (South Korea)		~+101%	Best major index globally
US 10Y yield	~4.44%	higher	2s10s curve about +30bp; bear-steepened in June
DXY (dollar index)	~101.3	higher	Highest since May 2025 after the hawkish hold
WTI crude	~$70/bbl	roughly flat	Full war round-trip
Brent crude	~$73/bbl (29 Jun)	roughly flat	Peaked near $126 during the war
Gold	~$4,090-4,110/oz	higher	Off war highs; still historically elevated

High-yield credit spreads sat near 285bp, well inside the 20-year average (~490bp), so there is no credit stress signal here. VIX and an exact 30 June gold print were not independently confirmed for this issue; volatility eased sharply as the war risk drained out. Crypto is not covered this issue.

The macro shock that set everything in motion

Everything downstream starts with the war. Fact: on 28 February 2026, the US and Israel launched large-scale strikes on Iran, opening what is now called the 2026 Iran war. Iran responded by disrupting the Strait of Hormuz, the chokepoint for roughly 20 to 25% of the world's seaborne crude and about 20% of its LNG (Wikipedia; NBC, 2026). A first ceasefire came on 7-8 April, and a memorandum of understanding on 14 June was signed by the two presidents on 17 June, setting a 60-day ceasefire and, on paper, reopening the strait (Britannica; CSIS; NPR, Jun 2026).

That arc explains the shape of the entire half. The war front-loaded an enormous energy trade into Q1, then re-lit inflation at the exact moment the market expected the Fed to be cutting. The collision between "sticky inflation" and "priced-for-cuts" is what turned an ordinary tape into a violent rotation. Understand the war and you understand why the money moved the way it did.

One detail that is not priced and matters enormously: the strait has not actually reopened. As of early July, commercial traffic through Hormuz was still running at roughly five ships a day against a normal ~93, with Iran citing Israeli strikes in Lebanon as a breach (NBC; Al Jazeera, 2026). Oil has round-tripped as if the risk is gone. The physical reality says otherwise. Hold that thought for the forward calendar.

Rotation 1: down the AI stack, from software to silicon to memory

This is the most important thing that happened in markets this half, and it is where the real alpha sits. For two years "the AI trade" meant a dozen mega-caps. In H1 2026 it split apart, and capital walked down the stack to the physical bottleneck.

The rotation, in one sentence: out of software and hyperscalers, into the companies that supply the buildout. Morningstar labeled it "AI disrupts the disruptors." Application software got re-rated on the fear that AI agents replace per-seat licenses: Salesforce (CRM) fell about 28% YTD, Workday (WDAY) about 22%, and even Microsoft (MSFT) was down roughly 17% YTD despite its AI position (Morningstar, 2026). At the same time, the hyperscalers that spend the money, Nvidia (NVDA), Amazon (AMZN), Alphabet (GOOGL), Meta (META), saw money rotate past them, to the layer that actually captures the pricing power.

Memory is the bottleneck, and the bottleneck has pricing power

Here is the alpha. In an AI buildout, the scarce physical input is not compute in the abstract, it is high-bandwidth memory (HBM) and the DRAM/NAND around it. Three companies, Samsung, SK Hynix, and Micron, control over 95% of DRAM, and they have spent 2026 reallocating wafers toward HBM for AI accelerators, starving everything else. Because HBM consumes roughly three times the wafer area per bit of standard DRAM, and new fabs take years, supply physically cannot respond inside 2026 (TrendForce, 13 Nov 2025). That hands the suppliers a once-in-a-cycle grip on price.

The numbers are genuinely historic:

DRAM contract prices rose 93 to 98% quarter-on-quarter in Q1 2026, with NAND up around 60%; Q2 guided prices kept climbing (DRAM +58 to 63%, NAND +70 to 75%) (TrendForce, 1 Jun & 31 Mar 2026).
HBM is sold out. SK Hynix and Micron have both said their HBM3E and HBM4 output is fully booked through calendar 2027, with demand visible into 2028 (CNBC, 25 Jun 2026).
The DRAM industry's Q1 revenue hit roughly $97bn, up about 81% quarter-on-quarter (TrendForce, 1 Jun 2026).

Micron (MU) is the cleanest expression of it. In its fiscal Q3 (the May quarter, reported around 24-25 June), Micron posted revenue of $41.46bn, up about 346% year-on-year, and adjusted EPS of $25.11 against roughly $20.49 expected. It guided the next quarter to $50bn of revenue at an ~86% gross margin and said HBM is booked through 2027 (CNBC, 25 Jun 2026; company transcript). The stock rose about 240% in Q2 alone, crossed a $1 trillion market cap on 26 May, and hit an all-time high near $1,255 on 25 June (CNBC, 30 Jun & 26 May 2026).

Analysis, and the reason this cycle may not rhyme with the last one: CEO Sanjay Mehrotra has emphasized multi-year Strategic Customer Agreements that lock both price and volume, a structural break from the one-year contracts and spot pricing that historically made memory a brutal boom-bust commodity. If those agreements hold, the 2026 memory cycle is less "commodity spike" and more "contracted supply agreement," which is a different, more durable animal. That is the bull case in one idea.

The enablers, and how big the moves were

The rotation into "AI enablers" produced some of the largest single-quarter equity moves you will ever see. Per CNBC (30 Jun 2026), Micron, Intel, and AMD together added roughly $2 trillion in market value in Q2:

Intel (INTC) rose about 216% in the quarter (roughly +$480bn) and is up about 278% year-to-date.
AMD nearly tripled in the quarter (roughly +$615bn).
Storage rode the same wave: Western Digital (WDC) and SanDisk (SNDK) posted several-hundred-percent year-to-date gains (source: stock aggregators; treat as reported, not primary-verified).

Speculation, clearly labeled: the "enablers over hyperscalers" trade is now consensus and no longer cheap. When a rotation gets a name and a Morningstar cover, the easy money is usually made. The late-June wobble (next section) is the tape starting to ask that question.

The bear case, stated honestly

We are not here to sell you a boom. The bear case is real and specific:

It is still memory. A commodity oligopoly whose record profits historically fund tomorrow's oversupply. Industry DRAM capex is rising toward roughly $61bn in 2026 (Micron ~$13.5bn, SK Hynix ~$20.5bn, Samsung ~$20bn), and the fabs that spend buys land as supply in 2027-2028. If AI end-demand disappoints, that is the setup for a classic capex hangover (TrendForce, 13 Nov 2025).
Valuations price no down-cycle. After 800%+ trailing-year moves, several of these names trade as if the cycle never turns. It always turns. The question is only timing.
Sentiment is fragile. On Friday 26 June, chip stocks sold off on worries about the rising cost of AI infrastructure: Intel -3%, SanDisk -10%, Arm -4%, Marvell -5%, and Micron gave back part of its post-earnings pop. Tellingly, Nvidia and Alphabet sat out the megacap bounce that day, the only two of the "Magnificent Seven" in the red (CNBC, 26 Jun 2026). The catalyst was a report that OpenAI might delay its IPO, plus Apple raising MacBook and iPad prices citing higher component costs, the first hint that memory's pricing power is becoming somebody else's margin problem.

The Yield Theory read: memory is the highest-conviction structural story in the market, and also the one most exposed to a 2027 air-pocket. The contracted-supply model is the thing to watch. If the Strategic Customer Agreements hold through the next demand scare, this is a multi-year re-rating. If they crack, it is 2018 again.

Rotation 2: out of the top, down the cap ladder and across the ocean

The second rotation is about breadth, and it is the healthiest thing that happened all half.

Small caps woke up. The Russell 2000 rose nearly 22%, its best first half since 1991, and briefly hit an all-time high of 3,033.75. This was not junk chasing: chip and chip-equipment names made up 16 of the Russell's 50 best performers, with Aehr Test Systems, Ichor Holdings, and MaxLinear each up more than 400%. Consensus 2026 earnings growth for the index climbed to about +38%, from roughly +23% at the start of the year (CNBC, 30 Jun 2026). Under the surface, the average S&P 500 member suffered a max drawdown of about 21% at some point this year even as the index rose, evidence of heavy rotation beneath a calm headline (Schwab, mid-2026).

The leadership, though, was not American. This is the point most US-centric investors are missing. On the same memory story, South Korea's KOSPI rose about 101% in H1, the best major index in the world, and SK Hynix overtook Samsung as the largest company in Korea for the first time in over 25 years (CNBC; KED Global, Jun 2026). Other markers of where the AI money actually went:

Taiwan's TSMC (TSM) +55.5%; Dutch equipment makers ASML +86.8% and ASM International +93.3%; the MSCI Emerging Markets tech index up more than 90% in H1 (CNBC, 1 Jul 2026).
MSCI Emerging Markets +24% overall; Nikkei 225 around +39% (CNBC, 2026).
The laggards matter too: Hong Kong's Hang Seng fell about 11%, and India was modestly negative, so this was not an "everything international" move. It was concentrated in the AI-hardware supply chain of North Asia.

Europe is a slower, structural story. Germany's 2026 budget marks a genuine regime change, roughly €520bn of spending with defense exempted from the debt brake and the military budget rising to about €108bn (CNBC; Military.com, 2026). But note the nuance, and the caution: the equity payoff from the defense theme was largely banked in 2025 (European banks had their best year on record, up about 76%), and in H1 2026 the narrow defense trade actually gave back ground (single-source reporting suggests Rheinmetall was down sharply YTD; treat as unconfirmed). The lesson: a great fiscal thesis and a great trade are not the same thing once the crowd has arrived.

The money followed. International drew roughly half of all equity fund inflows early in 2026, versus about 20% in 2025, and EM ETFs pulled in over $35bn year-to-date (iShares; SSGA, 2026). But watch the tell: those EM flows faded mid-year as the dollar firmed. Which brings us to the regime nobody is pricing.

The regime change nobody is pricing: Warsh's Fed

Fact: Kevin Warsh was sworn in as Fed chair on 22 May 2026, and the 16-17 June FOMC was his first. The Committee held at 3.50-3.75% by a 12-0 vote (Federal Reserve; CNBC, 17 Jun 2026). That is the boring part. The regime change is in everything around the decision:

The dots erased 2026 cuts. The June Summary of Economic Projections raised the year-end 2026 PCE inflation forecast to 3.6% (from 2.7% in March) and moved the year-end funds-rate median to 3.8% (from 3.4%), which implies roughly one hike, not a cut. Nine of eighteen dots sat at or above 3.875% (Federal Reserve, 17 Jun 2026).
He scrapped forward guidance. The statement was cut to around 130 words, Warsh eliminated forward guidance, and, in a genuine break with convention, he withheld his own dot from the projections. His stated view: markets "perform best when they react to incoming data" and worse when they are guessing the Fed's reaction function (CNBC; Fortune, 17 Jun 2026).

Analysis: this is the pivot beneath the rotation. A Fed anchored higher-for-longer, with an inflation problem and no forward guidance, does two things. It rewards real earnings and pricing power (chips selling into a supply-constrained boom) over long-duration promises (unprofitable software, rate-sensitive utilities), and it makes every data print a live event because the Fed will no longer pre-commit. It is also the single biggest threat to Rotation 2: Bank of America estimates that every additional 25bp hike knocks about 2% off Russell 2000 operating earnings, because small caps carry more floating-rate debt (CNBC, 30 Jun 2026).

The inflation is narrower than the headline. Headline CPI hit 4.2% in May (a three-year high) and headline PCE 4.1%, but core CPI rose only 0.2% month-on-month and the Dallas Fed trimmed-mean PCE was just 2.4% (BLS; BEA; Dallas Fed, 2026). Analysis: that gap is the whole ballgame. It says the inflation spike is an energy passthrough, not a broadening wage-price problem. If oil stays down, the print rolls over and Warsh never has to hike. If Hormuz re-closes, he might have to. The market is pricing the benign path.

The plumbing is calm. The 10Y sits near 4.44%, the 2s10s curve is about +30bp (bear-steepened as the long end led June's selloff), the dollar is at its highest since May 2025 (~101.3), and high-yield spreads near 285bp are well inside their historical average. No stress signal. Markets are pricing hawkish patience, not danger. The labor market backs that up: May payrolls were a firm +172k with unemployment at 4.3%, though the June ADP proxy softened to +98k and the official June jobs report is due the day this issue goes out (BLS; CNBC; ADP, 2026).

The bubble question is now on the central banks' desks

For most of the AI era, "bubble" talk was a permabear hobby. In H1 2026 it moved to the institutions that matter, and members should know exactly what they said.

The BIS, in its annual report (29 June), warned that the five largest hyperscalers are on pace for over $1 trillion of combined AI capex across 2025-26, already outpacing earnings and free cash flow and forcing debt issuance. It flagged "circular" vendor financing, hyperscalers taking equity in AI labs that then commit to buying chips, with "risks of the same asset being pledged multiple times," and drew explicit parallels to the canal, railway, and dot-com manias (BIS, 29 Jun 2026).
The scale is real: combined 2026 hyperscaler capex guidance runs around $700-725bn, up roughly 77% year-on-year, with Amazon near $200bn, Microsoft and Alphabet near $190bn, and Meta $125-145bn (Goldman Sachs; company guidance, 2026).
The valuation warning: long-term S&P 500 EPS-growth expectations have reached about 20.2%, exceeding even the 2000 peak of 18.6%, per Acadian's Owen Lamont, who said "2026 is looking like 1999." Jamie Dimon flagged "exuberance"; Goldman's James Covello called enterprise AI ROI "underwhelming so far" (Fortune, 8 Jun 2026).

The bull counter, for balance: JPMorgan's "what bubble?" note argues global AI capex of about $5.5 trillion through 2030 is "increasingly profitable," with hyperscaler operating cash flow set to surpass $900bn by 2027, though it raised its own estimate of debt financing to $4.1 trillion with loan-to-cost ratios above 85% (Fortune, 25 Jun 2026).

And the one genuinely reassuring fact: despite prices rising, the S&P 500 forward P/E actually fell, from roughly 22x in early January to around 20 by late June, because earnings did the lifting (FactSet; MacroMicro, 2026). Full-year 2026 EPS growth is tracking near 24%. This is a better-quality advance than the bubble headlines imply. The risk is not today's multiple. It is the leverage and circular financing sitting behind next year's capex.

Impact map

How the H1 forces map onto names and groups from here. Direction is our forward view; confidence is our own read, not a source claim.

Name / group	Ticker	Direction	Mechanism	Confidence
HBM / memory leaders	MU, SK Hynix (000660.KS), Samsung	↑	Sold-out HBM, contracted pricing, supply can't respond in 2026	High
Foundry / logic	TSM, INTC, AMD	↑	Buildout capex flows downstream; enabler rotation	Med-High
Memory storage	WDC, SNDK	↑	Same supply squeeze, NAND pricing	Medium
Hyperscalers	NVDA, MSFT, AMZN, GOOGL, META	→	Still growing, but capital rotated past them; capex-cost worry	Medium
AI-exposed software	CRM, WDAY, ADBE	↓/→	Agent-replaces-seat fear; recovery unproven	Low-Med
US small caps	RUT	↑ (rate risk)	Earnings broadening (+38% est.); -2% EPS per 25bp hike	Medium
North Asia AI supply chain	KOSPI, TSM, ASML	↑	The actual global leadership of the AI trade	Med-High
Energy	XLE, XOM, CVX	→ / ↑ tail	War premium gone; direction tied to Hormuz holding	Low-Med
Utilities / long duration	XLU	↓	Higher-for-longer plus risk-on both hurt	Medium
Data-center power	VST, D, nuclear/SMR names	↑	Power is the next bottleneck after memory	Med (early)

Note how many "obvious" reactions are already priced. The enabler trade in particular is crowded; the late-June selloff is the tape telling you the easy leg is done.

Disclosed trades of note

Legally disclosed public filings only (STOCK Act periodic reports and SEC Form 4s). This is not inside information, and correlation is not causation. Amounts are ranges, there is a reporting lag of up to ~45 days, and many trades are executed by a spouse or a managed account.

The Pelosi rotation (verified via Motley Fool). On 16 January 2026, filings for Nancy Pelosi (via Paul Pelosi) showed the exercise of call options into shares across NVDA ($250k-$500k), GOOGL and AMZN ($500k-$1m each), plus AI-adjacent names Tempus AI (TEM) and, notably, the AI-power play Vistra (VST). Analysis: this is itself a rotation, out of some 2025 winners and into AI-plus-power. The move into a merchant power name (VST) is the "electricity is the next bottleneck" thesis expressed in a portfolio a full quarter before it became a talking point.

The pre-war defense cluster (verified via finbold). A cluster of Republican senators and representatives, several with defense-committee proximity, bought defense primes in the weeks before the 28 February strikes:

Member (chamber/party)	Ticker	Buy/Sell	Amount range	Trade date	Note
Sen. Markwayne Mullin (R-OK)	RTX	Buy	$15k-$50k	29 Dec 2025	Senate Armed Services; same day bought CVX and COP
Sen. John Boozman (R-AR)	RTX	Buy	$1k-$15k	22 Dec 2025
Rep. Roger Williams (R-TX)	RTX	Buy	$1k-$15k	22 Dec 2025
Rep. Gil Cisneros (D-CA)	LMT	Buy	$1k-$15k	18 Nov 2025	Also a serial PLTR trader

Caveat: these buys landed two to three months before the strikes, during a period of rising tension. Whether that reflects committee information or ordinary macro positioning is not established by the filings. Note also that Sen. Mullin paired defense with energy (Chevron and ConocoPhillips) on the same December day, exactly the two sectors the war would move.

The insider signal is the absence of buying. Across eleven major chip names from 1 June onward, one dataset counted 63 insider sales worth about $155.6m against a single purchase of $374k, a roughly 416-to-1 sell-to-buy ratio by dollars (Kresmion, 26 Jun 2026). Analysis: before you read that as a top, note that the largest sales (Nvidia officers selling about $40m on 17 June, Palantir's Alex Karp selling roughly $54m) carry the fingerprints of scheduled 10b5-1 plans and option exercises, which are routine and low-signal. The more honest read is not "insiders are bailing" but "there is no cluster buying to confirm the rally." At these prices, nobody on the inside is backing up the truck.

(Reform watch: the bipartisan HONEST Act to ban congressional stock trading advanced out of Senate committee on an 8-7 vote, but we could not confirm any 2026 floor vote.)

What to watch (forward calendar)

Date (2026)	Event	Why it matters
2-3 July	June jobs report (BLS)	ADP proxy softened to +98k; a weak print against 4%+ CPI is the tension that could stay Warsh's hand
Mid-July	June CPI / PCE	The Fed just raised its own forecast; a hot core print hardens hike odds
2nd-3rd week July	Q2 earnings peak	Consensus near +24% S&P profit growth; the bar is high and memory/AI names must deliver
~30 July	Apple (AAPL) reports	Watch commentary on component (memory) costs squeezing hardware margins
28-29 July	FOMC decision	~30% priced for a hike; with no forward guidance, the statement and data around it are everything
~mid-August	Hormuz ceasefire clock	60 days from the 17 June signing; the strait is still barely flowing. A breakdown re-arms oil, inflation, and a hike
Ongoing	Memory contract renewals	The tell on whether Micron's multi-year pricing model actually holds through a demand scare

The Yield Theory view

If you take one thing from this issue, take this: the market that worked in 2024 is not the market you are in now. The passive, buy-the-mega-cap-index trade quietly stopped being the leadership trade in H1 2026. Leadership moved down the AI stack into memory, down the cap ladder into small caps, and out of the country entirely into North Asia. It did so under a Fed that has stopped promising to cut and an inflation impulse that is one Hormuz headline away from getting worse.

The rotation is real and, crucially, it is earnings-backed, not just multiple expansion. That is what separates it from 1999. But it is no longer cheap, the crowd has arrived in the enabler trade, and the two tail risks (the war returning, or the AI capex bill landing before the revenue) are both live and both underpriced. The honest posture into the back half is not "risk-off." It is "own the bottleneck, respect the breadth, and keep dry powder for the data print or the tanker headline that the consensus is currently pricing as if it cannot happen."

We will be back with a shorter update the moment the July FOMC or a Hormuz development demands one.

This is market research and commentary for informational purposes only, not investment advice. Do your own research; markets carry risk of loss. Figures are as of the 30 June 2026 US close unless stated, and were verified against public reporting at the time of writing. Several data points rest on secondary aggregators or single sources and are flagged as such in the text; treat anything time-stamped as a snapshot, because markets move.

Sources

CNBC, "Small-cap stocks enjoy best first half since 1991 as AI trade expands," 30 Jun 2026. https://www.cnbc.com/2026/06/30/small-cap-stocks-enjoy-best-first-half-since-1991-as-ai-trade-expands.html
CNBC, "Stock market news for July 1, 2026" (30 Jun close), Jun 2026. https://www.cnbc.com/2026/06/30/stock-market-today-live-updates.html
CNBC, "Tech leads first-half gains, but the biggest winners weren't in the US," 1 Jul 2026. https://www.cnbc.com/2026/07/01/tech-stocks-mag-7-emerging-markets-nvidia-chip-ai.html
TrendForce, "DRAM contract prices, Q1 2026," 1 Jun 2026. https://www.trendforce.com/presscenter/news/20260601-13070.html
TrendForce, "2Q26 memory pricing," 31 Mar 2026. https://www.trendforce.com/presscenter/news/20260331-12995.html
TrendForce, "Memory capex 2026," 13 Nov 2025. https://www.trendforce.com/presscenter/news/20251113-12780.html
CNBC, "Micron fiscal Q3 2026 earnings," 25 Jun 2026. https://www.cnbc.com/2026/06/25/micron-stock-3q-earnings-memory.html
CNBC, "AI chip rally in Q2 adds $2 trillion to Micron, Intel, AMD," 30 Jun 2026. https://www.cnbc.com/2026/06/30/ai-chip-rally-in-q2-adds-2-trillion-in-value-to-micron-intel-amd-.html
CNBC, "Micron crosses $1 trillion market cap," 26 May 2026. https://www.cnbc.com/2026/05/26/micron-stock-trillion-market-cap.html
Morningstar, "The Big 2026 Sector Rotation: AI Disrupts the Disruptors," 2026. https://www.morningstar.com/markets/markets-brief-big-2026-sector-rotation-ai-disrupts-disruptors
CNBC, "Nvidia, Alphabet sit out megacap tech bounce as chip stocks sink," 26 Jun 2026. https://www.cnbc.com/2026/06/26/global-tech-stocks-ai-infrastructure-costs-selloff-softbank-apple.html
Federal Reserve, "FOMC statement," 17 Jun 2026. https://www.federalreserve.gov/newsevents/pressreleases/monetary20260617a.htm
Federal Reserve, "Summary of Economic Projections," 17 Jun 2026. https://www.federalreserve.gov/monetarypolicy/fomcprojtabl20260617.htm
CNBC, "Five takeaways from Kevin Warsh's first meeting as Fed chair," 17 Jun 2026. https://www.cnbc.com/2026/06/17/here-are-the-five-big-takeaways-from-kevin-warshs-first-meeting-as-fed-chairman.html
Fortune, "Kevin Warsh's first Fed meeting: rates steady, forward guidance dropped," 17 Jun 2026. https://fortune.com/2026/06/17/kevin-warsh-first-fed-meeting-rates-steady-forward-guidance-dropped/
BLS, Consumer Price Index news release (May 2026). https://www.bls.gov/news.release/cpi.nr0.htm
BEA, "Personal Income and Outlays, May 2026," 25 Jun 2026. https://www.bea.gov/news/2026/personal-income-and-outlays-may-2026
Dallas Fed, Trimmed Mean PCE. https://www.dallasfed.org/research/pce
CNBC, "10-year Treasury yield falls below 4.5% as oil falls to pre-war levels," 24 Jun 2026. https://www.cnbc.com/2026/06/24/treasury-yields-oil-falls-pre-war-levels.html
CNBC, "Private payrolls rose 98,000 in June, less than expected," 1 Jul 2026. https://www.cnbc.com/2026/07/01/private-payrolls-rose-by-98000-in-june-less-than-expected-adp-reports.html
CNBC, "Oil prices, WTI back above $70 / Brent $73.21," 29 Jun 2026. https://www.cnbc.com/2026/06/29/oil-prices-wti-brent-crude-us-iran-strikes-strait-hormuz-talks.html
CNBC, "$166 a barrel: Middle East oil," 19 Mar 2026. https://www.cnbc.com/2026/03/19/166-a-barrel-middle-east-oil-gives-clue-to-where-all-prices-could-be-headed-if-iran-war-drags-on.html
NBC News, "Strait of Hormuz shipping traffic effectively at a standstill," 2026. https://www.nbcnews.com/world/iran/strait-hormuz-shipping-traffic-effectively-standstill-iran-ceasefire-rcna267391
Britannica, "2026 Iran war." https://www.britannica.com/event/2026-Iran-war
CSIS, "The United States and Iran Announce a Deal to End the War," Jun 2026. https://www.csis.org/analysis/united-states-and-iran-announce-deal-end-war-state-play
EIA, Short-Term Energy Outlook, June 2026. https://www.eia.gov/outlooks/steo/
Capital.com, "Oil price forecast" (Reuters survey, MS, Citi, Goldman), 30 Jun 2026. https://capital.com/en-int/market-updates/oil-price-forecast-30-06-2026
Fortune, "BIS warns on hyperscaler $1 trillion AI gamble," 29 Jun 2026. https://fortune.com/2026/06/29/bis-central-bank-warning-hyperscaler-data-center-1-trillion-gamble-recession/
Fortune, "2026 is looking like 1999," 8 Jun 2026. https://fortune.com/2026/06/08/ai-boom-tech-stocks-bubble-fears-earnings-growth-chipmakers-ipo/
Fortune, "What bubble? JPMorgan's $5.5 trillion AI capex call," 25 Jun 2026. https://fortune.com/2026/06/25/what-bubble-jpmorgan-5-5-trillion-ai-capex-explosion-profitable-for-now/
Tom's Hardware, "Big Tech AI spending plans reach $725 billion," 2026. https://www.tomshardware.com/tech-industry/big-tech/big-techs-ai-spending-plans-reach-725-billion
NVIDIA, Q1 FY2027 results, 20 May 2026. https://nvidianews.nvidia.com/news/nvidia-announces-financial-results-for-first-quarter-fiscal-2027
FactSet, Earnings Insight, 2026. https://insight.factset.com/sp-500-earnings-season-update-may-8-2026
KED Global, "SK Hynix overtakes Samsung," 22 Jun 2026. https://www.kedglobal.com/korean-chipmakers/newsView/ked202606220007
CNBC/Goldman Sachs, "Germany defense spending, fiscal stimulus," 18 Feb 2026. https://www.cnbc.com/2026/02/18/goldman-sachs-germany-defense-spending-military-fiscal-stimulus-equities.html
iShares, "2026 ETF market trends and flows," 2026. https://www.ishares.com/us/insights/inside-the-market/2026-etf-market-trends-and-flows
Motley Fool, "Nancy Pelosi's AI stock trades," Feb 2026. https://www.fool.com/investing/2026/02/01/nancy-pelosi-stocks-ai-stocks-nvda-stock/
finbold, "War stocks US politicians bought before Iran strikes," 2026. https://finbold.com/2-war-stocks-u-s-politicians-suspiciously-bought-just-before-iran-strikes/
Kresmion, "Daily brief: chip insider selling," 26 Jun 2026. https://kresmion.com/daily-brief/2026-06-26
Charles Schwab, 2026 Mid-Year Market Outlook. https://www.schwab.com/learn/story/us-stock-market-outlook