Kevin Warsh walked into his first Federal Open Market Committee meeting on June 17 and did something nobody expected — he said almost nothing. His policy statement was three paragraphs long. His press conference was the shortest in modern Fed history. And his personal dot? Missing entirely. He declined to submit one.

But what he left unsaid spoke volumes. Because while Warsh stayed quiet, nine of his eighteen colleagues screamed from the rooftops: rate hikes are coming.

Think about that for a second.

Three months ago, the median FOMC member was projecting a quarter-point cut in 2026. Now half the committee wants to raise rates — into an economy already carrying $31.5 trillion in public debt, with interest payments running above $1 trillion a year. The Fed stripped all forward guidance from its statement. The easing bias that had been baked into markets since late 2024 is gone.

If you're a paid-up subscriber to the idea that we've returned to the era of easy money, I urge you to reconsider.

Bank of America didn't wait long to connect the dots. On Monday, BofA shifted its forecast to three quarter-point hikes this year — taking the benchmark rate to 4.25%–4.5%. Their analysts were blunt: the Fed has been tolerating inflation above 2% for five years, and Warsh's team has lost patience. Deutsche Bank echoed the call, flagging a September hike as the most likely flashpoint.

Markets responded exactly how you'd expect. The two-year Treasury yield surged 16 basis points — its biggest one-day move in over a year — to 4.21%. The dollar index broke above 101 for the first time in 2026. And gold, which had already been sliding from its February peak near $5,650, fell below $4,075 an ounce — down more than 25% from that high.

And one more thing... the tech trade imploded. On Tuesday, South Korea's KOSPI index collapsed nearly 10%. Samsung and SK Hynix each lost more than 12%. Micron fell 10%. Nvidia shed 3.2%. The Nasdaq dropped 2.2% in a single session. The VanEck Semiconductor ETF cratered 6.5%. Investors who had been piling into AI-related names suddenly needed cash — and they sold everything to get it, including their gold positions.

Most investors are still treating this as a garden-variety rotation. They're wrong.

What you're watching is the collision of three forces that haven't intersected since 2022: a hawkish Fed, a deflating war premium, and an AI investment cycle that's starting to look dangerously overfunded. Big Tech is on pace to spend $630 billion on data centers and AI infrastructure in 2026 alone, according to Morgan Stanley. That's more than the GDP of Sweden. And after Broadcom's guidance disappointed, the market is finally asking the question it should have asked six months ago — when does this spending actually produce earnings?

So why does this matter to you? Because the playbook just changed. The Iran Memorandum of Understanding signed on June 17 — the same day Warsh held his meeting — is pulling oil toward $72 a barrel. That crushes the inflation-premium trade. But core inflation — in services, in housing, in wages — hasn't budged. Warsh sees this. The committee sees this. And they're preparing to act.

As Buffett has said, only when the tide goes out do you discover who's been swimming naked. The tide just went out.

Gold at $4,075. Oil at $72. The two-year yield at 4.21%. Plan accordingly.

On Monday, a company most people have never heard of made the most important critical minerals deal of the decade — and the market barely noticed.

Energy Fuels, a uranium and rare earth producer based in Denver, announced a definitive agreement to acquire Vacuumschmelze — a century-old German magnetics company known as VAC — for $1.9 billion in cash and stock. The deal creates the first fully integrated Western mine-to-magnet rare earth platform. From feedstock in Australia, to processing at Energy Fuels' White Mesa Mill in Utah, to permanent magnet manufacturing at VAC's plant in Sumter, South Carolina — the largest rare earth magnet facility of scale in the United States.

In other words, the West just built its first credible answer to China's stranglehold on the permanent magnet supply chain.

The timing is not a coincidence. Two days earlier, China's Commerce Ministry placed MP Materials and USA Rare Earth — two of Washington's flagship rare earth companies — on its export control list. The move was largely symbolic — both firms say they've already weaned off Chinese supply. But the message was unmistakable: Beijing is tightening control at every node of the value chain. Licensing requirements that were suspended after the October 2025 Busan summit are set to snap back in November. And China's MIIT recently published enforcement rules that would fine companies up to five times their illegal gains for breaching production quotas — and revoke licenses for those exceeding quotas by more than 30%.

If you think the critical minerals story peaked last year, you're not paying attention. The VAC deal comes backed by a $725 million conditional loan from the U.S. Office of Strategic Capital — a clear indicator that the Pentagon is underwriting this race. Energy Fuels' Sumter plant can currently produce 2,000 tonnes of permanent magnets annually. The pathway takes it to 12,000 tonnes. The customer pipeline? Over $2 billion annually, across auto, defense, robotics, and data center sectors.

Keep in mind: China still controls more than two-thirds of global rare earth mining and holds a near-monopoly on refining. Breaking that dependency takes years and billions. But this deal is the kind of structural move — mine to mill to magnet — that actually changes the board.

Six months ago, WTI crude was trading above $117 a barrel and the Strait of Hormuz was effectively closed. Today, oil sits near $72 — its lowest since before the war — and tankers are transiting the strait with their satellite transponders on.

That's the Iran peace dividend in a single data point.

The Islamabad Memorandum of Understanding, signed remotely by Trump and Iranian President Pezeshkian on June 17, established a 60-day ceasefire window and committed both sides to reopening commercial shipping through Hormuz. The International Maritime Organization confirmed receiving security assurances, and the IEA reported UAE oil exports in early June rebounded to nearly 85% of pre-war levels. Iran itself shipped over 30 million barrels in the past week — at discounted prices to China — and a new 60-day U.S. waiver now permits global buyers, including American refineries, to purchase Iranian crude.

But here's where it gets complicated. Iran and the U.S. are already publicly disputing what they actually agreed to. Vice President Vance claimed Tehran would allow IAEA inspectors to return to bombed nuclear sites — a claim Iranian officials flatly denied. The nuclear question — particularly the disposition of 400 kilograms of 60%-enriched uranium — remains unresolved. And violence flared again in southern Lebanon this week, with Israeli forces killing two people just days after a ceasefire was brokered.

A full reopening of Hormuz could release roughly 80 million barrels onto the market. If that happens, the floor for oil drops further. If the 60-day window collapses — and the structural incentives in both Washington, Tehran, and Jerusalem all point toward fragility — crude snaps back above $90 before you can blink.

Most people have already moved on from the Iran war. The market has priced in peace. I urge you to price in the possibility that it doesn't hold.

Last September, D.E. Shaw — one of the most sophisticated quantitative trading firms on the planet — trimmed its semiconductor exposure ahead of what it called an "air pocket" in AI monetization. On Tuesday, the air pocket arrived.

The global tech selloff that ripped through markets wasn't triggered by a single event. It was a crowded trade finally getting stress-tested. South Korea's KOSPI fell nearly 10% — dragged down by Samsung and SK Hynix, both off more than 12%. The contagion spread to Europe, where the Stoxx 600 Technology index dropped 3.2%. STMicroelectronics and ASMI fell more than 7%. In the U.S., the Nasdaq shed 2.2%. Sandisk cratered 11.7%. Micron — reporting earnings Wednesday — plunged 10%.

And then there was Alphabet. Google's stock fell 5% Monday — its worst day in more than a year — after its VP of engineering and co-lead of Gemini AI, Noam Shazeer, announced he was leaving for OpenAI. On the same day, DeepMind VP John Jumper defected to Anthropic. Two of Google's most important AI minds, gone in 48 hours.

The deeper issue is capital. Morgan Stanley projects Big Tech will spend $630 billion on AI infrastructure this year. That spending hasn't yet produced the earnings to justify the valuations. Wedbush's Dan Ives called it a "gut check moment" but maintained his bullish thesis. IBM — the anti-AI-hype trade — jumped 5% on Tuesday after a JPMorgan upgrade.

Think about what that tells you. In a market obsessed with the future, the stock that rallied was the one selling the present.

In January 2026, the Committee for a Responsible Federal Budget confirmed that the U.S. government had already borrowed $602 billion in the first three months of the fiscal year — including $145 billion in December alone. Their assessment was simple: we're on course for a nearly $2 trillion deficit.

They were right. Through May, the cumulative FY2026 deficit reached $1.2 trillion — the net effect of $3.7 trillion in revenue and $4.9 trillion in spending. The Treasury Department itself now projects we will borrow $2 trillion or more this fiscal year. The Office of Management and Budget puts it at $2.065 trillion. The CBO's baseline is $1.9 trillion. The bond market median? $1.95 trillion.

Now stack that against what happened at the Fed last week. Nine of nineteen officials want to raise interest rates. Bank of America forecasts three hikes. The two-year yield has ripped to 4.21%. And every quarter-point hike adds roughly $8 billion per year to the government's interest bill on its floating-rate debt.

Let me put this in plain English. The federal government spent more than $1 trillion servicing its debt in fiscal year 2025 — more than it spent on national defense. The national debt passed 100% of GDP in March. And the Fed is now signaling it may raise the cost of carrying that debt. That's dropping Mentos into a can of Diet Coke.

Here's what matters. First, the interest math is now self-reinforcing. Higher rates mean higher interest payments. Higher interest payments mean larger deficits. Larger deficits mean more Treasury issuance. More issuance means upward pressure on yields. Rinse and repeat.

Second, weak Treasury auctions in March already flashed a warning. Primary dealers absorbed 24% of a 2-year note auction — roughly double the normal share. That means the natural buyer base is thinning at exactly the moment supply is surging.

Third, the Iran peace dividend won't save the fiscal picture. Lower oil prices ease headline inflation, yes. But customs duties — which surged on the back of tariffs — contributed $109 billion in extra revenue in the first five months of FY26. If tariffs come off as part of broader trade deals, that revenue disappears. And Social Security spending is up $57 billion. Medicare outlays are up $44 billion. These are structural, not cyclical.

The Government Accountability Office said it plainly in April: the nation's fiscal path is "unsustainable" and poses "serious economic, security, and social challenges." They're right. These debts will never be repaid in real terms. The only question is how the adjustment happens — through inflation, through financial repression, or through a crisis that forces action.

If you don't own hard assets — gold, copper, real estate, productive businesses — I urge you to reconsider your allocation. Gold at $4,075 is 25% below its February peak. That's not a sell indicator. That's a gift from a market that's forgotten why it bought gold in the first place.

Horse, meet water.

The May Personal Consumption Expenditures report — the Fed's preferred inflation gauge — lands Friday morning. If core PCE comes in hot, the September rate hike goes from probable to certain. If it surprises to the downside, Warsh gets breathing room. Every asset class pivots on this number. Watch it.

Micron reports Wednesday evening after a brutal 10% decline. Memory chips have been the tell for the entire AI hardware cycle. If Micron's guidance disappoints, the semiconductor selloff extends into July. If they deliver, this becomes the buying opportunity Wedbush has been talking about. Either way, the reaction tells you more than the number.

The Islamabad MOU started a 60-day window to negotiate final terms on nuclear inspections, sanctions relief, and the Strait of Hormuz. We're now one week in and the U.S. and Iran are already disputing what was agreed. If you're positioned for sustained low oil prices, keep your stops tight. This deal has more fault lines than the San Andreas.

Copper is testing a support zone that has held multiple times since 2021. Goldman Sachs recently raised its year-end target to $13,735 per tonne. If you believe in the electrification thesis — EVs, AI data centers, grid buildout — copper at these levels is the entry point. If China's demand continues softening, the floor drops to $5.50. Watch the LME inventories this week.