Larry McDonald — A Former Lehman Trader Sees the Same Warning Signs Again

FCG is a fund holding natural-gas companies. McDonald says these stocks are "really breaking out" — outperforming the broad market — and still cheap.

His thesis: data centers for AI need huge amounts of power, and natural gas is a cheap, ready supply. So gas is the "power" side of the AI trade — owning the fuel behind AI rather than the chips.

Tourmaline is a large Canadian natural-gas producer. He uses it to explain "trapped" gas: in some regions there aren't enough pipelines to move the gas to market, so it sits there and stays very cheap.

His insight is that AI data centers will move to where the cheap, stranded gas is, rather than forcing the gas to come to them. That turns trapped gas from a problem into an advantage — and makes producers like Tourmaline winners in a multi-year gas boom.

Chevron is one of the biggest oil & gas companies. He names it (with Exxon) as a destination for the roughly $4 trillion that has been flowing out of the Nasdaq-100 tech names.

He calls this "the great migration" — money leaving expensive tech and moving into companies that own real, physical assets like oil and gas, which do well in an inflationary, stagflation-leaning world.

Exxon is the other oil-and-gas supermajor he names alongside Chevron.

Same logic: it's a big, cash-generating owner of energy assets, exactly the kind of name catching the money rotating out of tech and into hard assets.

Glencore is a global mining and commodity-trading company — it owns and trades physical metals and materials. It's a foreign (international) stock, which he favors as cheaper than U.S. names.

He groups it with the miners that "own lots of assets" — the type of hard-asset business that tends to beat tech and growth stocks during stagflation (rising prices plus slow growth).

BHP is one of the world's largest miners, producing iron ore, copper and more — a core "owns real stuff in the ground" holding.

It's a centerpiece of his "great migration" basket: as money leaves paper assets (stocks and bonds) for hard assets, global value miners like BHP are where he expects it to land.

Freeport-McMoRan is a major copper miner. Copper is essential for electricity and the power grid, which ties directly into the electrification and AI-power buildout.

He includes it in the hard-asset basket money is rotating into — a real-asset producer positioned for the commodity bull market he expects.

GDX is a fund of gold-mining companies. He sold it in January when the market got frothy (traders were buying eight times as many bullish silver bets as bearish ones — a sign of crowd euphoria).

Now, after a sharp pullback toward the "100-day moving average" (a common gauge of the medium-term trend), he's buying back. His rule: in a young bull market, dips toward that line are good entry points. He still sees gold as under-owned — only ~1.25% of household wealth versus ~3% in the 1980s.

SLV is a fund backed by physical silver. Same playbook as the gold miners: he sold in January during the froth and is buying back into the pullback.

His big-picture case is that hard assets are still badly under-owned (around 1.25% of household wealth today versus ~3% in the 1980s), leaving plenty of room to grow as money keeps rotating in.

SIL holds silver-mining companies (rather than the metal itself). It's part of the same gold/silver miner basket he sold in January and started rebuying in this washout.

He frames the drop as the "tourists" and "weak hands" — short-term, unserious buyers — getting flushed out, while serious investors step in to buy the dip.

IBIT is a fund that holds Bitcoin for you, so you can own it through a normal brokerage account. This is McDonald's first-ever Bitcoin purchase.

Two reasons. First, the Bitcoin-to-gold ratio (how many ounces of gold one Bitcoin is worth) fell from about 38 to about 13; historically, when it gets that low he'd sell some gold and buy Bitcoin. Second, ETFs like this have "democratized" ownership — spreading Bitcoin across many holders instead of a handful of big families, which he thinks makes it less prone to a sudden crash if one whale needs to sell. He notes there are cheaper ways to own Bitcoin, but the ETF is the easiest for ordinary investors.

Micron makes memory chips (DRAM) — the working memory inside computers and AI servers. There's a shortage, so memory prices have soared, and Micron is making a fortune.

His phrase: "Micron is robbing the Mag 7." The AI giants must buy huge quantities of this now-expensive memory, which inflates their costs and squeezes their profits. So Micron is a beneficiary of the very dynamic that's hurting the big tech names — interesting as a symptom, not a buy/sell call here.

Caterpillar makes the heavy construction equipment used to build things — including data centers. He notes its machines are in short supply.

The point: scarce equipment drives up the cost of building data centers, which adds to the cost squeeze on the big AI companies. Like Micron, he cites it to illustrate why Mag-7 profit margins are under pressure, not as a stock recommendation.

BlackRock is the world's largest asset manager and the firm behind the IBIT Bitcoin ETF.

Here he mentions it positively in passing: its Bitcoin ETF helped spread ownership of Bitcoin across many investors (instead of ~18 families controlling 60% of it). That broader ownership base is part of why he's comfortable buying Bitcoin now — a supporting point rather than a view on BlackRock the stock.

UNG is a fund meant to track the price of natural gas, giving ordinary investors easy exposure without a futures-trading account.

His caution: to track gas, the fund constantly has to "roll" futures contracts — sell expiring ones and buy later-dated ones — which usually costs money each time. Over the long run this "decay" quietly eats away returns, so the fund lags the actual commodity. Fine for a quick trade, weaker as a long-term hold.

USO is the oil equivalent of UNG — a fund that aims to track the price of oil for people who can't trade futures directly.

Same warning: the constant rolling of futures contracts creates a slow drag on returns, so it works for a short-term move but underperforms owning the commodity over the long haul.

Meta (Facebook, Instagram) is one of the "Magnificent 7" tech giants. It's down more than 20%, worse than the group.

The reason: it's spending enormous sums on AI data centers while the cost of memory chips and construction soars — and all that spending squeezes profits. That margin pressure is what's dragging the stock down.

Nvidia makes the chips at the heart of the AI boom and was, with Microsoft, one of the two biggest stocks in the S&P 500.

It's down about 19% as the AI-spending story sours and money rotates out of these crowded top-two index holdings. When the most over-owned names start to unwind, he expects them to keep underperforming as cash moves toward hard assets.

Microsoft is the other half of that top-two S&P concentration (with Nvidia, ~14–15% of the index). It's down roughly 28–30% from its highs.

Strikingly, the S&P overall is only down ~6% even though its two biggest stocks fell that much — proof, he says, of a "broadening": money leaving the giant tech names and spreading into hard-asset stocks instead.

Block (formerly Square) is Jack Dorsey's payments/fintech company. He cites it as the poster child for AI-driven layoffs: Dorsey cut 45% of the workforce — and the stock rose 30% on the news.

That reaction, he warns, encourages "copycat" companies to slash jobs too. The bigger worry is the wave of layoffs hitting consumer spending and feeding the slowdown — so it's a flag for the broader disruption, not a buy.

TLT holds long-term U.S. government bonds (those maturing in 20+ years). "Everyone's losing money in TLT," he says — long-dated bonds have steadily bled since 2022, in the U.S., U.K. and France alike.

The deeper point: the classic "60/40" portfolio relied on bonds rising when stocks fell. Since 2022 that offset has broken — stocks and bonds now fall together — so long bonds no longer protect you, and the money keeps fleeing into hard assets.

XLF is the main fund tracking U.S. financial-sector stocks (banks, insurers, etc.). McDonald recommended betting against (shorting) the financials.

His reasoning is "two punches at once": banks are exposed to the brewing private-credit mess, and they also face disruption as AI hits the software companies they've lent to. The group has underperformed the S&P by the most since the 2008 crisis — though it's now so beaten down it's deeply oversold short-term.

MetLife is a large life insurer. Insurers, he argues, are the ultimate "bag holders" of the private-credit boom: chasing higher returns ("reaching for yield"), they loaded up on private loans that were stamped with overly generous credit ratings.

He likens it to 2008, when junk mortgages got rated safe — and says clients are shorting MetLife-type insurers most exposed to private credit, betting they'll be hit when the bad loans are revealed.