The Trillion-Dollar Bet That the Cycle Is Dead

The market just handed a memory company a trillion-dollar valuation at roughly a hundred times trailing free cash flow, on a yield under one percent.¹² Strip the letters AI off that sentence and it is a bet that the most violently cyclical business in technology has quietly stopped being cyclical.

On Wednesday’s main show, Hunt, Jason, and Mike spent an hour on the opposite problem — how you get to a two-trillion-dollar valuation on a company with almost no free cash flow at all. That was the SpaceX S-1.³ This week the market answered the mirror image for Micron. It looked at the cash that actually exists, and it priced that cash as if it can never fall again.

Here is the part the screens are not showing you. As recently as fiscal 2023, Micron lost almost six billion dollars in a single year — and its gross margin went negative, which is the polite way of saying it sold memory for less than it cost to make.⁴ That was not 1998. That was two years ago. The cycle before it: fourteen billion in net income in 2018, gone to a fraction of that by 2020.⁵ Boom, glut, collapse, repeat. The most reliable pattern in semiconductors, and Micron has run it twice in the last eight years.

The mechanism is not mysterious. High memory prices pull in capacity. Capacity becomes supply. Supply kills the price. A hundred times trailing free cash flow is the market betting that mechanism has been switched off — that high-bandwidth memory, the stuff that feeds the AI accelerators, is different enough to outrun the cycle for years instead of quarters. The bull case is real, and worth stating fairly: Micron’s entire 2026 HBM output is already sold out,⁶ management says it can fill only half to two-thirds of what its largest customers are asking for,⁷ and trailing free cash flow has grown more than fivefold off the 2023 trough.⁸ The stock rose nineteen percent the day it crossed a trillion, and nearly ninety percent in a month.⁹ ¹⁰ None of that is fake. The only question is whether it is permanent.

Now put Broadcom next to it, because it reports Wednesday and the contrast is the whole point.¹¹ Broadcom is guiding AI-chip revenue up a hundred and forty percent, to roughly eleven billion dollars in a single quarter¹² — and it goes into that print at about seventy times trailing free cash flow, on more than thirty billion dollars of cash it has already banked.¹³¹⁴ The cashflow read is in Marcus’s column below; short version, one of these names is a bet on a shortage and the other is a bet on a standard. Micron needs the cycle to stay broken. Broadcom gets paid whether the buyer is Google, Meta, or Anthropic, and it needs nothing about memory pricing to be true. Page one of the Cash Flow Memo has both names. The market is paying the fatter multiple for the one carrying all the cycle risk.

What changes the read is not a hiring chart or a single beat. It is the first crack in sold out. Broadcom’s print on Wednesday, June third, is the near-term referee — if a hundred-and-forty-percent AI guide actually holds, the standard is winning and the shortage trade has competition. The Micron test runs longer: watch HBM contract pricing into 2026 and the first time a competitor qualifies into the big accelerator sockets. The thesis breaks the day sold out quietly becomes renegotiated. That is always how a memory cycle turns. Not with a warning. With a renegotiation.

For a long-cycle owner the discipline is simple and unpopular. You are not paid to decide whether AI memory demand is real this year. It is. You are paid to ask what Micron earns across the whole cycle, trough included, and whether a trillion-dollar tag survives the year the glut comes back. In 2023 the answer to that question was a six-billion-dollar loss. The market has decided that year was the last of its kind. The market has decided that before.

Wall Street’s consensus on Micron’s trillion-dollar tag: AI broke the memory cycle. The memory cycle has broken that consensus before — twice since 2018, and the second time the gross margin went negative.

The Tape — W2622

Universe of 94 cashflow-memo names, snap dates 2026-05-26 → 2026-05-30. Composite is rank-sum percentile of FCF Yield + NTM Revenue Growth (higher = better balance). Financials shown separately.

Telltales Yield — Top 10

* 1 | UBER | Technology | 80 | 7.5% | 15.2% | 13.2 | -2.0%

* 2 | FCX | Basic Materials | 78 | 6.3% | 19.8% | 16.0 | 6.0%

* 3 | REGN | Healthcare | 77 | 8.8% | 10.6% | 11.3 | -3.8%

* 4 | LNTH | Healthcare | 73 | 6.6% | 13.2% | 15.1 | -3.6%

* 5 | CRM | Technology | 72 | 8.5% | 9.6% | 11.8 | 6.1%

* 6 | NOW | Technology | 68 | 3.8% | 18.5% | 26.1 | 21.8%

* 7 | ABNB | Consumer Cyclical | 67 | 6.5% | 10.6% | 15.3 | 0.7%

* 8 | PYPL | Financial Services | 67 | 16.5% | 4.1% | 6.1 | 1.2%

* 9 | AM | Energy | 67 | 8.5% | 6.0% | 11.8 | -5.5%

* 10 | LNG | Energy | 66 | 7.9% | 6.7% | 12.6 | -6.6%

From the Cashflow Desk — Marcus Graham

Micron’s trillion-dollar tag and Broadcom’s get talked about as the same AI trade. They are not. Micron sits at 102x trailing FCF — a multiple that only pays off if the memory cycle is dead, in a business that lost $5.8B two fiscal years ago when the last glut hit. Broadcom goes into Wednesday’s print near 68x and needs nothing about memory pricing to cooperate; it gets paid on the custom-silicon standard whether the buyer is Google, Meta, or Anthropic. One name is a bet on a shortage. The other is a bet on a standard, and the standard is the cheaper multiple. The test is Wednesday, June 3: if Broadcom’s AI guide holds, the cycle-proof story suddenly has a cheaper rival. We re-anchor Micron when the next HBM contract-pricing print lands.

Telltales Yield — Bottom 10

* 1 | VG | Energy | 2 | -8.2% | -9.2% | — | -12.9%

* 2 | LEN | Consumer Cyclical | 18 | 0.1% | 3.4% | 1589.4 | 1.0%

* 3 | NKE | Consumer Cyclical | 19 | 1.8% | 0.6% | 56.0 | 3.5%

* 4 | FANG | Energy | 20 | 2.8% | -6.6% | 36.4 | -4.6%

* 5 | XOM | Energy | 22 | 3.3% | -9.0% | 30.6 | -6.2%

* 6 | SBUX | Consumer Cyclical | 25 | 2.4% | 2.3% | 40.8 | -3.8%

* 7 | CVX | Energy | 29 | 4.1% | -13.0% | 24.5 | -4.7%

* 8 | WMT | Consumer Defensive | 29 | 1.6% | 4.7% | 64.0 | -3.8%

* 9 | INTC | Technology | 35 | -0.4% | 10.7% | — | -4.3%

* 10 | COST | Consumer Defensive | 36 | 2.2% | 7.9% | 45.8 | -7.0%

This Week’s Reporters

* AVGO | Technology | 2026-06-03 | 1.5% | 57.1% | 58

* FIVE | Consumer Defensive | 2026-06-03 | 3.0% | 9.9% | 48

Sector Medians

* Sector | N | Median Composite | Median FCF Yld | Median NTM Growth

* Financial Services | 4 | 61 | 3.8% | 11.6%

* Healthcare | 10 | 56 | 6.4% | 10.0%

* Technology | 15 | 54 | 1.6% | 25.1%

* Communication Services | 11 | 54 | 4.1% | 4.5%

* Basic Materials | 3 | 50 | 6.3% | 6.5%

* Industrials | 8 | 49 | 3.3% | 9.0%

* Consumer Defensive | 5 | 48 | 3.0% | 7.9%

* Consumer Cyclical | 12 | 45 | 3.3% | 4.8%

* Energy | 21 | 44 | 5.9% | 0.7%

* Utilities | 1 | 42 | 2.4% | 9.5%

Debt / FCF Watch (highest leverage on TTM FCF)

* Ticker | Sector | Net Debt / FCF | FCF Yld | Composite

* TRGP | Energy | 20.8 | 1.2% | 44

* NEE | Utilities | 16.2 | 2.4% | 42

* MTDR | Energy | 12.8 | 2.7% | 51

* CHTR | Communication Services | 10.6 | 8.0% | 50

* ET | Energy | 10.0 | 5.4% | 44

* DE | Industrials | 9.4 | 4.4% | 56

* EPD | Energy | 9.4 | 3.3% | 43

* FDX | Industrials | 8.8 | 3.6% | 46

Weekly Price Movement

Top 5 (week-over-week price) | Ticker | Sector | Price | 1-wk % | |——–|——–|——:|——-:| | SNOW | Technology | $255.55 | 48.4% | | MU | Technology | $971.00 | 29.3% | | NOW | Technology | $124.37 | 21.8% | | ORCL | Technology | $225.78 | 17.5% | | PLTR | Technology | $156.54 | 14.4% |

Bottom 5 (week-over-week price) | Ticker | Sector | Price | 1-wk % | |——–|——–|——:|——-:| | VG | Energy | $12.04 | -12.9% | | OKE | Energy | $83.94 | -10.7% | | KMI | Energy | $31.08 | -8.0% | | TRGP | Energy | $255.07 | -7.8% | | CF | Basic Materials | $112.35 | -7.7% |

Financials (shown separately — FCF metric not meaningful)

* Ticker | Price | 52-wk Position | Div Yld

* JPM | $299.31 | 51% | 2.1%

* MS | $208.00 | 100% | 2.0%

* GS | $1025.56 | 100% | 1.9%

* IBKR | $86.97 | 96% | 0.1%

Data Gaps

90 of 90 non-Financial names ranked. 0 dropped for missing FCF yield or NTM revenue growth.

Source: cashflow-memo master_2026-05-30.csv. NTM growth from FMP analyst-estimates consensus. Composite is a percentile rank, not a recommendation.

▶ Explore the interactive Tape →

The Cashflow Memo

Five P...