Sell Your Underwear to Buy More Shares

Manias, markets, and the people who start them

In Seoul this winter, a twenty-four-year-old software developer named Na Se-bin put roughly $47,000—nearly everything she had—into the stock market. She has won and lost a month’s salary in the space of a second. She keeps going anyway. She and her co-workers joke that they should sell their underwear to buy more shares.

I have read that sentence in a dozen languages and four centuries. In 1720, a clerk on the rue Quincampoix sold his coach to buy more Mississippi stock. In 1929, an Iowa insurance salesman borrowed against his house to buy more Wright Aeronautical. The costume changes.

The joke about the underwear does not.

What’s happening across Asia right now is the largest version of that joke I’ve seen in my career.

Taiwan’s market has doubled in a year. South Korea’s has tripled. Japan’s Nikkei is up more than eighty percent. In Taipei, taxi drivers trade stocks mid-ride. In Seoul, a thirty-five-year-old elementary school teacher has watched a $300,000 portfolio grow fivefold and is now deciding between a Mercedes S-class and a Tesla. More than 180,000 trading accounts for children eighteen and younger were opened at a single Korean brokerage in the first three months of the year. One of them needed a parent’s signature.

These are real numbers attached to a real thing. That is what makes them dangerous. Aviation was real in 1929 too. The planes flew. The mail got carried. Lindbergh’s achievement was genuine, his character impeccable, and the boom that bore his name destroyed six hundred thousand people anyway.

The honest story and the dishonest story produce the same collapse. The semiconductors are real. The AI build-out is real. None of that has ever been the question.

The question is always the same one, and you already know how I’m going to phrase it. Who is the teller?

Because there is one. There is always one. And in this mania he is not hiding—he is doing an eighteen-day victory lap across Asia in a leather jacket.

His name is Jensen Huang, and he runs Nvidia.

Read the dispatches and notice what people are actually buying. Not chips. Not gross margins. Not the discounted cash flows of a contract foundry. They are buying the man who flew into Taipei and called the island the epicenter of the AI revolution and promised to spend $150 billion a year there.

They are buying the man who, when Korea’s main index cratered more than eight percent in a single session while he was in town, shrugged in the Seoul heat and said everyone should be happy, because now you can buy the stock more cheaply.

Sit with that line. The market had just erased a fortune in an afternoon, and the most important man in the industry reframed the loss as a sale. And it worked. He was not lying—I think he believes every word, which is precisely what makes him the kind of figure I’ve spent four hundred years of history chasing. The knowing fraudster is a different animal, and a less interesting one. The dangerous one is the true believer whose conviction is so total it becomes yours before you’ve finished your coffee.

They did not buy Nvidia. They bought Huang’s certainty.

And here is the part the bubble taxonomies miss, the part that took me too long to learn at my own desk. The crowd is not a herd of fools being hypnotized by a salesman. Look closely at the people in this story and you find they are reasoning. The Taiwanese insurance agent who put more than half his salary into chip stocks, watched it quadruple, and bought a four-bedroom apartment—he is not stupid. He told a reporter that none of it would have happened without semiconductors, and he is correct. The banker, the teacher, the developer: each one is running the same arithmetic Huang is running, arriving at the same answer, feeling the same warm private glow of conviction.

That is the secret of every teller I have ever studied. He does not implant a foreign idea. He crystallizes the one you were already half-telling yourself. Huang succeeds not because his audience is gullible but because his audience is built from the same material he is—creatures that process an uncertain future by turning it into a story, then mistaking the story for the future. He says the words out loud. The crowd was already mouthing them.

You can measure how far the story has detached from the arithmetic by where it shows up. A TSMC-branded rice cooker—the company’s corporate colors, a little circuit-board logo—sells secondhand for $312, more than four times what employees pay. The red holiday envelopes the company hands out, the contents removed, go for fifteen dollars apiece.

When a manufacturer’s gift wrapping trades at a premium on the resale market, you are no longer pricing chips. You are pricing belief in the people who make them. In Taipei, convenience stores sell lottery tickets that pay out in shares of Nvidia. The lottery and the stock market have stopped being metaphors for each other.

I am not telling you the AI boom is a fraud or that it ends tomorrow. I don’t know when it ends. Neither does anyone selling you a date. What I know is the structure, because I’ve now watched it wear the dot-com costume, the housing costume, the crypto costume, and I’ll watch it wear another one after this. A genuine technology. A coherent story. A single figure whose conviction the crowd adopts as its own. Prices that float free of the thing underneath and ride on the belief alone. And then, eventually, the moment nobody can schedule: belief in the man falters, and the price falls not gradually, the way a thesis dies under contrary evidence, but all at once, as if the story had never made sense.

The one figure in the whole piece who gives me pause is a twenty-one-year-old semiconductor student in Kumamoto named Ryoki Nao. He’s surrounded by classmates who invest. He watches the market. He doesn’t buy. He wants to graduate and earn real money first.

He may be the only person in the story who understands the arithmetic and still won’t let it tell him what to do. Or he’s just early, and in eighteen months he’ll be in a margin account like everyone else, selling his underwear.

History says the second one. History almost always says the second one.

Watch the next mania form—and one is always forming—and don’t waste your first question on the valuation or the macro or the chart. Ask the only question that has ever mattered.

Who is telling you the story you already wanted to hear?