Park Ha Biological's 93% Crash Wasn't an Accident — It Was the Float

On July 7, 2025, shares of a Chinese skincare company called Park Ha Biological Technology closed at $41.01. The next day, they fell to $2.99. That is a 93% drop in one trading session. Not a slow decline. Not a bad earnings number. The price simply disappeared.

The company, which trades on the Nasdaq under the ticker BYAH (formerly PHH), made about $1.2 million in revenue in the first half of 2025. Its market capitalization before the crash was roughly $1 billion. After the crash, it was around $78 million.

The lawsuit that followed claims the collapse wasn't a normal market event. It alleges the company's IPO was intentionally structured — with a tiny public float, just 1.2 million shares out of 26.2 million total — to make exactly this kind of price manipulation possible. That the float was less than 5% of the company. That the stock was then promoted through WhatsApp groups, social media ads, and impersonators posing as financial advisors from places like Merrill Lynch, telling people the stock would go up 200–300% because of an unannounced partnership with L'Oreal.

The weird part isn't the pump-and-dump itself — those are old news. The weird part is that the IPO structure makes the scheme mechanically workable, and Nasdaq's listing rules let it happen.

How a tiny float turns a skincare brand into a leveraged bet

The IPO closed on December 30, 2024. Park Ha sold 1.2 million shares at $4.00 each, raising $4.8 million. After the offering, there were 26.2 million shares outstanding. The public — every broker, retail investor, and fund on the planet — collectively owned 4.6% of the company.

That means the tradable pool of shares was microscopic. When the supply of shares available to trade is that small, you don't need a lot of buying pressure to move the price. A few hundred thousand shares changing hands can swing the ticker by 20%, 30%, 50%. The stock doesn't need real demand — it just needs real sellers to sit still.

(The IPO prospectus warned, in standard boilerplate, that small-float stocks "may" experience extreme price swings. The lawsuit alleges the warning was misleading because it treated the float as an accident instead of a feature, and omitted that the stock was being targeted by a coordinated promotion scheme.)

From the IPO price of $4, the stock rose to $41 — a tenfold increase — without any material corporate developments. Revenue hadn't changed. No new product. No new market. No L'Oreal partnership. Just buying pressure in a pool so small that it didn't take much money to create a price.

And here is the plumbing detail that matters: the company is a Cayman Islands holding company that channels money through a Hong Kong subsidiary to Chinese operating companies. The founder and CEO, Xiaoqiu Zhang, controls about 73% of the voting power. It is classified as a "controlled company" under Nasdaq rules, which exempts it from standard corporate governance requirements. The structure doesn't create the pump-and-dump — but it means fewer independent directors, fewer disclosure obligations, and less internal friction if manipulation is happening.

The crash was the settlement

On July 8, 2025, volume exploded to 8.9 million shares — eight times the prior day's volume. Shares worth nearly $1 billion in market value evaporated. The closing price of $2.99 fell below the $4.00 IPO price, meaning the company was worth less publicly than it had raised in its offering.

The company didn't issue a press release. The complaint says it didn't comment at all on the unusual trading.

A 93% one-day collapse is not a re-rating. It is a settlement. The people who had been coordinating the buying — through WhatsApp groups, fake profiles, impersonators — had been sitting on shares accumulated at prices well below $41. When they sold simultaneously, the price had nowhere to go. There was no depth in the order book because there was no float. The buyers who had been told "this one's going to 80" were the ones left holding shares worth pennies of what they paid.

The class action lawsuit, filed as Robert Thomas v Park Ha, defines its class period from December 27, 2024 to July 8, 2025. Investors who purchased during that window and suffered losses can seek appointment as lead plaintiff. The deadline to do so is September 28, 2026. The lawsuit names the company, the CEO, the CFO, a director, the Cayman parent entity, and the underwriters Dawson James and D Boral Capital.

(For the reader wondering about their own exposure: if you bought BYAH between late December 2024 and July 8, 2025, you may be in the class. The lawsuit operates on a contingency basis, meaning investors don't pay attorneys up front. Whether a recovery is possible depends on whether there is actually money to recover — and at $2.41 per share with a micro-market-cap, the company's balance sheet is not a deep pocket. The real recovery target would be the officers, directors, and underwriters personally.)

This was not a single case. It was a template.

Park Ha was one of at least four Chinese small-cap stocks targeted by the same promotion mechanism in the summer of 2025. The others included Top KingWin, PicoCELA, and EPWK Holdings. Together these stocks lost roughly $3.7 billion in market value. The pattern was identical: tiny float, social media promotion, WhatsApp groups, impersonation, parabolic rise, catastrophic collapse.

The Inquirer reported that a Philadelphia-based underwriter called Bancroft Capital LLC handled at least 11 of these micro-IPOs between spring 2024 and September 2025. All 11 of those stocks now trade below their IPO prices. Nine trade 54–98% below. The group lost over $2 billion from peak prices. A securities fraud lawsuit names Bancroft as a defendant, though no regulator has accused the firm of wrongdoing.

Jay Ritter, a securities researcher at the University of Florida, put it bluntly: penny-stock IPOs are "generally targeted at unsophisticated retail investors" and "almost all of them crash and burn."

Nasdaq treated it as a structural problem, not an isolated one. In September 2025 — two months after the Park Ha collapse — Nasdaq proposed raising the minimum IPO size to $15 million generally and $25 million for companies operating primarily in China. The SEC approved the rule on May 14, 2026. It took effect in June. The threshold change effectively ended the pipeline of miniature Chinese IPOs on Nasdaq.

(Non-retroactively, of course. Park Ha was already listed, already crashed, already a ghost. The new rules close the door for future companies, not for the ones that already got through.)

What is left now

The stock trades at $2.41 as of September 10, 2026, up 6% on the day but down roughly 94% from its July 2025 peak. Daily volume is 73,000 shares. Turnover was $182,000. There is almost no liquidity — the kind of thin market where a few thousand shares can still move the price around, though nobody with real capital is playing.

The company made a $19.8 million net loss in the first half of 2025. Revenue grew to $1.24 million, but the loss was driven by an accounting charge for warrant impairment — a reminder that Park Ha filed a secondary offering with warrants in April 2025, raising additional capital while the stock was inflated. In May 2026, it filed an F-3 shelf registration for up to $300 million in future securities, a disclosure that is mechanically required but practically meaningless when the company's entire market cap is around $60 million.

The business itself — a Chinese franchise-based skincare brand with 43 franchisees as of April 2024, declining to fewer after terminations and downgrades — is not the story here. The story is that the company was listed on the Nasdaq through an IPO structure that turned it into a manipulable security, and that structure was a known vulnerability that the exchange only fixed after billions of dollars in investor losses.

If you are not a holder, the investment case is straightforward: this is a shell with a tiny operating business sitting on a collapsed stock price, with no clear catalyst, a class action lawsuit, and a regulatory environment that has been tightened specifically to prevent companies like this from listing in the first place. There is no obvious mechanism for value creation.

If you are a holder — and many people bought between $15 and $41 during the promotion period — the lawsuit is the only potential path to recovery. But class action recoveries against Chinese holding companies with tiny balance sheets are uncommon. The defendants' personal wealth and the underwriters' insurance are the only realistic recovery sources, and those require both a successful litigation outcome and collectible defendants.

The real takeaway isn't about Park Ha. It's about a classification boundary that Nasdaq only recognized after the damage was done: an IPO that sells less than 5% of a company to the public isn't really an IPO in the sense that it connects a business to capital markets. It's a mechanism for creating a tradable shell, and when that shell has a float so small that it can be moved with minimal money, it becomes a pump-and-dump waiting to happen. The $25 million minimum Nasdaq now requires of Chinese IPOs is, in effect, a price floor on the size of the float. Companies like Park Ha won't be able to do this anymore.

The question for investors is whether the structural lesson is worth the personal cost of learning it.