Taiwan Banned Short-Selling Three Times in a Month. The First Two Didn't Work.
Taiwan's third attempt to halt the AI stock slide bans short-selling below the previous close for every listed stock, starting Monday. The benchmark is down almost 18% in a week. Here's why it probably won't change the direction, and what actually will.
Can a rule change stop a repricing? Taiwan is about to find out. Regulators there just banned short-selling below the previous close for every listed stock, effective Monday. It's the third attempt this month to halt a slide in the AI trade that's now gone global.
What the tape shows
The numbers tell the story. Taiwan's benchmark lost almost 18% in a single week. Three interventions in roughly four weeks. The first two, lighter curbs on short-selling, did nothing. Prices kept falling. So on Saturday the regulator went further and pulled its most direct tool: no shorting any stock below its previous close, starting Monday.
That's a hard floor on bearish positioning, at least on paper. It doesn't touch derivatives. It doesn't touch foreign institutional flows. And it doesn't touch the actual reason the market is selling off.
Why Taiwan matters here
Taiwan isn't a random market. It's the plumbing of the global AI trade. TSMC alone is roughly a third of the index weight, and it builds the chips every hyperscaler on the planet wants. When Taiwan sneezes, Nvidia catches a cold. So a short-selling ban here isn't a local story. It's a read on how much stress is building in the most crowded trade on earth.
Here's what matters: short bans historically don't work. The 2008 bans in the US and UK didn't stop the crash. They just pushed selling into futures and ETFs. Same playbook across Asia in 1998. Markets fall because holders want out, not because shorts push them down. Banning the short side doesn't change the thesis, it just hides the votes.
Frankly, the third attempt tells you more about the panic in the room than the first two did.
What traders are saying
According to the desks I've talked to, the ban is being read as a warning shot, not a cure. Positioning is getting pushed into options and offshore futures. The reality is that if the underlying AI capex story is intact, the ban is noise. If it isn't, the ban buys a few days at best.
The more interesting question isn't whether Taiwan's ban holds the line. It's what the scramble says about how thin the bid has gotten in AI names generally.
What to watch next
Three things over the next two weeks. First, Monday's open. If the Taiex gaps up and holds, the ban bought time. If it opens weak anyway, the market is telling you the ban is irrelevant.
Second, watch TSMC's ADR in New York. It trades through the Taipei close and doesn't care about short-selling rules. That's the honest price.
Third, watch the derivatives. If futures open interest spikes while cash volume dries up, shorts haven't gone away. They've just moved.
The takedown of a crowded trade rarely gets stopped by a rulebook. It gets stopped by buyers showing up with conviction. Right now, nobody's stepping in front of that train.
Related Articles
Key Terms Explained
Financial contracts whose value is based on an underlying asset.
Contracts to buy or sell an asset at a specific price on a future date.
The total number of outstanding derivative contracts (like futures or options) that haven't been settled.
Contracts giving the right, but not obligation, to buy (call) or sell (put) an asset at a set price before expiration.