Hedge Funds Dumped Tech at Near-Record Pace, Then Chips Jumped 7%

Hedge funds sold technology stocks at a pace matching the highest levels on record since 2016 — and the very morning that data surfaced, on November 9, the Philadelphia Semiconductor Index rose 6.9%, its best day since July 30. The timing is not a coincidence. It is the signature of forced deleveraging unwinding, and it tells chip investors a lot about what just happened to their funds.

Goldman Sachs prime brokerage data through November 6, 2026, showed technology was the most heavily net-sold sector, both globally and in the U.S. The iShares Semiconductor ETF (NASDAQ:SOXX) closed at $146.76, up $10.23, or 7.49%, and slipped to $146.41 after hours. The VanEck Semiconductor ETF (NASDAQ:SMH) gained $23.45, or 5.05%, to $487.41, and the Invesco QQQ Trust (NASDAQ:QQQ) rose 3.54% to $679.35.

The stakes for chip fund holders are straightforward: how much of this move reflected forced selling, and how much reflected a genuine rethink of AI demand. The AI buildout thesis still rests on the companies supplying power, cooling, and networking to data centers — the infrastructure layer beyond the chipmakers themselves.

What does the prime brokerage data actually show?

The selling was broad and violent. Tech led net sales globally and in the U.S., and that selling produced the biggest wave of single-stock net selling in four months. Funds also dumped U.S. industrials at the fastest pace since April 2020.

Among U.S. fundamental long/short funds, gross leverage fell 8.7 percentage points to 207.5% — the largest drop since March 2020. Gross leverage is a fund's total long plus short positions measured against its capital, so a drop this size means funds cut both sides at once.

Net leverage, which measures how much a fund is leaning on the market's direction, fell 3.7 points to 43.3%. That was the largest drop since the April 2025 "Liberation Day" tariff week. By November 6, funds had turned net buyers — mostly to cover index and ETF shorts.

The takeaway: this was funds slashing risk on both sides of their books, not a considered bet against chips.

Why did chip funds bounce harder than the market?

Context matters here. The selling came during a post-midterm rout. On November 5, the Nasdaq-100 had its deepest drop since the Covid pandemic and lost $2.3 trillion in market value in a single day.

Then the rebound. On November 9, the S&P 500 rose 2.4%, or 173.08 points, to 7,549.94. The Nasdaq Composite gained 2.7% to 25,253.14, and the Nasdaq-100 added 3.4%.

The mechanism behind the outsized chip rally is mechanical, not sentimental. Funds cut their shorts along with their longs, and buying back those shorts can push prices up quickly even when nobody has new confidence in the stocks.

The rally was also unusually broad. Bespoke posted on X on November 9: "The semis are up 5% even though its biggest member is down 5%. Wild times." A rally that wide likely helped SOXX more than SMH, because a fund leaning harder on its largest holding would trail on a day like that.

In plain terms: short covering, not renewed conviction, did most of the lifting.

Are strategists convinced the selling is done?

They are split. JPMorgan's Dubravko Lakos-Bujas said the unwind of long momentum positions against the market was the largest in about 40 years, and he called the pullback largely a "deleveraging event." Lakos-Bujas and his team kept their year-end S&P 500 target of 8,000.

Morgan Stanley's Michael Wilson said the correction is in its final phase, with the S&P 500 approaching his downside target of 7,100 to 7,300. Citigroup strategists went the other way and cut U.S. stocks and global tech to neutral.

Light positioning can support a rebound, but it does not tell you when selling will end. Funds can start selling again if chip guidance or AI spending plans weaken.

The takeaway: the desks agree on what happened — deleveraging — but not on whether it's finished.

What does washed-out positioning mean for SOXX holders?

The evidence points to a positioning-driven selloff rather than a verdict on the sector. Gross leverage fell far more than net leverage, which looks more like forced risk-cutting than a permanent change in how funds view chips.

The numbers keep that read in perspective. SOXX was still up 46.45% year to date after a 21.29% drop over the past month. SMH was up 35.34% for the year after a 19.21% monthly decline.

QQQ has carried smaller swings over the same stretch. It fell 9.57% over the month, is up 10.59% for the year, and holds software, retail, and consumer stocks alongside chipmakers — a buffer the pure-play chip funds don't have.

What comes next?

The next test arrives fast: the October CPI report at 8:30 a.m. ET on November 10. A hot inflation reading followed by a SOXX close below $136.53 — where it started on November 9 — would break the deleveraging story. That combination would mean something more fundamental is wrong with the chip trade, and holders will find out within one trading session.