After a sharp sell-off in global technology stocks, hedge funds are withdrawing from this once highly favored investment sector at a pace close to record levels.
As the US midterm elections sparked concerns about the outlook for artificial intelligence (AI) infrastructure expansion, institutional investors accelerated their reduction of technology stock exposure and significantly lowered their leverage levels.
According to Bloomberg on Monday, data from Goldman Sachs Group (NYSE: GS) prime brokerage showed that the technology sector last week recorded the highest share of net selling among global and US hedge funds, with its net selling ratio approaching the highest level since Goldman began tracking in 2016.
Data as of November 6 also showed that hedge funds' net selling of individual stocks hit a four-month high. Meanwhile, US industrial stocks also saw concentrated selling, with the net selling pace reaching the fastest since April 2020.
Tech Stocks Hit by Concentrated Selling, Hedge Funds Sharply De-leverage
The most notable change in this round of position reduction was reflected in hedge funds' leverage levels. Goldman Sachs data showed that last week, the gross leverage of US fundamental equity long-short strategy funds fell 8.7 percentage points to 207.5%, the largest single-week decline since March 2020.
Net leverage dropped 3.7 percentage points to 43.3%, the largest single-week decline since the week when the US announced "Liberation Day" tariff measures in April 2025.
Gross leverage measures the overall risk exposure of a fund's long and short positions, while net leverage reflects the difference between long and short exposure. The simultaneous decline in both indicators shows that funds are not only reducing technology holdings but also comprehensively compressing portfolio risk.
This round of position reduction occurred after a sharp decline in US technology stocks. Last Thursday, the Nasdaq 100 plunged 6.4%, the largest single-day drop since March 2020, wiping out approximately $2.3 trillion in market value.
For funds with highly concentrated holdings in AI, chip, and data center-related companies, the rapid market reversal put significant pressure on previously accumulated gains.
AI Buildout Prospects Questioned, Selling Spreads to Industrial Sector
The US midterm elections became an important trigger for this round of position reduction. After Democrats won control of Congress in the elections, the market began reassessing potential approval restrictions, power access costs, and local regulatory changes that data center projects might face.
Previously, investors widely expected that large technology companies would continue to expand AI capital expenditures, thereby driving demand growth for chips, servers, power equipment, and industrial infrastructure. However, as construction progress for related projects faces new uncertainties, the market began questioning how many data centers can actually be built as planned in the coming years.
This change also explains why the selling was not limited to technology companies. Some companies in the industrial sector also benefited from expectations of AI infrastructure expansion. When investors lower their judgment on the future pace of data center construction, related equipment and infrastructure suppliers may also be affected.
However, the current position reduction does not mean that AI investment demand has already experienced a comprehensive contraction. Some institutions believe that large cloud computing companies are still continuously increasing capital expenditures, and the recent sharp market adjustment also includes the impact of concentrated liquidation of leveraged funds.
Signs of Buying Back Emerged Last Friday, Market Pressure Not Yet Fully Relieved
Despite the historic sell-off in technology stocks, Goldman Sachs data also showed that the market began to show some signs of stabilization late last week.
Goldman Sachs equity trading desk data showed that hedge funds ultimately recorded net buying overall last week, mainly driven by buying on Friday. But it is worth noting that these purchases mainly came from short covering of macro positions such as indices and exchange-traded funds (ETFs), rather than large-scale re-accumulation of individual technology stocks.
Therefore, the near-record net selling of technology stocks and the overall net buying by hedge funds are not contradictory. They reflect capital flows across different asset scopes and trade types.
At present, whether institutional investors are willing to re-increase AI-related stock exposure remains uncertain. As technology stock valuations, data center construction pace, and future earnings expectations come under renewed scrutiny, changes in hedge funds' risk appetite may continue to affect market volatility.
Last week's position reduction has shown that the adjustment of the AI trade is not just a price decline in some technology stocks, but also involves a systematic contraction of institutional portfolio leverage and risk exposure.
Whether related capital can flow back in the future will depend on improvements in market liquidity and whether corporate earnings and AI investment demand can deliver on previous growth expectations.
Comments