Goldman Sachs top quantitative analyst: Buying the dip in chip/momentum stocks, now leverage is lower and positions are lighter.

Goldman Sachs' top quantitative strategist believes the time has come to go long on semiconductors and momentum stocks.

Shawn Tuteja, head of ETF and custom basket volatility trading in Goldman Sachs' Global Banking & Markets division, noted in a recent report that the current nearly 20% drawdown in the momentum factor, particularly in the semiconductor and AI sectors, is primarily driven by technical and structural factors.

He believes that with reduced leverage within the system and more balanced institutional positioning, investors have a tactical opportunity for "tentative positioning" on the long side of semiconductors and momentum stocks.

The S&P 500 year-end target of over 8,000 points remains a market consensus with little controversy.

Goldman Sachs' 2027 EPS forecast is $385, while the market consensus is $398. Taking the midpoint and applying a P/E of 20.5x still yields a level above 8,000 points, less than 7% from the current closing price.

Technical Drivers Behind the Momentum Factor Drawdown

Tuteja believes that the over 20% drawdown in the momentum factor fundamentally reflects an improvement in market pricing efficiency, rather than a breakdown in the AI trade narrative.

(Goldman Sachs High Beta Momentum Stock Basket Volatility-Adjusted Drawdown)

He notes that momentum strategies historically underperform in the second half of July. With the market fully aware of this pattern, the "rotation trade" this year began on July 1 instead of waiting until July 18.

In terms of volatility-adjusted drawdown magnitude, the current decline closely aligns with the trajectory of previous sell-offs in the momentum factor under similar realized volatility environments, only that the market reached this level faster.

Client feedback from Goldman's cash trading desk corroborates this assessment.

According to data from Goldman's TMT desk, client sentiment currently stands at approximately 7.5 to 8 out of 10, cooling from 9.5 to 10 a month ago. Clients generally agree that the core driver of this pullback is technical.

Several clients noted that the SOX semiconductor index rallied approximately 100% between April and May, with only two corrections of about 5% during that period. A period of consolidation thereafter is considered reasonable.

They also observed that if the market were truly "pronouncing a top" for the AI trade, one would expect to see broader declines in the AI sector. However, stocks like Dell and CRDO remain elevated, and large-cap SaaS and IT services names have not exhibited a significant bounce, which is inconsistent with a topping signal.

Leverage Receding, Structural Pressures Being Digested

The total AUM of U.S. leveraged semiconductor ETFs has plunged from a mid-June peak of approximately $157 billion to roughly $104 billion as of July 8, a cumulative reduction of about $53 billion. This deleveraging process not only represents capital outflows but also directly alters market microstructure.

(Total AUM of U.S. Leveraged Semiconductor ETFs)

At the peak, the daily short gamma exposure generated solely by semiconductor leveraged ETFs was approximately $2.8 billion, meaning that on a single-day 3% rally in the semiconductor sector, leveraged rebalancing would require buying roughly $8.5 billion in semiconductor stocks. This daily gamma exposure has now shrunk to about $1.9 billion.

Notably, if the implied size compression were solely due to spot price declines, the complex's AUM would have fallen to approximately $89 billion. The actual figure of $104 billion implies that various investor groups have bought an additional $15 billion in leveraged semiconductor ETFs during the decline, indicating persistent "buy-the-dip" behavior.

Meanwhile, hedge fund positioning data also points to deepening deleveraging.

Goldman Sachs Prime Services data shows that over the past few weeks, fundamental long/short funds have significantly reduced their AI and momentum exposure, with gross leverage falling to its lowest decile in nearly a year. Fundamental long/short managers are down 2.2% since June 22 but still up roughly 15.5% year-to-date.

Additionally, financing costs for institutional clients have risen notably. Reports indicate that last week, one-month financing rates for popular names like SK Hynix and Samsung briefly reached Fed Funds Rate +12%, further reducing the appeal of holding positions in a sideways market.

Single-Stock Volatility Premium at Historic Extremes, Compression Window Opening

On the volatility front, the current premium of single-stock implied volatility over index implied volatility has risen to the highest level in the past 20 years.

Goldman Sachs data shows that the weighted average implied volatility (3-month tenor) of the top 50 S&P 500 constituents is approximately 26 volatility points higher than the index's implied volatility. This implies extremely low index implied correlation, meaning that when the semiconductor sector falls, sectors like healthcare and financials must necessarily rise.

(3-Month Weighted Average Single-Stock Implied Volatility vs. 3-Month Index Implied Volatility)

Tuteja points out that monthly at-the-money straddle strategies on semiconductor ETFs (rolled weekly, delta-hedged daily) have generated consistent positive returns since the start of the AI trade, while identical strategies on the S&P 500 have generated consistently negative returns — a stark contrast.

(20-Day Rolling Realized Volatility of Semiconductor ETFs Since the Start of the AI Trade)

He believes that with reduced leverage in the system, shrinking daily short gamma exposure, and stabilizing positioning structures, implied and realized volatility in the semiconductor sector should begin to compress.

Currently, single-stock call skew relative to put skew remains attractive, making it suitable for investors looking to hold long positions while capping upside and hedging downside, potentially via collar or put-spread collar strategies.

Upside Expectations Narrowing, Fundamental Focus Shifts to Earnings Season

Compared to a month ago, when some AI stocks doubled or even tripled in a short period, the market's expectations for upside magnitude have become more rational.

Goldman Sachs research had previously heavily promoted collar strategies of "selling out-of-the-money calls to pay for put protection." These strategies saw strong inquiries in April and May but few executions, as investors were generally worried about missing out on upside by selling calls.

Tuteja notes that hyperscaler cloud capex expectations for 2026 were revised up by roughly $100 billion in Q2, a key catalyst driving semiconductor momentum in April and May. However, Goldman Sachs research expects no similar magnitude of capex upward revision during the current earnings season.

As July earnings season approaches, client focus will center on AI investment returns, open-source model developments, and token usage trends.

Tuteja also points out that the biggest variable affecting the medium-term market landscape will be the trajectory of the Fed's potential rate hiking cycle.

The prevailing view among equity accounts is zero rate hikes, current market pricing implies two, while several macro accounts are betting on four or more. The implications of these three paths for market leadership will be fundamentally different.

Risk Disclaimer

        Markets are risky, and investment requires caution. This article does not constitute personal investment advice and does not consider the specific investment objectives, financial situations, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investment based on this content is at your own risk.