In early April 2026, Goldman Sachs chief global equity strategist Peter Oppenheimer published a note to clients that cut through the noise of a turbulent market moment with a single, striking assertion: US technology stocks were presenting a generational buying opportunity. The kind that appears once in decades, not once in a market cycle.
The note landed at a moment when that claim required genuine conviction to make. Tech stocks had just endured their worst stretch of relative performance against the broader market in 50 years.
The Roundhill Magnificent Seven ETF, tracking equal exposure to:
- Alphabet
- Amazon
- Apple
- Meta
- Microsoft
- Nvidia
- Tesla
All had slid roughly 11% for the year. The iShares US Technology ETF had pulled back sharply. Big Tech, for years the undisputed engine of market returns, had been left behind by sectors that had spent most of the previous decade as afterthoughts: energy, basic resources, healthcare, and industrials.
Goldman’s position was that the selloff had created something rare. A disconnect between what tech companies were actually earning and what the market was willing to pay for those earnings. And that disconnect, Oppenheimer argued, was the opportunity.
What Created the Selloff
Understanding Goldman’s argument requires understanding what drove the underperformance in the first place. Several forces converged in 2025 and early 2026 to shake investor confidence in a sector that had seemed structurally immune to extended weakness.
The DeepSeek disruption
The emergence of DeepSeek. The Chinese AI model that demonstrated competitive performance at dramatically lower cost than leading US models, rattled the assumption that AI leadership required the scale of capital only the largest US companies could deploy. If a highly capable AI model could be built at a fraction of the cost previously assumed, the multi-hundred-billion-dollar capital expenditure commitments made by Amazon, Microsoft, Alphabet, and Meta would suddenly look less like moats and more like potential overinvestment.
Hyperscaler spending concerns
The sheer scale of AI infrastructure investment committed by the largest technology companies, including data centres, chips, power infrastructure, triggered investor concern about return timelines. Capital was flowing out of these companies at unprecedented rates, and the market began questioning when and whether those investments would generate the returns being implicitly priced into valuations.
AI disruption of software
The same AI capabilities that justified massive infrastructure investment were also disrupting the software companies that had been among the most reliably valued components of the tech sector. If AI could automate functions that software tools previously handled, the long-term revenue outlook for many software businesses became genuinely uncertain.
The Iran conflict and macro pressure
The ongoing US-Israel conflict with Iran has driven oil prices significantly higher, creating inflationary pressure and raising questions about global economic growth. In this environment, investors rotated toward sectors with direct commodity exposure and away from growth-oriented technology stocks.
The combined effect was a reassessment, investors re-rated long-neglected “old economy” sectors upward while simultaneously re-rating technology downward, producing the 50-year low in relative performance that Goldman identified as the starting point for its analysis.
What Goldman Actually Found in the Numbers
Oppenheimer and his team did not simply argue that tech had fallen and would therefore rise. They made a specific valuation argument grounded in 4 distinct data points.
1. The PEG ratio reset
The price-to-earnings-to-growth ratio, which compares a stock’s current price against expected earnings growth, had reset to a level not seen since the aftermath of the dot-com bubble in 2003 to 2005. The PEG ratio for US technology had fallen below that of the global aggregate market, meaning investors were paying less for each unit of tech earnings growth than for the average global company.
Goldman described this as tech’s trailing PEG implying “future earnings will be much weaker,” a level of embedded pessimism that, historically, has marked buying opportunities rather than the beginning of structural decline.
2. Valuation parity with the S&P 500
The valuation of the hyperscalers, Amazon, Alphabet, and their peers, had converged with the rest of the S&P 500. For companies that were generating significantly higher earnings growth than the market average, trading at the same multiple was, in Goldman’s framing, a mispricing.
Jed Ellerbroek, portfolio manager at Argent Capital Management, put the specific anomaly in concrete terms: Walmart was trading at a higher price-to-earnings ratio than Amazon. “That’s a bit of an anomaly,” Ellerbroek told The Daily Upside. “Typically, the faster-growing companies trade at higher multiples.”
3. The IT sector P/E is below that of consumer staples and industrials
Globally, the IT sector’s price-to-earnings ratio had fallen below that of consumers discretionary, consumer staples, and industrial sectors. This was a historically unusual configuration, sectors with lower growth rates trading at higher multiples than a sector producing some of the fastest earnings growth in the market.
4. Record the gap between performance and earnings
Goldman identified a record-level gap between how tech stocks had performed and how tech companies’ underlying earnings had actually developed. This gap, performance deteriorating while earnings growth remained strong, is precisely the configuration that value-oriented investors look for: price dislocated from fundamental value.
The Earnings Story That the Market Was Ignoring
Central to Goldman’s argument was that the selloff had been driven by sentiment, narrative, and sector rotation rather than by deteriorating business fundamentals. The fundamental picture for tech earnings, Goldman argued, remained strong, and was in fact strengthening.
Within the S&P 500, the IT sector was projected to generate 44% earnings per share growth for Q1 2026, accounting for 87% of the entire index’s earnings per share growth. This was not a sector in fundamental decline. It was a sector whose stock prices had disconnected from its earnings trajectory.
Earnings revisions across tech, adjustments made by analysts to their forward earnings estimates, remained more positive than for any other sector. Even as investors sold tech stocks, analysts who followed individual companies were raising their estimates for what those companies would earn.
Morningstar, during the same period, specifically named Microsoft as one of the most undervalued stocks available, alongside Broadcom and NXP Semiconductors. “We remain confident in secular tailwinds in tech, including cloud computing, artificial intelligence and the long-term expansion of semiconductor demand,” Morningstar’s analysts wrote. “After months of poor performance, we see software as offering the most upside within the sector.”
AI Spending: The Tailwind the Market Underestimated
A key component of Goldman’s analysis was that the market had mispriced the long-term value of the AI capital expenditure wave rather than correctly identifying it as overinvestment.
Corporate AI spending in 2026 was projected to remain at historically elevated levels, with forecasts pointing toward continued acceleration rather than pullback. Goldman’s Oppenheimer and his team acknowledged the concerns over capital expenditure but noted: “Analyst estimates for the magnitude of the earnings tailwind created by those investments have only increased during the past few weeks.”
The argument was essentially this: the market was pricing tech stocks as if AI investment would not generate returns. The analysts who follow these companies in detail were revising their estimates upward, indicating they believed the returns were coming. The gap between the market’s implied pessimism and the analysts’ forward estimates represented the opportunity Goldman was identifying.
Why the Iran Conflict Made Tech More Attractive, Not Less
One of the counterintuitive dimensions of Goldman’s April 2026 note was the argument that the ongoing conflict with Iran, which had contributed to market volatility and tech’s underperformance, was actually a reason to favour tech over other sectors.
The mechanism was interest rates. The longer the disruption to Strait of Hormuz shipping continued, the greater the risk of a “perceived growth shock” that would limit central banks’ ability to raise interest rates or even prompt them to cut. Technology companies’ cash flows are relatively insensitive to near-term economic growth. The demand for cloud computing, AI services, and software does not collapse in a mild recession the way commodity demand or consumer spending might. But technology stocks benefit significantly from lower interest rates, because lower rates increase the present value of future earnings, and tech earnings are disproportionately weighted toward the future.
Oppenheimer wrote that “given the relative insensitivity of the cash flows in the technology sector to economic growth, and the benefit it would derive on any rally in bond yields, this sector might prove to be more defensive over the next few months“. In a geopolitically uncertain environment where rate expectations were shifting, tech’s defensive characteristics had quietly improved.
No Bubble: The Historical Comparison
Goldman was explicit in addressing the most common concern about technology valuations: whether the current situation resembled the dot-com bubble of 1999 to 2000.
The comparison, Goldman argued, did not hold. Current valuations, even at their pre-selloff peaks, were lower than what existed before the 2000 tech bubble burst. The “Nifty Fifty” bust of the 1970s, another historical analogue for overvalued market darlings, also saw higher valuations than what the current tech sector carries. The market had not been flooded with tech IPOs, a hallmark of late-stage bubble behaviour, and the companies at the centre of the current AI investment cycle were generating real revenue, real earnings, and real cash flow rather than the speculative projections that characterised the dot-com era.
Goldman’s framing was that this was not a bubble deflating. It was a growth sector that had been re-rated downward by rotation, sentiment, and macro noise, and that the re-rating had overshot.
What the Opportunity Looked Like in Practice
For investors trying to act on Goldman’s analysis, the landscape was specific. The following were the most cited names and instruments connected to the opportunity Goldman identified:
The Magnificent Seven including Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla, had individually reached valuation levels that multiple analysts described as anomalous given their earnings growth rates. The Roundhill Magnificent Seven ETF offered equal-weighted exposure to all 7.
Microsoft was specifically named by Morningstar as among the most undervalued large-cap stocks. Broadcom and NXP Semiconductors were also cited as undervalued, with Morningstar pointing to cloud computing, AI infrastructure, and long-term semiconductor demand as the structural tailwinds supporting their outlook.
Software companies, the sub-sector most directly affected by concerns about AI disruption to their business models, were identified by Morningstar as offering the most upside within tech, on the premise that the market had overreacted to disruption risk.
The iShares US Technology ETF had pulled back sharply from its highs, offering broad sector exposure at the discounted valuation Goldman was pointing to.
Goldman’s own framing was sector-level rather than stock-specific. The argument was that technology as a category had been mispriced relative to its earnings growth, and that the entry point created by the selloff was historically rare.
Conclusion
Goldman Sachs’ April 2026 call on US technology stocks was one of the more clearly articulated contrarian positions of the year. The logic was straightforward: the worst relative performance in 50 years had created the best relative valuation in decades, while the underlying earnings of technology companies continued to grow at rates that outpaced every other sector in the market. The gap between what tech stocks were priced at and what tech companies were earning was, in Goldman’s assessment, not a warning sign. It was the opportunity. Whether the timing proved correct would depend on how quickly the market moved to close that gap.
