Goldman Sachs highlighted that unusually strong corporate profits have helped sustain elevated valuations in U.S. equities, but cautioned that the ongoing investment surge into artificial intelligence infrastructure could put pressure on profitability metrics for the largest technology companies in coming years.
Year-to-date the S&P 500 has risen 9%, driven mainly by higher earnings expectations rather than a widening of valuation multiples. The index is trading at roughly 21 times forward earnings, a level noted by the bank as well above its long-run average.
In a recent report Goldman found that S&P 500 return on equity - a key measure of corporate profitability - hit a record 22% in the first quarter of 2026, exceeding the prior peak recorded in 2021. The increase has been propelled in large part by rising profit margins across the index.
The report emphasized the outsized role of mega-cap technology firms in this performance. Collectively, the largest technology companies now record a combined ROE of 44%. Goldman singled out the seven biggest technology names - Nvidia, Apple, Microsoft, Amazon, Alphabet, Meta Platforms, and Broadcom - noting that their combined ROE has increased by nine percentage points over the past three years.
Despite that contribution to headline profitability, the bank said the AI infrastructure buildout is likely to create new headwinds for margins and return on equity among hyperscale tech firms. Analysts cited in the report expect ROE across the largest technology companies to fall by an average of seven percentage points next year as spending on data centers, computing hardware, and AI-related assets accelerates.
Goldman noted that hyperscale technology companies are becoming more asset-intensive as they expand capacity to support AI workloads. As a result, depreciation and amortization expense is projected to climb from about 7% of revenue in 2022 to roughly 12% by 2027. The report also describes an increased reliance on debt and equity financing to fund these investment plans.
The bank pointed to consensus estimates that major hyperscalers will commit approximately $770 billion to capital expenditures in 2026 - an amount equivalent to roughly 100% of their operating cash flow, according to the report.
Nonetheless, Goldman retained a constructive long-term view on AI's potential to boost corporate profitability across the economy. The report said productivity gains from AI adoption could eventually lift sales, earnings per employee, and profit margins across a broad set of industries.
The analysts further observed that semiconductor companies stand to be among the principal beneficiaries of the AI spending cycle, supported by solid pricing power and elevated margins within that sector.
Key points
- Record S&P 500 ROE: The index's ROE reached 22% in Q1 2026, above the previous high in 2021.
- Mega-cap influence: The seven largest technology companies have pushed combined ROE to 44% and raised their combined ROE by nine percentage points over three years.
- AI capex trade-off: Accelerating investment in data centers and AI infrastructure is expected to depress ROE for major tech firms, while benefiting semiconductor suppliers.
Risks and uncertainties
- ROE compression for large tech firms - Analysts expect the largest technology companies to see average ROE declines of about seven percentage points next year, tied to heavier capital spending.
- Rising non-cash charges - Depreciation and amortization are projected to increase from roughly 7% of revenue in 2022 to about 12% by 2027, which could weigh on reported margins.
- Financing strain from capex - Consensus estimates place hyperscaler capex at around $770 billion in 2026, near 100% of operating cash flow, increasing dependence on debt and equity financing to fund expansion.