• HSBC (HSBC) expects strong Q3 earnings and accelerating AI demand to extend the equity bull market.
  • S&P 500 forward valuations have fallen 16.5% since late 2025, while earnings guidance is improving at its fastest pace since 2021.
  • The bank favors AI hyperscalers, citing surging computing demand and resilient GPU rental prices.

HSBC is urging investors not to bet against the AI trade ahead of third-quarter earnings, arguing that improving corporate profits and rising demand for AI computing could prolong the equity rally. In an October 13 strategy note, Alastair Pinder, HSBC’s head of emerging-market and global equity strategy, made the case for staying invested, with large cloud providers—the “hyperscalers”—its preferred near-term beneficiaries.

The note’s central claim is that valuations have moderated meaningfully. The S&P 500’s forward price-to-earnings ratio has fallen 16.5% since late 2025, and 27% of constituents have seen multiple declines exceeding 20% this year. That’s a decline in valuation ratios, not a 16.5% fall in the index. HSBC says company guidance is improving at its fastest pace since 2021 and anticipates another “beat-and-raise” reporting season—results above expectations followed by stronger forecasts.

AI demand remains a key pillar. HSBC points to accelerating inference demand—the computing used to run AI models—and firm GPU rental prices as evidence that spending on AI infrastructure is not slowing. The report specifically refers to rental prices, rather than necessarily the purchase prices of graphics processors. These figures are HSBC’s reported analysis, as relayed by Investing.com; they should not be mistaken for independently verified market-wide measurements or guaranteed outcomes.

The Earnings Test

The short-term catalyst is clear: Q3 results must deliver. HSBC’s thesis requires earnings surprises, stronger guidance, and evidence that AI demand is converting into revenue—not merely larger spending commitments. The bank’s own Q3 earnings release is scheduled for October 27, so those results were not available at the time of the strategy note.

HSBC is a London-headquartered banking group with approximately 41 million customers across 56 markets and $1.8 trillion in customer deposits at the end of 2025. Its latest published results, for the first half of 2026, showed revenue of $37.7 billion, up 11% year over year, and profit before tax of $19.5 billion, up 23%. Banking net interest income rose to $22.9 billion, while expected credit losses increased to $2.4 billion. The common equity tier 1 capital ratio stood at 14.1%. Much of the reported profit increase reflected a more favorable comparison in exceptional items; constant-currency pretax profit excluding those items increased by $1.1 billion to $20.4 billion.

CEO Georges Elhedery, who took office in September 2024, continues to oversee a restructuring into four main businesses, alongside disposals and cost reductions. October 12 reporting also revealed plans to replace senior employees’ fixed-pay allowances with more performance-linked bonuses from January 2028, without reducing base salaries. Separately, recent reports describe substantial proposed cuts to HSBC’s UK wealth-advisory workforce as it introduces AI-enabled services—a reminder that the bank’s investment enthusiasm for AI is paired with a more difficult operational reality.

From Spending to Monetization

The investment debate is shifting from whether companies will spend heavily on AI to whether that spending produces durable revenue and cash flow. Reuters (TRI) reported in August that cloud growth was accelerating and capacity constraints persisted, prompting investors to increase exposure to hyperscalers as well as semiconductor suppliers. That supports the direction of HSBC’s argument, though it does not establish that every AI-linked stock is attractively valued.

The spending scale remains substantial. In June, Reuters cited JPMorgan (JPM) estimates that five major companies, including Microsoft (MSFT), Alphabet (GOOGL) and Amazon (AMZN), expected combined capital expenditure of approximately $730 billion in 2026. Benefits extend beyond technology to construction, industrial equipment and energy suppliers involved in building and powering data centers.

There is also a cash-flow constraint: Reuters’ August analysis estimated hyperscalers’ annual operating cash flow would rise by about $340 billion between 2025 and 2027, while capital expenditure would increase by roughly $534 billion. Strong demand can therefore coexist with pressure on cash available to shareholders.

HSBC also argues that higher rates should have a limited immediate effect on S&P 500 earnings because approximately 80% of constituent companies’ debt is long-term and fixed-rate. It identifies $64 billion in customs refunds as potential margin support. Both are claims from its strategy note; the available report does not detail the refunds’ timing or distribution across companies.

Energy availability is a longer-term constraint. The International Energy Agency’s base case projects global data-center electricity consumption rising from approximately 415 terawatt-hours in 2024 to 945 terawatt-hours in 2030. That creates opportunities for utilities and grid investment, but also makes electricity supply a critical condition for sustained AI expansion. The IEA expects the US and China to account for nearly 80% of the increase in data-center electricity consumption through 2030.

HSBC’s latest position continues a strategy it articulated earlier this year. Its Q3 investment outlook identified “invest in the AI-led future” as a priority, pointing to semiconductors, data centers and AI adopters; September commentary likewise emphasized accelerating adoption and broadening earnings. The October note makes upcoming earnings the next test of that thesis.

The recurring historical comparison is the 1999–2000 dot-com boom. A September Reuters commentary argued that today’s profitable technology leaders provide a stronger earnings foundation than many speculative dot-com companies had—an argument against a simplistic comparison, not proof that current prices or capital-spending plans are sustainable.

For now, HSBC’s message is unambiguous: don’t fight the AI trade. But the bank’s own analysis implies that the next leg higher depends on earnings delivery, spending discipline, and the price investors pay. Whether Q3 results validate that optimism will become clear in the coming weeks.

Correction: An earlier version of this article misstated the nature of the 16.5% decline. It refers to a decline in the S&P 500’s forward price-to-earnings ratio, not a fall in the index itself.